SlideShare une entreprise Scribd logo
1  sur  29
Télécharger pour lire hors ligne
Accelerometer-Based Transportation
Mode Detection on Smartphones
Alice Valentini
alice.valentini7@studio.unibo.it
1
Transportation mode detection
The capability to capture transportation behaviour on smartphones would have an
impact on:
❏ Urban planning
❏ Monitoring and addressing the spread of diseases
❏ Localization and positioning algorithms
❏ Calculating CO2 footprint or level of physical activity
2
Transportation mode detection
Previous studies on transportation mode detection used GPS.
❏ Accurate when signal’s available but not energy-efficient.
❏ Unavailable when view to satellites is obstructed.
❏ Can not make accurate distinction between motorised transportation.
3
Transportation mode detection
Variation in GSM and WiFi signal could be an alternative to GPS.
❏ More energy-efficient.
❏ Susceptible to varying of WiFi access point density and GSM cell size.
❏ Not reliable outside urban areas.
4
Transport mode detection
This study focuses on the use of accelerometer for transport mode detection
because of its properties:
❏ Very low power consumption.
❏ Measures user’s movements directly and does not depend on external sources
❏ Highly detailed information about phone movement.
5
Transportation mode detection
❏ This method uses horizontal and vertical acceleration data.
❏ Different type of vehicles can be identified from their acceleration and braking
periods.
6
How does an accelerometer work?
❏ An accelerometer is a sensor that measures proper acceleration.
❏ They are incorporated in smartphones for detecting the device orientation.
❏ It behaves as a mass on a spring.
7
Preprocessing and gravity estimation
❏ Gravity affects accelerometer signals.
❏ Preprocessing include calculating gravity
eliminated projections of vertical and horizontal
acceleration.
8
Preprocessing and gravity estimation
❏ The algorithm used for estimating the gravity component uses a threshold for
detecting periods where the device is stationary.
❏ The threshold is dynamically adjusted according to the movement patterns as
during activities like walking or bicycling the measurements contain large
variations for a sustained period.
❏ To reduce the influence of orientation changes, the estimate of the gravity
component is reseted when a large shift in orientation is observed.
9
Feature extraction
Features are extracted on three levels of granularity:
❏ Frame-based
❏ Peak-based
❏ Segment-based
10
❏ The frame-based features are able to capture characteristics of
high-frequency motion, e.g., physical movements during pedestrian activity.
❏ From each frame, 27 features are extracted from both vertical and horizontal
representations.
Frame-based features
11
Peak-based features
❏ They capture movements with lower frequencies such as acceleration and
braking periods of motorised vehicles.
❏ These are essential for distinguishing between different motorised
transportation modalities.
❏ To extract these features, peak-areas that correspond to acceleration and
braking periods are identified in the horizontal acceleration projection.
12
Peak-based features
❏ The beginning of a peak-area is identified by first detecting significant
changes in the horizontal acceleration.
❏ The end of a peak-area corresponds in a significant decrease in the horizontal
acceleration.
❏ Once the boundaries have been identified 10 peak features are extracted.
13
Peak-based features
Peak areas detected from gravity eliminated horizontal acceleration during a metro
ride. 14
Segment-based features
❏ Segment-based features characterize patterns of acceleration and
deceleration periods.
❏ Features are:
❏ frequency of acceleration and breaking periods,
❏ frequency and duration of the intermittent stationary periods,
❏ In total 14 segment-based features are considered.
15
Segment-based features
16
Segment-based features
Peak and segment-based features describe the movement patterns of vehicles
instead of those of the users.
17
Classification
18
Full list of the 78 features considered.
Classification
19
The transportation mode detection is decomposed hierarchically into 3 classifiers:
Classification
For classifying a frame classifiers use the machine learning algorithm AdaBoost.
It combines more weak classifiers into one strong classifier.
20
Segment-based classification
❏ It is assumed that within a non-pedestrian segment the transportation modality
remains unchanged.
❏ Therefore, transition between consecutive stationary or motorised
transportation modes requires pedestrian activity.
❏ The transportation modality of the segment is predicted based on its history
using frame and peak features observed.
21
Kinematic Motion classifier
❏ Uses frame-based accelerometer features to distinguish between pedestrian
and other modalities.
❏ The most effective features are variance for both horizontal and vertical
representations.
❏ Accuracy is over 99%.
22
Stationary Classifier
❏ Uses peak and frame-based features for distinguishing between stationary
and motorised periods.
❏ Its precision is over 95%.
23
Motorised Classifier
❏ Distinguishes between transportation modalities.
❏ The correct modality is detected with
approximately 80% precision.
24
Evaluation
25
Evaluation
❏ 67 hours of transportation data were collected among two scenarios.
❏ The mean precision is over 80%. The variance is relatively small.
26
Evaluation
❏ Power consumption
Continuous transportation mode detection
requires minimal power consumption.
The average power consumption of the system is
85 mW.
27
Discussion and improvements
❏ Detection latency is the main limitation and can be reduced using also other
sensors (GPS, GSM or WiFi).
❏ The application could be switched off during extended stationary behaviour.
❏ The system is susceptible to interference from extraneous kinematic events
(user interaction, orientation changes…).
28
References
[1] Hemminki, Samuli, Petteri Nurmi, and Sasu Tarkoma. "Accelerometer-based
transportation mode detection on smartphones." Proceedings of the 11th ACM
Conference on Embedded Networked Sensor Systems. ACM, 2013.
[2] Mizell, David. "Using gravity to estimate accelerometer orientation." Proc. 7th
IEEE Int. Symposium on Wearable Computers (ISWC 2003). Vol. 252. 2003.
[3] Figo, Davide, et al. "Preprocessing techniques for context recognition from
accelerometer data." Personal and Ubiquitous Computing 14.7 (2010): 645-662.
29

