SlideShare une entreprise Scribd logo
1  sur  60
Télécharger pour lire hors ligne
CRF-F: D P F 
         S S E
            B L, D F  L L


                   Hannes Schulz

                University of Freiburg, ACS



                       Feb 2008
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
Intro               Transformation of Directed Model to CRF               Application             Experimental Results


 C: S E
 C D M A  S E



                 ut−2                           ut−1


         xt−2                   xt−1                                 xt


                                  ...                                       ...          n
                 1    2                  n
                                        zt−1                   1
                                                              zt       2
                                                                      zt                zt
                zt−1 zt−1



        P (xt |u1:t −1 , z1:t ) = ηP (zt |xt )          P (xt |ut −1 , xt −1 )P (xt −1 |u1:t −2 , z1:t −1 ) dxt −1
Intro       Transformation of Directed Model to CRF    Application             Experimental Results


 D  D M P




        p (zt |xt ) =            n
                                 i =1   p (zti |xt )                 p (xt +1 |xt , u)
Intro                Transformation of Directed Model to CRF           Application             Experimental Results


 D  D M P



             p (zt |xt ) =                n
                                          i =1   p (zti |xt )                        p (xt +1 |xt , u)
            i
        P (zt |xt)                               ˆi
                                                 zt             zmax




           zrand
Intro                    Transformation of Directed Model to CRF    Application             Experimental Results


 D  D M P

                     p (zt |xt ) =            n
                                              i =1   p (zti |xt )                 p (xt +1 |xt , u)
            i
        P (zt |xt)                ˆi
                                  zt         zmax




           zrand
Intro                    Transformation of Directed Model to CRF    Application                Experimental Results


 D  D M P

                     p (zt |xt ) =            n
                                              i =1   p (zti |xt )                 p (xt +1 |xt , u)
            i
        P (zt |xt)                ˆi
                                  zt         zmax
                                                                                            δrot2


           zrand
                                                                                               xt

                                                                                  δtrans



                                                                                    δrot1
                                                                              xt−1

                                                                         u = (δrot1 , δrot2 , δtrans )
                                                                         executed with gaussian
                                                                         noise
Intro        Transformation of Directed Model to CRF            Application        Experimental Results


 A P   D A



                                                       p (zti |xt ) are not cond. independent

        zt
                            xt
Intro                         Transformation of Directed Model to CRF            Application        Experimental Results


 A P   D A


               ut−2             ut−1


        xt−2          xt−1                  xt                          p (zti |xt ) are not cond. independent
                                                                        Sensor models can only be
                       ...                       ...    n
            1    2            n
                             zt−1       1    2         zt
           zt−1 zt−1                   zt   zt                          generated seperatly for each beam


             i
         P (zt |xt)                    ˆi
                                       zt              zmax




            zrand
Intro                         Transformation of Directed Model to CRF            Application        Experimental Results


 A P   D A


               ut−2             ut−1


        xt−2          xt−1                  xt                          p (zti |xt ) are not cond. independent
                                                                        Sensor models can only be
                       ...                       ...    n
            1    2            n
                             zt−1       1    2         zt
           zt−1 zt−1                   zt   zt                          generated seperatly for each beam
                                                                        Assumption that measurements
                                                                        are independent: “Works
             i
         P (zt |xt)                    ˆi
                                       zt              zmax
                                                                        surprisingly well”. . . if. . .


            zrand
Intro                         Transformation of Directed Model to CRF            Application        Experimental Results


 A P   D A


               ut−2             ut−1


        xt−2          xt−1                  xt                          p (zti |xt ) are not cond. independent
                                                                        Sensor models can only be
                       ...                       ...    n
            1    2            n
                             zt−1       1    2         zt
           zt−1 zt−1                   zt   zt                          generated seperatly for each beam
                                                                        Assumption that measurements
                                                                        are independent: “Works
             i
         P (zt |xt)                    ˆi
                                       zt              zmax
                                                                        surprisingly well”. . . if. . .
                                                                             increasing uncertainty (tweaking)
                                                                             using every 10th measurement
            zrand                                                            ...
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
Intro                    Transformation of Directed Model to CRF           Application     Experimental Results


 I: CRF



                                                                   Undirected graphical models

               ut−2          ut−1


        xt−2          xt−1                 xt


                      zt−1                 zt
Intro                    Transformation of Directed Model to CRF           Application     Experimental Results


 I: CRF



                                                                   Undirected graphical models
                                                                   Every (possible) dependency
               ut−2          ut−1                                  represented by edge

        xt−2          xt−1                 xt


                      zt−1                 zt
Intro                    Transformation of Directed Model to CRF           Application      Experimental Results


 I: CRF



                                                                   Undirected graphical models
                                                                   Every (possible) dependency
               ut−2          ut−1                                  represented by edge
                                                                   Distribution defined over products
        xt−2          xt−1                 xt
                                                                   of functions over cliques
                      zt−1                 zt
Intro                    Transformation of Directed Model to CRF           Application       Experimental Results


 I: CRF



                                                                   Undirected graphical models
                                                                   Every (possible) dependency
               ut−2          ut−1                                  represented by edge
                                                                   Distribution defined over products
        xt−2          xt−1                 xt
                                                                   of functions over cliques
                      zt−1                 zt                      Functions are called clique
                                                                   potentials
Intro                    Transformation of Directed Model to CRF            Application       Experimental Results


 I: CRF



                                                                   Undirected graphical models
                                                                   Every (possible) dependency
               ut−2          ut−1                                  represented by edge
                                                                   Distribution defined over products
        xt−2          xt−1                 xt
                                                                   of functions over cliques
                      zt−1                 zt                      Functions are called clique
                                                                   potentials
                                                                   Clique potentials represent
                                                                   compatibility of their variables
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
Intro              Transformation of Directed Model to CRF              Application             Experimental Results


 CRF-M  S E

                                    ut−2                     ut−1


                            xt−2                 xt−1                          xt


                                                  zt−1                         zt



                                                                T
                                                 1
        p (x0:T |z1:T , u0:T −1 ) =                                   ϕp (xt , xt −1 , ut −1 )ϕm (xt , zt )
                                         Z (z1:T , u1:T −1 )
                                                               t =1
Intro              Transformation of Directed Model to CRF              Application             Experimental Results


 CRF-M  S E

                                    ut−2                     ut−1


                            xt−2                 xt−1                          xt


                                                  zt−1                         zt



                                                                T
                                                 1
        p (x0:T |z1:T , u0:T −1 ) =                                   ϕp (xt , xt −1 , ut −1 )ϕm (xt , zt )
                                         Z (z1:T , u1:T −1 )
                                                               t =1


              Z (·):    all trajectories        ϕp (·)ϕm (·)
Intro              Transformation of Directed Model to CRF              Application             Experimental Results


 CRF-M  S E

                                    ut−2                     ut−1


                            xt−2                 xt−1                          xt


                                                  zt−1                         zt



                                                                T
                                                 1
        p (x0:T |z1:T , u0:T −1 ) =                                   ϕp (xt , xt −1 , ut −1 )ϕm (xt , zt )
                                         Z (z1:T , u1:T −1 )
                                                               t =1