Contenu connexe

Tendances

Navigation system by using gis and gps
Navigation system by using gis and gpsNavigation system by using gis and gps
Navigation system by using gis and gpsIAEME Publication
 
PhD Dissertation Proposal
PhD Dissertation ProposalPhD Dissertation Proposal
PhD Dissertation Proposaljairo_sandoval
 
Modal Space Controller for Hydraulically Driven Six Degree of Freedom Paralle...
Modal Space Controller for Hydraulically Driven Six Degree of Freedom Paralle...Modal Space Controller for Hydraulically Driven Six Degree of Freedom Paralle...
Modal Space Controller for Hydraulically Driven Six Degree of Freedom Paralle...Dr. Amarjeet Singh
 
Spot speed studies.ppt .mylan nejyar
Spot speed studies.ppt .mylan nejyarSpot speed studies.ppt .mylan nejyar
Spot speed studies.ppt .mylan nejyarMalika khalil
 
INS/GPS Integrated Navigation Technology for Hypersonic UAV
INS/GPS Integrated Navigation Technology for Hypersonic UAVINS/GPS Integrated Navigation Technology for Hypersonic UAV
INS/GPS Integrated Navigation Technology for Hypersonic UAVNooria Sukmaningtyas
 
5. renewing of indonesia highway capacity manual urban road segment with traf...
5. renewing of indonesia highway capacity manual urban road segment with traf...5. renewing of indonesia highway capacity manual urban road segment with traf...
5. renewing of indonesia highway capacity manual urban road segment with traf...wandi rusfiandi
 
Traffic studies volume study
Traffic studies volume studyTraffic studies volume study
Traffic studies volume studyAglaia Connect
 
Challenging Paraglide Control System
Challenging Paraglide Control SystemChallenging Paraglide Control System
Challenging Paraglide Control SystemCLES-FACIL
 
4. lecture 3 data capturing techniques - total station and gps
4. lecture 3   data capturing techniques - total station and gps4. lecture 3   data capturing techniques - total station and gps
4. lecture 3 data capturing techniques - total station and gpsFenTaHun6
 
L20 Weaving Merging and Diverging Movements
L20 Weaving Merging and Diverging MovementsL20 Weaving Merging and Diverging Movements
L20 Weaving Merging and Diverging MovementsHossam Shafiq I
 
Ieeepro techno solutions 2013 ieee embedded project model predictive contro...
Ieeepro techno solutions   2013 ieee embedded project model predictive contro...Ieeepro techno solutions   2013 ieee embedded project model predictive contro...
Ieeepro techno solutions 2013 ieee embedded project model predictive contro...srinivasanece7
 
Lecture 04 Capacity for TWSC (Traffic Engineering هندسة المرور & Dr. Usama Sh...
Lecture 04 Capacity for TWSC (Traffic Engineering هندسة المرور & Dr. Usama Sh...Lecture 04 Capacity for TWSC (Traffic Engineering هندسة المرور & Dr. Usama Sh...
Lecture 04 Capacity for TWSC (Traffic Engineering هندسة المرور & Dr. Usama Sh...Hossam Shafiq I
 