              Z (·): all trajectories ϕp (·)ϕm (·)
              How to define ϕp (·) and ϕm (·)?
Intro               Transformation of Directed Model to CRF         Application         Experimental Results


 T P P φp

                                                    ut −1 = (δrot1 , δtrans , δrot2 ) odometry
                                                    ut −1 = (δrot1 , δtrans , δrot2 ) derived odometry
                                                    ˆ        ˆ       ˆ        ˆ
                  δrot2
                                                                                  2
                                   Before: Gaussian noise N uti −1 , σi
                     xt

        δtrans



          δrot1
   xt−1
Intro               Transformation of Directed Model to CRF         Application          Experimental Results


 T P P φp

                                                    ut −1 = (δrot1 , δtrans , δrot2 ) odometry
                                                    ut −1 = (δrot1 , δtrans , δrot2 ) derived odometry
                                                    ˆ        ˆ       ˆ        ˆ
                  δrot2
                                                                                   2
                                   Before: Gaussian noise N uti −1 , σi
                     xt
                                                            (δrot1 − δrot1 )2
                                                                       ˆ
                                                                                  
                                                                                  
                                                                                   
        δtrans
                                                                                  
                                fp (xt , xt −1 , ut −1 ) =  (δtrans − δtrans )2
                                                                                  
                                                           
                                                           
                                                           
                                                                      ˆ           
                                                                                   
                                                                                   
                                                                                   
                                                                                      3 features
                                                                                  
                                                              (δrot2 − δrot2 )2
                                                                       ˆ
                                                           
                                                                                  
                                                                                   

          δrot1
   xt−1
Intro               Transformation of Directed Model to CRF         Application           Experimental Results


 T P P φp

                                                    ut −1 = (δrot1 , δtrans , δrot2 ) odometry
                                                    ut −1 = (δrot1 , δtrans , δrot2 ) derived odometry
                                                    ˆ        ˆ       ˆ        ˆ
                  δrot2
                                                                                   2
                                   Before: Gaussian noise N uti −1 , σi
                     xt
                                                            (δrot1 − δrot1 )2
                                                                       ˆ
                                                                                  
                                                                                  
                                                                                   
        δtrans
                                                                                  
                                fp (xt , xt −1 , ut −1 ) =  (δtrans − δtrans )2
                                                                                  
                                                           
                                                           
                                                           
                                                                      ˆ           
                                                                                   
                                                                                   
                                                                                   
                                                                                      3 features
                                                                                  
                                                              (δrot2 − δrot2 )2
                                                                       ˆ
                                                           
                                                                                  
                                                                                   

          δrot1                φp (xt , xt −1 , ut −1 ) = exp wp , fp (xt , xt −1 , ut −1 )
   xt−1
Intro               Transformation of Directed Model to CRF             Application               Experimental Results


 T P P φp

                                                    ut −1 = (δrot1 , δtrans , δrot2 ) odometry
                                                    ut −1 = (δrot1 , δtrans , δrot2 ) derived odometry
                                                    ˆ        ˆ       ˆ        ˆ
                  δrot2
                                                                                          2
                                   Before: Gaussian noise N uti −1 , σi
                     xt
                                                            (δrot1 − δrot1 )2
                                                                       ˆ
                                                                                          
                                                                                          
                                                                                           
        δtrans
                                                                                          
                                fp (xt , xt −1 , ut −1 ) =  (δtrans − δtrans )2
                                                                                          
                                                           
                                                           
                                                           
                                                                      ˆ                   
                                                                                           
                                                                                           
                                                                                           
                                                                                              3 features
                                                                                          
                                                              (δrot2 − δrot2 )2
                                                                       ˆ
                                                           
                                                                                          
                                                                                           

          δrot1                φp (xt , xt −1 , ut −1 ) = exp wp , fp (xt , xt −1 , ut −1 )
   xt−1                                               1                 (a − a )2
                                                                             ˆ
                                         N a,                 = exp −
                                                    σ2                    2σ2

                                  Gaussian noise N uti −1 ,                1
                                                                         −2wpi        if wp < 0
                                                                                          i
Intro       Transformation of Directed Model to CRF                   Application   Experimental Results


 R: S M   N¨ B A
                                  

            i
        P (zt |xt)                                                   ˆi
                                                                     zt             zmax




           zrand

                                                       n
                                   p (zt |xt ) =             p (zti |xt )
                                                      i =1
Intro              Transformation of Directed Model to CRF             Application    Experimental Results


 M P φm
                                                                             i
                                                                         P (zt |xt)    ˆi
                                                                                       zt           zmax
                                            n
                                                              
                                                              
        φm (xt , zt ) = exp  wm ,               fm (zt , xt ) 
                                                  i
                            
                                                              
                                                               
                            
                                                              
                                                               
                                          i =0
                                                                            zrand


                             (¬mti ∧ ¬mti )cti (zti − zti )2
                                      ˆ               ˆ
                                                                  
                    
                    
                                                                  
                                                                   
                                                                   
                                 i      i       i
                                                                  
                    
                    
                    
                    
                                     ˆ
                             (¬mt ∧ ¬mt )¬ct
                                                                   
                                                                   
                                                                   
                                                                   
                                                                   
    fm (zt , xt ) = 
     i
                                                                  
                             (¬mti ∧ mti )
                                      ˆ
                    
                                                                  
                                                                   
                                                                   
                    
                    
                                                                  
                                                                   
                                                                   
                             ( mti ∧ ¬mti )
                                                                  
                    
                    
                    
                    
                                     ˆ                            
                                                                   
                                                                   
                                                                   
                                                                   
                                                                  
                             ( mti ∧ mti )
                                     ˆ
                                                                  
Intro              Transformation of Directed Model to CRF             Application    Experimental Results


 M P φm
                                                                             i
                                                                         P (zt |xt)    ˆi
                                                                                       zt           zmax
                                            n
                                                              
                                                              
        φm (xt , zt ) = exp  wm ,               fm (zt , xt ) 
                                                  i
                            
                                                              
                                                               
                            
                                                              
                                                               
                                          i =0
                                                                            zrand


                             (¬mti ∧ ¬mti )cti (zti − zti )2
                                      ˆ               ˆ
                                                                  
                    
                    
                                                                  
                                                                   
                                                                   
                                 i      i       i
                                                                  
                    
                    
                    
                    
                                     ˆ
                             (¬mt ∧ ¬mt )¬ct
                                                                   
                                                                   
                                                                   
                                                                   
                                                                   
    fm (zt , xt ) = 
     i
                                                                  
                             (¬mti ∧ mti )
                                      ˆ
                    
                                                                  
                                                                   
                                                                   
                    
                    
                                                                  
                                                                   
                                                                   
                             ( mti ∧ ¬mti )
                                                                  
                    
                    
                    
                    
                                     ˆ                            
                                                                   
                                                                   
                                                                   
                                                                   
                                                                  
                             ( mti ∧ mti )
                                     ˆ
                                                                  


         mti ∈ {1, 0} measured zmax
Intro              Transformation of Directed Model to CRF             Application    Experimental Results


 M P φm
                                                                             i
                                                                         P (zt |xt)    ˆi
                                                                                       zt           zmax
                                            n
                                                              
                                                              
        φm (xt , zt ) = exp  wm ,               fm (zt , xt ) 
                                                  i
                            
                                                              
                                                               
                            
                                                              
                                                               
                                          i =0
                                                                            zrand


                             (¬mti ∧ ¬mti )cti (zti − zti )2
                                      ˆ               ˆ
                                                                  
                    
                    
                                                                  
                                                                   
                                                                   
                                 i      i       i
                                                                  
                    
                    
                    
                    
                                     ˆ
                             (¬mt ∧ ¬mt )¬ct
                                                                   
                                                                   
                                                                   
                                                                   
                                                                   
    fm (zt , xt ) = 
     i
                                                                  
                             (¬mti ∧ mti )
                                      ˆ
                    
                                                                  
                                                                   
                                                                   
                    
                    
                                                                  
                                                                   
                                                                   
                             ( mti ∧ ¬mti )
                                                                  
                    
                    
                    
                    
                                     ˆ                            
                                                                   
                                                                   
                                                                   
                                                                   
                                                                  
                             ( mti ∧ mti )
                                     ˆ
                                                                  


         mti ∈ {1, 0} measured zmax
         mti ∈ {1, 0} expected zmax
         ˆ
Intro              Transformation of Directed Model to CRF             Application    Experimental Results


 M P φm
                                                                             i
                                                                         P (zt |xt)    ˆi
                                                                                       zt           zmax
                                            n
                                                              
                                                              
        φm (xt , zt ) = exp  wm ,               fm (zt , xt ) 
                                                  i
                            
                                                              
                                                               
                            
                                                              
                                                               
                                          i =0
                                                                            zrand


                             (¬mti ∧ ¬mti )cti (zti − zti )2
                                      ˆ               ˆ
                                                                  
                    
                    
                                                                  
                                                                   
                                                                   
                                 i      i       i
                                                                  
                    
                    
                    
                    
                                     ˆ
                             (¬mt ∧ ¬mt )¬ct
                                                                   
                                                                   
                                                                   
                                                                   
                                                                   
    fm (zt , xt ) = 
     i
                                                                  
                             (¬mti ∧ mti )
                                      ˆ
                    
                                                                  
                                                                   
                                                                   
                    
                    
                                                                  
                                                                   
                                                                   
                             ( mti ∧ ¬mti )
                                                                  
                    
                    
                    
                    
                                     ˆ                            
                                                                   
                                                                   
                                                                   
                                                                   
                                                                  
                             ( mti ∧ mti )
                                     ˆ
                                                                  


         mti ∈ {1, 0} measured zmax
         mti ∈ {1, 0} expected zmax
         ˆ
          cti ∈ {1, 0} zti − zti < 20cm
                             ˆ
Intro              Transformation of Directed Model to CRF             Application    Experimental Results


 M P φm
                                                                             i
                                                                         P (zt |xt)    ˆi
                                                                                       zt           zmax
                                            n
                                                              
                                                              
        φm (xt , zt ) = exp  wm ,               fm (zt , xt ) 
                                                  i
                            
                                                              
                                                               
                            
                                                              
                                                               
                                          i =0
                                                                            zrand


                             (¬mti ∧ ¬mti )cti (zti − zti )2
                                      ˆ               ˆ
                                                                  
                    
                    
                                                                  
                                                                   
                                                                   
                                 i      i       i
                                                                  
                    
                    
                    
                    
                                     ˆ
                             (¬mt ∧ ¬mt )¬ct
                                                                   
                                                                   
                                                                   
                                                                   
                                                                   
    fm (zt , xt ) = 
     i
                                                                  
                             (¬mti ∧ mti )
                                      ˆ
                    
                                                                  
                                                                   
                                                                   
                    
                    
                                                                  
                                                                   
                                                                   
                             ( mti ∧ ¬mti )
                                                                  
                    
                    
                    
                    
                                     ˆ                            
                                                                   
                                                                   
                                                                   
                                                                   
                                                                  
                             ( mti ∧ mti )
                                     ˆ
                                                                  


         mti ∈ {1, 0} measured zmax
         mti ∈ {1, 0} expected zmax
         ˆ
          cti ∈ {1, 0} zti − zti < 20cm
                             ˆ
Intro              Transformation of Directed Model to CRF             Application    Experimental Results


 M P φm
                                                                             i
                                                                         P (zt |xt)    ˆi
                                                                                       zt           zmax
                                            n
                                                              
                                                              
        φm (xt , zt ) = exp  wm ,               fm (zt , xt ) 
                                                  i
                            
                                                              
                                                               
                            
                                                              
                                                               
                                          i =0
                                                                            zrand


                             (¬mti ∧ ¬mti )cti (zti − zti )2
                                      ˆ               ˆ
                                                                  
                    
                    
                                                                  
                                                                   
                                                                   
                                 i      i       i
                                                                  
                    
                    
                    
                    
                                     ˆ
                             (¬mt ∧ ¬mt )¬ct
                                                                   
                                                                   
                                                                   
                                                                   
                                                                   
    fm (zt , xt ) = 
     i
                                                                  
                             (¬mti ∧ mti )
                                      ˆ
                    
                                                                  
                                                                   
                                                                   
                    
                    
                                                                  
                                                                   
                                                                   
                             ( mti ∧ ¬mti )
                                                                  
                    
                    
                    
                    
                                     ˆ                            
                                                                   
                                                                   
                                                                   
                                                                   
                                                                  
                             ( mti ∧ mti )
                                     ˆ
                                                                  


         mti ∈ {1, 0} measured zmax
         mti ∈ {1, 0} expected zmax
         ˆ
          cti ∈ {1, 0} zti − zti < 20cm
                             ˆ
Intro              Transformation of Directed Model to CRF             Application    Experimental Results


 M P φm
                                                                             i
                                                                         P (zt |xt)    ˆi
                                                                                       zt           zmax
                                            n
                                                              
                                                              
        φm (xt , zt ) = exp  wm ,               fm (zt , xt ) 
                                                  i
                            
                                                              
                                                               
                            
                                                              
                                                               
                                          i =0
                                                                            zrand


                             (¬mti ∧ ¬mti )cti (zti − zti )2
                                      ˆ               ˆ
                                                                  
                    
                    
                                                                  
                                                                   
                                                                   
                                 i      i       i
                                                                  
                    
                    
                    
                    
                                     ˆ
                             (¬mt ∧ ¬mt )¬ct
                                                                   
                                                                   
                                                                   
                                                                   
                                                                   
    fm (zt , xt ) = 
     i
                                                                  
                             (¬mti ∧ mti )
                                      ˆ
                    
                                                                  
                                                                   
                                                                   
                    
                    
                                                                  
                                                                   
                                                                   
                             ( mti ∧ ¬mti )
                                                                  
                    
                    
                    
                    
                                     ˆ                            
                                                                   
                                                                   
                                                                   
                                                                   
                                                                  
                             ( mti ∧ mti )
                                     ˆ
                                                                  


         mti ∈ {1, 0} measured zmax
         mti ∈ {1, 0} expected zmax
         ˆ
          cti ∈ {1, 0} zti − zti < 20cm
                             ˆ
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
Intro          Transformation of Directed Model to CRF          Application   Experimental Results