Spot speed studies and speed delay time survey
Spot speed studies and speed delay time surveySpot speed studies and speed delay time survey
Spot speed studies and speed delay time surveySai Santosh Yakkali
 
碩一工研院研究成果
碩一工研院研究成果碩一工研院研究成果
碩一工研院研究成果Shaun Lin
 
Functional Simulation of the Integrated Onboard System For a Commercial Launc...
Functional Simulation of the Integrated Onboard System For a Commercial Launc...Functional Simulation of the Integrated Onboard System For a Commercial Launc...
Functional Simulation of the Integrated Onboard System For a Commercial Launc...irjes
 
Joe Dodds - Alcohol Consumption on Mode Choice
Joe Dodds - Alcohol Consumption on Mode ChoiceJoe Dodds - Alcohol Consumption on Mode Choice
Joe Dodds - Alcohol Consumption on Mode ChoiceJoseph Chow
 

Tendances (20)

Navigation system by using gis and gps
Navigation system by using gis and gpsNavigation system by using gis and gps
Navigation system by using gis and gps
 
Traffic Speed Analysis
Traffic Speed   AnalysisTraffic Speed   Analysis
Traffic Speed Analysis
 
PhD Dissertation Proposal
PhD Dissertation ProposalPhD Dissertation Proposal
PhD Dissertation Proposal
 
Modal Space Controller for Hydraulically Driven Six Degree of Freedom Paralle...
Modal Space Controller for Hydraulically Driven Six Degree of Freedom Paralle...Modal Space Controller for Hydraulically Driven Six Degree of Freedom Paralle...
Modal Space Controller for Hydraulically Driven Six Degree of Freedom Paralle...
 
Spot speed studies.ppt .mylan nejyar
Spot speed studies.ppt .mylan nejyarSpot speed studies.ppt .mylan nejyar
Spot speed studies.ppt .mylan nejyar
 
INS/GPS Integrated Navigation Technology for Hypersonic UAV
INS/GPS Integrated Navigation Technology for Hypersonic UAVINS/GPS Integrated Navigation Technology for Hypersonic UAV
INS/GPS Integrated Navigation Technology for Hypersonic UAV
 
Monitoring Open Pit
Monitoring Open PitMonitoring Open Pit
Monitoring Open Pit
 
5. renewing of indonesia highway capacity manual urban road segment with traf...
5. renewing of indonesia highway capacity manual urban road segment with traf...5. renewing of indonesia highway capacity manual urban road segment with traf...
5. renewing of indonesia highway capacity manual urban road segment with traf...
 
Rail Deflection
Rail DeflectionRail Deflection
Rail Deflection
 
Traffic studies volume study
Traffic studies volume studyTraffic studies volume study
Traffic studies volume study
 
L4 volume studies
L4 volume studiesL4 volume studies
L4 volume studies
 
Challenging Paraglide Control System
Challenging Paraglide Control SystemChallenging Paraglide Control System
Challenging Paraglide Control System
 
4. lecture 3 data capturing techniques - total station and gps
4. lecture 3   data capturing techniques - total station and gps4. lecture 3   data capturing techniques - total station and gps
4. lecture 3 data capturing techniques - total station and gps
 
L20 Weaving Merging and Diverging Movements
L20 Weaving Merging and Diverging MovementsL20 Weaving Merging and Diverging Movements
L20 Weaving Merging and Diverging Movements
 
Ieeepro techno solutions 2013 ieee embedded project model predictive contro...
Ieeepro techno solutions   2013 ieee embedded project model predictive contro...Ieeepro techno solutions   2013 ieee embedded project model predictive contro...
Ieeepro techno solutions 2013 ieee embedded project model predictive contro...
 
Lecture 04 Capacity for TWSC (Traffic Engineering هندسة المرور & Dr. Usama Sh...
Lecture 04 Capacity for TWSC (Traffic Engineering هندسة المرور & Dr. Usama Sh...Lecture 04 Capacity for TWSC (Traffic Engineering هندسة المرور & Dr. Usama Sh...
Lecture 04 Capacity for TWSC (Traffic Engineering هندسة المرور & Dr. Usama Sh...
 
Spot speed studies and speed delay time survey
Spot speed studies and speed delay time surveySpot speed studies and speed delay time survey
Spot speed studies and speed delay time survey
 
碩一工研院研究成果
碩一工研院研究成果碩一工研院研究成果
碩一工研院研究成果
 
Functional Simulation of the Integrated Onboard System For a Commercial Launc...
Functional Simulation of the Integrated Onboard System For a Commercial Launc...Functional Simulation of the Integrated Onboard System For a Commercial Launc...
Functional Simulation of the Integrated Onboard System For a Commercial Launc...
 