 U  CRF    P F



    At each time step t:
         Prediction
             Move particles according to gaussian noise
             determined by wp
             Same as sampling from N uti −1 ,
                                     ˆ                    1
                                                            i
                                                         −2wp

        Correction
             Particle at xt gets weight φm (xt , zt )
             Resample (includes normalization)
Intro          Transformation of Directed Model to CRF          Application   Experimental Results


 U  CRF    P F



    At each time step t:
         Prediction
             Move particles according to gaussian noise
             determined by wp
                                                                                    u
             Same as sampling from N uti −1 ,
                                     ˆ                    1
                                                            i
                                                         −2wp

        Correction
             Particle at xt gets weight φm (xt , zt )
             Resample (includes normalization)
Intro          Transformation of Directed Model to CRF           Application     Experimental Results


 U  CRF    P F



    At each time step t:
         Prediction
             Move particles according to gaussian noise
             determined by wp                                                   moved
             Same as sampling from N uti −1 ,
                                     ˆ                     1
                                                         −2wpi                 particles
        Correction
             Particle at xt gets weight φm (xt , zt )
             Resample (includes normalization)
Intro          Transformation of Directed Model to CRF           Application   Experimental Results


 U  CRF    P F



    At each time step t:
         Prediction
             Move particles according to gaussian noise
             determined by wp                                                  added
             Same as sampling from N uti −1 ,                                  noise
                                                           1
                                     ˆ                   −2wpi


        Correction
             Particle at xt gets weight φm (xt , zt )
             Resample (includes normalization)
Intro          Transformation of Directed Model to CRF           Application      Experimental Results


 U  CRF    P F



    At each time step t:
         Prediction
             Move particles according to gaussian noise
             determined by wp                                                  ...sense...
             Same as sampling from N uti −1 ,
                                     ˆ                     1
                                                         −2wpi


        Correction
             Particle at xt gets weight φm (xt , zt )
             Resample (includes normalization)
Intro          Transformation of Directed Model to CRF           Application    Experimental Results


 U  CRF    P F



    At each time step t:
         Prediction
             Move particles according to gaussian noise
             determined by wp                                                  weights
             Same as sampling from N uti −1 ,
                                     ˆ                     1
                                                         −2wpi


        Correction
             Particle at xt gets weight φm (xt , zt )
             Resample (includes normalization)
Intro          Transformation of Directed Model to CRF          Application     Experimental Results


 U  CRF    P F



    At each time step t:
         Prediction
             Move particles according to gaussian noise
             determined by wp
                                                                              resample
             Same as sampling from N uti −1 ,
                                     ˆ                    1
                                                            i
                                                         −2wp

        Correction
             Particle at xt gets weight φm (xt , zt )
             Resample (includes normalization)
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
Intro     Transformation of Directed Model to CRF     Application   Experimental Results


 D     wp  wm

                                   Drive around in test area
Intro     Transformation of Directed Model to CRF     Application    Experimental Results


 D     wp  wm

                                   Drive around in test area
                                   Use high-quality scanmatcher to generate
                                   “ground truth” trajectory x∗
Intro     Transformation of Directed Model to CRF     Application      Experimental Results


 D     wp  wm

                                   Drive around in test area
                                   Use high-quality scanmatcher to generate
                                   “ground truth” trajectory x∗
                                                                                ˆ
                                   Using arbitrary weights, generate trajectory x
                                   with CRF-filter
Intro     Transformation of Directed Model to CRF     Application      Experimental Results


 D     wp  wm

                                   Drive around in test area
                                   Use high-quality scanmatcher to generate
                                   “ground truth” trajectory x∗
                                                                                ˆ
                                   Using arbitrary weights, generate trajectory x
                                   with CRF-filter
                                   Use difference of summed features as weight
                                   update(−) :
                                   wk = wk −1 + α ( f (x∗ , u, z) − f (x, u, z))
                                                                       ˆ
Intro     Transformation of Directed Model to CRF     Application      Experimental Results


 D     wp  wm

                                   Drive around in test area
                                   Use high-quality scanmatcher to generate
                                   “ground truth” trajectory x∗
                                                                                ˆ
                                   Using arbitrary weights, generate trajectory x
                                   with CRF-filter
                                   Use difference of summed features as weight
                                   update(−) :
                                   wk = wk −1 + α ( f (x∗ , u, z) − f (x, u, z))
                                                                       ˆ
                                   Decrease α if new Filter cannot track
Intro       Transformation of Directed Model to CRF     Application      Experimental Results


 D     wp  wm

                                     Drive around in test area
                                     Use high-quality scanmatcher to generate
                                     “ground truth” trajectory x∗
                                                                                  ˆ
                                     Using arbitrary weights, generate trajectory x
                                     with CRF-filter
                                     Use difference of summed features as weight
                                     update(−) :
                                     wk = wk −1 + α ( f (x∗ , u, z) − f (x, u, z))
                                                                         ˆ
                                     Decrease α if new Filter cannot track
                                     loop

        Adapts weights to task, sensor dependencies/environment,
        sensor noise, particle filter parameters
Intro       Transformation of Directed Model to CRF               Application                 Experimental Results


 L A


        Averaged Perceptron Algorithm (Collins 2002) for tagging

                   w k = w k −1 + α                   f (x∗ , u, z) −           f (x, u, z)
                                                                                   ˆ

        Proven to converge even in presence of errors in training data
        Intuition of learning algorithm:
            If PF works correctly, then

                                      f (xn , un−1 , zn ) =
                                          ∗
                                                                  f (xn , un−1 , zn )
                                                                     ˆ

            f i occurs less often in x∗ than in x → decrease influence of f i
                                                ˆ
            on particle filter by decreasing w i
O


  1   I: S E U D M

  2   T  D M  CRF
        Short Introduction to CRF
        CRF-Model for State Estimation

  3   A
        CRF-Filter Algorithm
        Learning the Parameters


  4   E R
Intro     Transformation of Directed Model to CRF              Application       Experimental Results


 E R

                                        Properties of the learned weights
                                                    Norm of weight vector decreases with
                                                    number of laser beams in z
                                                    believes the features/measurements less
                                                    equivalent to initially introduced
                                                    “tweaking”?!
Intro     Transformation of Directed Model to CRF              Application         Experimental Results


 E R

                                        Properties of the learned weights
                                                    Norm of weight vector decreases with
                                                    number of laser beams in z
                                                    believes the features/measurements less
                                                    equivalent to initially introduced
                                                    “tweaking”?!
                                        Two specialized CRF-filters compared to
                                        generative particle filter trained using
                                        expectation maximization

                                                              Tracking       Global
                                                              Error          Localization
                                                                             Accuracy
                                           Generative         7.52 cm        30%
                                           CRF-Filter         7.07 cm        96%
Intro           Transformation of Directed Model to CRF   Application   Experimental Results


 C



        1   A CRF is an alternative, undirected graphical model
Intro           Transformation of Directed Model to CRF   Application   Experimental Results


 C



        1   A CRF is an alternative, undirected graphical model
        2   CRF-Filters use a continuous CRF for recursive state
            estimation
Intro           Transformation of Directed Model to CRF   Application   Experimental Results


 C



        1   A CRF is an alternative, undirected graphical model
        2   CRF-Filters use a continuous CRF for recursive state
            estimation
        3   . . . can be trained to maximize filter performance depending
            on the task
Intro           Transformation of Directed Model to CRF   Application   Experimental Results


 C



        1   A CRF is an alternative, undirected graphical model
        2   CRF-Filters use a continuous CRF for recursive state
            estimation
        3   . . . can be trained to maximize filter performance depending
            on the task
        4   . . . can deal with correlated measurements
Intro           Transformation of Directed Model to CRF   Application   Experimental Results


 C



        1   A CRF is an alternative, undirected graphical model
        2   CRF-Filters use a continuous CRF for recursive state
            estimation
        3   . . . can be trained to maximize filter performance depending
            on the task
        4   . . . can deal with correlated measurements
        5   . . . do not explicitly account for dependencies between sensor
            data

Contenu connexe

En vedette

Realidad Virtual y sus reglas legales
Realidad Virtual y sus reglas legalesRealidad Virtual y sus reglas legales
Realidad Virtual y sus reglas legalesAbanlex
 
Missing school seminar
Missing school seminarMissing school seminar
Missing school seminarcolwilliamson
 
December 2009 TeachStreet Teacher Webinar & Meet-up
December 2009 TeachStreet Teacher Webinar & Meet-upDecember 2009 TeachStreet Teacher Webinar & Meet-up
December 2009 TeachStreet Teacher Webinar & Meet-upTeachStreet
 
Cloenda del Curs d'introducció a Perl 2011
Cloenda del Curs d'introducció a Perl 2011Cloenda del Curs d'introducció a Perl 2011
Cloenda del Curs d'introducció a Perl 2011Alex Muntada Duran
 
Search: The Purest Form of Interaction Design
Search: The Purest Form of Interaction DesignSearch: The Purest Form of Interaction Design
Search: The Purest Form of Interaction DesignChiara Fox Ogan
 
The Raw Remnant - Vegan Health Seminar
The Raw Remnant - Vegan Health SeminarThe Raw Remnant - Vegan Health Seminar
The Raw Remnant - Vegan Health SeminarGranville Glasco
 
You belong here and you can be successful
You belong here and you can be successfulYou belong here and you can be successful
You belong here and you can be successfulcolwilliamson
 
Introduction to Twitter (w/ Allen Klosowski)
Introduction to Twitter (w/ Allen Klosowski)Introduction to Twitter (w/ Allen Klosowski)
Introduction to Twitter (w/ Allen Klosowski)TeachStreet
 
Marketing management
Marketing managementMarketing management
Marketing managementsutrisno2629
 
Assessment
AssessmentAssessment
AssessmentRan Yang
 
Seedlounge 20110126 oj
Seedlounge 20110126   ojSeedlounge 20110126   oj
Seedlounge 20110126 ojOlaf Jacobi
 
Matthew Hamada Timeline
Matthew Hamada TimelineMatthew Hamada Timeline
Matthew Hamada Timelineguest8a85f5
 
RoboCup Introduction
RoboCup IntroductionRoboCup Introduction
RoboCup Introductioncijat
 
Presentation1[1]
Presentation1[1]Presentation1[1]
Presentation1[1]elsiegel
 

En vedette (19)

About Me
About MeAbout Me
About Me
 
Tudlo journey
Tudlo journeyTudlo journey
Tudlo journey
 
Realidad Virtual y sus reglas legales
Realidad Virtual y sus reglas legalesRealidad Virtual y sus reglas legales
Realidad Virtual y sus reglas legales
 
Dades i operadors
Dades i operadorsDades i operadors
Dades i operadors
 
Missing school seminar
Missing school seminarMissing school seminar
Missing school seminar
 
Biografía Pdte
Biografía PdteBiografía Pdte
Biografía Pdte
 
December 2009 TeachStreet Teacher Webinar & Meet-up
December 2009 TeachStreet Teacher Webinar & Meet-upDecember 2009 TeachStreet Teacher Webinar & Meet-up
December 2009 TeachStreet Teacher Webinar & Meet-up
 
Cloenda del Curs d'introducció a Perl 2011
Cloenda del Curs d'introducció a Perl 2011Cloenda del Curs d'introducció a Perl 2011
Cloenda del Curs d'introducció a Perl 2011
 
Search: The Purest Form of Interaction Design
Search: The Purest Form of Interaction DesignSearch: The Purest Form of Interaction Design
Search: The Purest Form of Interaction Design
 
The Raw Remnant - Vegan Health Seminar
The Raw Remnant - Vegan Health SeminarThe Raw Remnant - Vegan Health Seminar
The Raw Remnant - Vegan Health Seminar
 
You belong here and you can be successful
You belong here and you can be successfulYou belong here and you can be successful
You belong here and you can be successful
 
Introduction to Twitter (w/ Allen Klosowski)
Introduction to Twitter (w/ Allen Klosowski)Introduction to Twitter (w/ Allen Klosowski)
Introduction to Twitter (w/ Allen Klosowski)
 
Marketing management
Marketing managementMarketing management
Marketing management
 
Assessment
AssessmentAssessment
Assessment
 
Staying Positive
Staying PositiveStaying Positive
Staying Positive
 
Seedlounge 20110126 oj
Seedlounge 20110126   ojSeedlounge 20110126   oj
Seedlounge 20110126 oj
 
Matthew Hamada Timeline
Matthew Hamada TimelineMatthew Hamada Timeline
Matthew Hamada Timeline
 
RoboCup Introduction
RoboCup IntroductionRoboCup Introduction
RoboCup Introduction
 
Presentation1[1]
Presentation1[1]Presentation1[1]
Presentation1[1]
 

Similaire à CRF-Filters: Discriminative Particle Filters for Sequential State Estimation

Portfolios and Risk Premia for the Long Run
Portfolios and Risk Premia for the Long RunPortfolios and Risk Premia for the Long Run
Portfolios and Risk Premia for the Long Runguasoni
 
Statistical inference for agent-based SIS and SIR models
Statistical inference for agent-based SIS and SIR modelsStatistical inference for agent-based SIS and SIR models
Statistical inference for agent-based SIS and SIR modelsJeremyHeng10
 
Static feedback stabilization of nonlinear systems with single sensor and sin...
Static feedback stabilization of nonlinear systems with single sensor and sin...Static feedback stabilization of nonlinear systems with single sensor and sin...
Static feedback stabilization of nonlinear systems with single sensor and sin...ISA Interchange
 
An Empirical Study of Exposure at Default
An Empirical Study of Exposure at DefaultAn Empirical Study of Exposure at Default
An Empirical Study of Exposure at DefaultMichael Jacobs, Jr.
 
Elastic Modulus And Residual Stress Of Thin Films
Elastic Modulus And Residual Stress Of Thin FilmsElastic Modulus And Residual Stress Of Thin Films
Elastic Modulus And Residual Stress Of Thin Filmserikgherbert
 
Regression Theory
Regression TheoryRegression Theory
Regression TheorySSA KPI
 

Similaire à CRF-Filters: Discriminative Particle Filters for Sequential State Estimation (6)

Portfolios and Risk Premia for the Long Run
Portfolios and Risk Premia for the Long RunPortfolios and Risk Premia for the Long Run
Portfolios and Risk Premia for the Long Run
 
Statistical inference for agent-based SIS and SIR models
Statistical inference for agent-based SIS and SIR modelsStatistical inference for agent-based SIS and SIR models
Statistical inference for agent-based SIS and SIR models
 
Static feedback stabilization of nonlinear systems with single sensor and sin...
Static feedback stabilization of nonlinear systems with single sensor and sin...Static feedback stabilization of nonlinear systems with single sensor and sin...
Static feedback stabilization of nonlinear systems with single sensor and sin...
 
An Empirical Study of Exposure at Default
An Empirical Study of Exposure at DefaultAn Empirical Study of Exposure at Default
An Empirical Study of Exposure at Default
 
Elastic Modulus And Residual Stress Of Thin Films
Elastic Modulus And Residual Stress Of Thin FilmsElastic Modulus And Residual Stress Of Thin Films
Elastic Modulus And Residual Stress Of Thin Films
 
Regression Theory
Regression TheoryRegression Theory
Regression Theory
 

Dernier

Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyKhushali Kathiriya
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodJuan lago vázquez
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoffsammart93
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsNanddeep Nachan
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024The Digital Insurer
 
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...Orbitshub
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Zilliz
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesrafiqahmad00786416
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdfSandro Moreira
 
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfOrbitshub
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...DianaGray10
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...apidays
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfsudhanshuwaghmare1
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...Zilliz
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAndrey Devyatkin
 
Platformless Horizons for Digital Adaptability
Platformless Horizons for Digital AdaptabilityPlatformless Horizons for Digital Adaptability
Platformless Horizons for Digital AdaptabilityWSO2
 
WSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering DevelopersWSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering DevelopersWSO2
 
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...apidays
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
 

Dernier (20)

Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectors
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challenges
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
 
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Platformless Horizons for Digital Adaptability
Platformless Horizons for Digital AdaptabilityPlatformless Horizons for Digital Adaptability
Platformless Horizons for Digital Adaptability
 
WSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering DevelopersWSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering Developers
 
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 

CRF-Filters: Discriminative Particle Filters for Sequential State Estimation

  • 1. CRF-F: D P F  S S E B L, D F  L L Hannes Schulz University of Freiburg, ACS Feb 2008
  • 2. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 3. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 4. Intro Transformation of Directed Model to CRF Application Experimental Results C: S E C D M A  S E ut−2 ut−1 xt−2 xt−1 xt ... ... n 1 2 n zt−1 1 zt 2 zt zt zt−1 zt−1 P (xt |u1:t −1 , z1:t ) = ηP (zt |xt ) P (xt |ut −1 , xt −1 )P (xt −1 |u1:t −2 , z1:t −1 ) dxt −1
  • 5. Intro Transformation of Directed Model to CRF Application Experimental Results D  D M P p (zt |xt ) = n i =1 p (zti |xt ) p (xt +1 |xt , u)
  • 6. Intro Transformation of Directed Model to CRF Application Experimental Results D  D M P p (zt |xt ) = n i =1 p (zti |xt ) p (xt +1 |xt , u) i P (zt |xt) ˆi zt zmax zrand
  • 7. Intro Transformation of Directed Model to CRF Application Experimental Results D  D M P p (zt |xt ) = n i =1 p (zti |xt ) p (xt +1 |xt , u) i P (zt |xt) ˆi zt zmax zrand
  • 8. Intro Transformation of Directed Model to CRF Application Experimental Results D  D M P p (zt |xt ) = n i =1 p (zti |xt ) p (xt +1 |xt , u) i P (zt |xt) ˆi zt zmax δrot2 zrand xt δtrans δrot1 xt−1 u = (δrot1 , δrot2 , δtrans ) executed with gaussian noise
  • 9. Intro Transformation of Directed Model to CRF Application Experimental Results A P   D A p (zti |xt ) are not cond. independent zt xt
  • 10. Intro Transformation of Directed Model to CRF Application Experimental Results A P   D A ut−2 ut−1 xt−2 xt−1 xt p (zti |xt ) are not cond. independent Sensor models can only be ... ... n 1 2 n zt−1 1 2 zt zt−1 zt−1 zt zt generated seperatly for each beam i P (zt |xt) ˆi zt zmax zrand
  • 11. Intro Transformation of Directed Model to CRF Application Experimental Results A P   D A ut−2 ut−1 xt−2 xt−1 xt p (zti |xt ) are not cond. independent Sensor models can only be ... ... n 1 2 n zt−1 1 2 zt zt−1 zt−1 zt zt generated seperatly for each beam Assumption that measurements are independent: “Works i P (zt |xt) ˆi zt zmax surprisingly well”. . . if. . . zrand
  • 12. Intro Transformation of Directed Model to CRF Application Experimental Results A P   D A ut−2 ut−1 xt−2 xt−1 xt p (zti |xt ) are not cond. independent Sensor models can only be ... ... n 1 2 n zt−1 1 2 zt zt−1 zt−1 zt zt generated seperatly for each beam Assumption that measurements are independent: “Works i P (zt |xt) ˆi zt zmax surprisingly well”. . . if. . . increasing uncertainty (tweaking) using every 10th measurement zrand ...
  • 13. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 14. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 15. Intro Transformation of Directed Model to CRF Application Experimental Results I: CRF Undirected graphical models ut−2 ut−1 xt−2 xt−1 xt zt−1 zt
  • 16. Intro Transformation of Directed Model to CRF Application Experimental Results I: CRF Undirected graphical models Every (possible) dependency ut−2 ut−1 represented by edge xt−2 xt−1 xt zt−1 zt
  • 17. Intro Transformation of Directed Model to CRF Application Experimental Results I: CRF Undirected graphical models Every (possible) dependency ut−2 ut−1 represented by edge Distribution defined over products xt−2 xt−1 xt of functions over cliques zt−1 zt
  • 18. Intro Transformation of Directed Model to CRF Application Experimental Results I: CRF Undirected graphical models Every (possible) dependency ut−2 ut−1 represented by edge Distribution defined over products xt−2 xt−1 xt of functions over cliques zt−1 zt Functions are called clique potentials
  • 19. Intro Transformation of Directed Model to CRF Application Experimental Results I: CRF Undirected graphical models Every (possible) dependency ut−2 ut−1 represented by edge Distribution defined over products xt−2 xt−1 xt of functions over cliques zt−1 zt Functions are called clique potentials Clique potentials represent compatibility of their variables
  • 20. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 21. Intro Transformation of Directed Model to CRF Application Experimental Results CRF-M  S E ut−2 ut−1 xt−2 xt−1 xt zt−1 zt T 1 p (x0:T |z1:T , u0:T −1 ) = ϕp (xt , xt −1 , ut −1 )ϕm (xt , zt ) Z (z1:T , u1:T −1 ) t =1
  • 22. Intro Transformation of Directed Model to CRF Application Experimental Results CRF-M  S E ut−2 ut−1 xt−2 xt−1 xt zt−1 zt T 1 p (x0:T |z1:T , u0:T −1 ) = ϕp (xt , xt −1 , ut −1 )ϕm (xt , zt ) Z (z1:T , u1:T −1 ) t =1 Z (·): all trajectories ϕp (·)ϕm (·)
  • 23. Intro Transformation of Directed Model to CRF Application Experimental Results CRF-M  S E ut−2 ut−1 xt−2 xt−1 xt zt−1 zt T 1 p (x0:T |z1:T , u0:T −1 ) = ϕp (xt , xt −1 , ut −1 )ϕm (xt , zt ) Z (z1:T , u1:T −1 ) t =1 Z (·): all trajectories ϕp (·)ϕm (·) How to define ϕp (·) and ϕm (·)?
  • 24. Intro Transformation of Directed Model to CRF Application Experimental Results T P P φp ut −1 = (δrot1 , δtrans , δrot2 ) odometry ut −1 = (δrot1 , δtrans , δrot2 ) derived odometry ˆ ˆ ˆ ˆ δrot2 2 Before: Gaussian noise N uti −1 , σi xt δtrans δrot1 xt−1
  • 25. Intro Transformation of Directed Model to CRF Application Experimental Results T P P φp ut −1 = (δrot1 , δtrans , δrot2 ) odometry ut −1 = (δrot1 , δtrans , δrot2 ) derived odometry ˆ ˆ ˆ ˆ δrot2 2 Before: Gaussian noise N uti −1 , σi xt  (δrot1 − δrot1 )2 ˆ      δtrans   fp (xt , xt −1 , ut −1 ) =  (δtrans − δtrans )2       ˆ      3 features   (δrot2 − δrot2 )2 ˆ     δrot1 xt−1
  • 26. Intro Transformation of Directed Model to CRF Application Experimental Results T P P φp ut −1 = (δrot1 , δtrans , δrot2 ) odometry ut −1 = (δrot1 , δtrans , δrot2 ) derived odometry ˆ ˆ ˆ ˆ δrot2 2 Before: Gaussian noise N uti −1 , σi xt  (δrot1 − δrot1 )2 ˆ      δtrans   fp (xt , xt −1 , ut −1 ) =  (δtrans − δtrans )2       ˆ      3 features   (δrot2 − δrot2 )2 ˆ     δrot1 φp (xt , xt −1 , ut −1 ) = exp wp , fp (xt , xt −1 , ut −1 ) xt−1
  • 27. Intro Transformation of Directed Model to CRF Application Experimental Results T P P φp ut −1 = (δrot1 , δtrans , δrot2 ) odometry ut −1 = (δrot1 , δtrans , δrot2 ) derived odometry ˆ ˆ ˆ ˆ δrot2 2 Before: Gaussian noise N uti −1 , σi xt  (δrot1 − δrot1 )2 ˆ      δtrans   fp (xt , xt −1 , ut −1 ) =  (δtrans − δtrans )2       ˆ      3 features   (δrot2 − δrot2 )2 ˆ     δrot1 φp (xt , xt −1 , ut −1 ) = exp wp , fp (xt , xt −1 , ut −1 ) xt−1 1 (a − a )2 ˆ N a, = exp − σ2 2σ2 Gaussian noise N uti −1 , 1 −2wpi if wp < 0 i
  • 28. Intro Transformation of Directed Model to CRF Application Experimental Results R: S M   N¨ B A  i P (zt |xt) ˆi zt zmax zrand n p (zt |xt ) = p (zti |xt ) i =1
  • 29. Intro Transformation of Directed Model to CRF Application Experimental Results M P φm i P (zt |xt) ˆi zt zmax n     φm (xt , zt ) = exp  wm , fm (zt , xt )  i         i =0 zrand (¬mti ∧ ¬mti )cti (zti − zti )2 ˆ ˆ         i i i        ˆ (¬mt ∧ ¬mt )¬ct      fm (zt , xt ) =  i   (¬mti ∧ mti ) ˆ            ( mti ∧ ¬mti )        ˆ        ( mti ∧ mti ) ˆ  
  • 30. Intro Transformation of Directed Model to CRF Application Experimental Results M P φm i P (zt |xt) ˆi zt zmax n     φm (xt , zt ) = exp  wm , fm (zt , xt )  i         i =0 zrand (¬mti ∧ ¬mti )cti (zti − zti )2 ˆ ˆ         i i i        ˆ (¬mt ∧ ¬mt )¬ct      fm (zt , xt ) =  i   (¬mti ∧ mti ) ˆ            ( mti ∧ ¬mti )        ˆ        ( mti ∧ mti ) ˆ   mti ∈ {1, 0} measured zmax
  • 31. Intro Transformation of Directed Model to CRF Application Experimental Results M P φm i P (zt |xt) ˆi zt zmax n     φm (xt , zt ) = exp  wm , fm (zt , xt )  i         i =0 zrand (¬mti ∧ ¬mti )cti (zti − zti )2 ˆ ˆ         i i i        ˆ (¬mt ∧ ¬mt )¬ct      fm (zt , xt ) =  i   (¬mti ∧ mti ) ˆ            ( mti ∧ ¬mti )        ˆ        ( mti ∧ mti ) ˆ   mti ∈ {1, 0} measured zmax mti ∈ {1, 0} expected zmax ˆ
  • 32. Intro Transformation of Directed Model to CRF Application Experimental Results M P φm i P (zt |xt) ˆi zt zmax n     φm (xt , zt ) = exp  wm , fm (zt , xt )  i         i =0 zrand (¬mti ∧ ¬mti )cti (zti − zti )2 ˆ ˆ         i i i        ˆ (¬mt ∧ ¬mt )¬ct      fm (zt , xt ) =  i   (¬mti ∧ mti ) ˆ            ( mti ∧ ¬mti )        ˆ        ( mti ∧ mti ) ˆ   mti ∈ {1, 0} measured zmax mti ∈ {1, 0} expected zmax ˆ cti ∈ {1, 0} zti − zti < 20cm ˆ
  • 33. Intro Transformation of Directed Model to CRF Application Experimental Results M P φm i P (zt |xt) ˆi zt zmax n     φm (xt , zt ) = exp  wm , fm (zt , xt )  i         i =0 zrand (¬mti ∧ ¬mti )cti (zti − zti )2 ˆ ˆ         i i i        ˆ (¬mt ∧ ¬mt )¬ct      fm (zt , xt ) =  i   (¬mti ∧ mti ) ˆ            ( mti ∧ ¬mti )        ˆ        ( mti ∧ mti ) ˆ   mti ∈ {1, 0} measured zmax mti ∈ {1, 0} expected zmax ˆ cti ∈ {1, 0} zti − zti < 20cm ˆ
  • 34. Intro Transformation of Directed Model to CRF Application Experimental Results M P φm i P (zt |xt) ˆi zt zmax n     φm (xt , zt ) = exp  wm , fm (zt , xt )  i         i =0 zrand (¬mti ∧ ¬mti )cti (zti − zti )2 ˆ ˆ         i i i        ˆ (¬mt ∧ ¬mt )¬ct      fm (zt , xt ) =  i   (¬mti ∧ mti ) ˆ            ( mti ∧ ¬mti )        ˆ        ( mti ∧ mti ) ˆ   mti ∈ {1, 0} measured zmax mti ∈ {1, 0} expected zmax ˆ cti ∈ {1, 0} zti − zti < 20cm ˆ
  • 35. Intro Transformation of Directed Model to CRF Application Experimental Results M P φm i P (zt |xt) ˆi zt zmax n     φm (xt , zt ) = exp  wm , fm (zt , xt )  i         i =0 zrand (¬mti ∧ ¬mti )cti (zti − zti )2 ˆ ˆ         i i i        ˆ (¬mt ∧ ¬mt )¬ct      fm (zt , xt ) =  i   (¬mti ∧ mti ) ˆ            ( mti ∧ ¬mti )        ˆ        ( mti ∧ mti ) ˆ   mti ∈ {1, 0} measured zmax mti ∈ {1, 0} expected zmax ˆ cti ∈ {1, 0} zti − zti < 20cm ˆ
  • 36. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 37. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 38. Intro Transformation of Directed Model to CRF Application Experimental Results U  CRF    P F At each time step t: Prediction Move particles according to gaussian noise determined by wp Same as sampling from N uti −1 , ˆ 1 i −2wp Correction Particle at xt gets weight φm (xt , zt ) Resample (includes normalization)
  • 39. Intro Transformation of Directed Model to CRF Application Experimental Results U  CRF    P F At each time step t: Prediction Move particles according to gaussian noise determined by wp u Same as sampling from N uti −1 , ˆ 1 i −2wp Correction Particle at xt gets weight φm (xt , zt ) Resample (includes normalization)
  • 40. Intro Transformation of Directed Model to CRF Application Experimental Results U  CRF    P F At each time step t: Prediction Move particles according to gaussian noise determined by wp moved Same as sampling from N uti −1 , ˆ 1 −2wpi particles Correction Particle at xt gets weight φm (xt , zt ) Resample (includes normalization)
  • 41. Intro Transformation of Directed Model to CRF Application Experimental Results U  CRF    P F At each time step t: Prediction Move particles according to gaussian noise determined by wp added Same as sampling from N uti −1 , noise 1 ˆ −2wpi Correction Particle at xt gets weight φm (xt , zt ) Resample (includes normalization)
  • 42. Intro Transformation of Directed Model to CRF Application Experimental Results U  CRF    P F At each time step t: Prediction Move particles according to gaussian noise determined by wp ...sense... Same as sampling from N uti −1 , ˆ 1 −2wpi Correction Particle at xt gets weight φm (xt , zt ) Resample (includes normalization)
  • 43. Intro Transformation of Directed Model to CRF Application Experimental Results U  CRF    P F At each time step t: Prediction Move particles according to gaussian noise determined by wp weights Same as sampling from N uti −1 , ˆ 1 −2wpi Correction Particle at xt gets weight φm (xt , zt ) Resample (includes normalization)
  • 44. Intro Transformation of Directed Model to CRF Application Experimental Results U  CRF    P F At each time step t: Prediction Move particles according to gaussian noise determined by wp resample Same as sampling from N uti −1 , ˆ 1 i −2wp Correction Particle at xt gets weight φm (xt , zt ) Resample (includes normalization)
  • 45. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 46. Intro Transformation of Directed Model to CRF Application Experimental Results D     wp  wm Drive around in test area
  • 47. Intro Transformation of Directed Model to CRF Application Experimental Results D     wp  wm Drive around in test area Use high-quality scanmatcher to generate “ground truth” trajectory x∗
  • 48. Intro Transformation of Directed Model to CRF Application Experimental Results D     wp  wm Drive around in test area Use high-quality scanmatcher to generate “ground truth” trajectory x∗ ˆ Using arbitrary weights, generate trajectory x with CRF-filter
  • 49. Intro Transformation of Directed Model to CRF Application Experimental Results D     wp  wm Drive around in test area Use high-quality scanmatcher to generate “ground truth” trajectory x∗ ˆ Using arbitrary weights, generate trajectory x with CRF-filter Use difference of summed features as weight update(−) : wk = wk −1 + α ( f (x∗ , u, z) − f (x, u, z)) ˆ
  • 50. Intro Transformation of Directed Model to CRF Application Experimental Results D     wp  wm Drive around in test area Use high-quality scanmatcher to generate “ground truth” trajectory x∗ ˆ Using arbitrary weights, generate trajectory x with CRF-filter Use difference of summed features as weight update(−) : wk = wk −1 + α ( f (x∗ , u, z) − f (x, u, z)) ˆ Decrease α if new Filter cannot track
  • 51. Intro Transformation of Directed Model to CRF Application Experimental Results D     wp  wm Drive around in test area Use high-quality scanmatcher to generate “ground truth” trajectory x∗ ˆ Using arbitrary weights, generate trajectory x with CRF-filter Use difference of summed features as weight update(−) : wk = wk −1 + α ( f (x∗ , u, z) − f (x, u, z)) ˆ Decrease α if new Filter cannot track loop Adapts weights to task, sensor dependencies/environment, sensor noise, particle filter parameters
  • 52. Intro Transformation of Directed Model to CRF Application Experimental Results L A Averaged Perceptron Algorithm (Collins 2002) for tagging w k = w k −1 + α f (x∗ , u, z) − f (x, u, z) ˆ Proven to converge even in presence of errors in training data Intuition of learning algorithm: If PF works correctly, then f (xn , un−1 , zn ) = ∗ f (xn , un−1 , zn ) ˆ f i occurs less often in x∗ than in x → decrease influence of f i ˆ on particle filter by decreasing w i
  • 53. O 1 I: S E U D M 2 T  D M  CRF Short Introduction to CRF CRF-Model for State Estimation 3 A CRF-Filter Algorithm Learning the Parameters 4 E R
  • 54. Intro Transformation of Directed Model to CRF Application Experimental Results E R Properties of the learned weights Norm of weight vector decreases with number of laser beams in z believes the features/measurements less equivalent to initially introduced “tweaking”?!
  • 55. Intro Transformation of Directed Model to CRF Application Experimental Results E R Properties of the learned weights Norm of weight vector decreases with number of laser beams in z believes the features/measurements less equivalent to initially introduced “tweaking”?! Two specialized CRF-filters compared to generative particle filter trained using expectation maximization Tracking Global Error Localization Accuracy Generative 7.52 cm 30% CRF-Filter 7.07 cm 96%
  • 56. Intro Transformation of Directed Model to CRF Application Experimental Results C 1 A CRF is an alternative, undirected graphical model
  • 57. Intro Transformation of Directed Model to CRF Application Experimental Results C 1 A CRF is an alternative, undirected graphical model 2 CRF-Filters use a continuous CRF for recursive state estimation
  • 58. Intro Transformation of Directed Model to CRF Application Experimental Results C 1 A CRF is an alternative, undirected graphical model 2 CRF-Filters use a continuous CRF for recursive state estimation 3 . . . can be trained to maximize filter performance depending on the task
  • 59. Intro Transformation of Directed Model to CRF Application Experimental Results C 1 A CRF is an alternative, undirected graphical model 2 CRF-Filters use a continuous CRF for recursive state estimation 3 . . . can be trained to maximize filter performance depending on the task 4 . . . can deal with correlated measurements
  • 60. Intro Transformation of Directed Model to CRF Application Experimental Results C 1 A CRF is an alternative, undirected graphical model 2 CRF-Filters use a continuous CRF for recursive state estimation 3 . . . can be trained to maximize filter performance depending on the task 4 . . . can deal with correlated measurements 5 . . . do not explicitly account for dependencies between sensor data