Joe Dodds - Alcohol Consumption on Mode Choice
Joe Dodds - Alcohol Consumption on Mode ChoiceJoe Dodds - Alcohol Consumption on Mode Choice
Joe Dodds - Alcohol Consumption on Mode Choice
 

Similaire à Accelerometer-Based Transportation Mode Detection

gps system and application in mining
gps system and application in mininggps system and application in mining
gps system and application in miningSATYABRATA NAYAK
 
IRJET- Automated Guided Vehicle using Servo Motor in Indoor Positioning S...
IRJET-  	  Automated Guided Vehicle using Servo Motor in Indoor Positioning S...IRJET-  	  Automated Guided Vehicle using Servo Motor in Indoor Positioning S...
IRJET- Automated Guided Vehicle using Servo Motor in Indoor Positioning S...IRJET Journal
 
LANE CHANGE DETECTION AND TRACKING FOR A SAFE-LANE APPROACH IN REAL TIME VISI...
LANE CHANGE DETECTION AND TRACKING FOR A SAFE-LANE APPROACH IN REAL TIME VISI...LANE CHANGE DETECTION AND TRACKING FOR A SAFE-LANE APPROACH IN REAL TIME VISI...
LANE CHANGE DETECTION AND TRACKING FOR A SAFE-LANE APPROACH IN REAL TIME VISI...cscpconf
 
Traffic Signals Safety and Efficiency Project – Stage Two Implementation
Traffic Signals Safety and Efficiency Project – Stage Two ImplementationTraffic Signals Safety and Efficiency Project – Stage Two Implementation
Traffic Signals Safety and Efficiency Project – Stage Two ImplementationJumpingJaq
 
Application of GPS in mining
Application of GPS in miningApplication of GPS in mining
Application of GPS in miningVansajRaj2
 
booysen_vehicle_paper automotive 2015.pdf
booysen_vehicle_paper automotive 2015.pdfbooysen_vehicle_paper automotive 2015.pdf
booysen_vehicle_paper automotive 2015.pdfYogi Adi Wijaya
 
Design of automatic navigation control system for agricultural vehicle
Design of automatic navigation control system for agricultural vehicleDesign of automatic navigation control system for agricultural vehicle
Design of automatic navigation control system for agricultural vehicleeSAT Journals
 
Spot speed study, transport planning
Spot speed study, transport planningSpot speed study, transport planning
Spot speed study, transport planningPratyush Kumar
 
Lane Detection and Obstacle Aviodance
Lane Detection and Obstacle AviodanceLane Detection and Obstacle Aviodance
Lane Detection and Obstacle AviodanceNishanth Sriramoju
 
Lane Detection and Obstacle Aviodance Revised
Lane Detection and Obstacle Aviodance RevisedLane Detection and Obstacle Aviodance Revised
Lane Detection and Obstacle Aviodance RevisedPhanindra Amaradhi
 
paper Presentation
paper Presentationpaper Presentation
paper PresentationPranesh nair
 
COORDINATION OF ACTUATED SIGNALS FOR A CORRIDOR
COORDINATION OF ACTUATED SIGNALS FOR A CORRIDORCOORDINATION OF ACTUATED SIGNALS FOR A CORRIDOR
COORDINATION OF ACTUATED SIGNALS FOR A CORRIDORRakesh Venkateswaran
 
Ijett v5 n1p108
Ijett v5 n1p108Ijett v5 n1p108
Ijett v5 n1p108Athulya MB
 
AGV Presentation (2) (1).ppt
AGV Presentation (2) (1).pptAGV Presentation (2) (1).ppt
AGV Presentation (2) (1).pptBetshaTizazu
 
4. traffic engineering
4. traffic engineering4. traffic engineering
4. traffic engineeringholegajendra
 
An IoT based Dynamic Traffic Signal Control
An IoT based Dynamic Traffic Signal ControlAn IoT based Dynamic Traffic Signal Control
An IoT based Dynamic Traffic Signal ControlGauthamSK4
 
文献紹介2改
文献紹介2改文献紹介2改
文献紹介2改Souhei Hirai
 

Similaire à Accelerometer-Based Transportation Mode Detection (20)

gps system and application in mining
gps system and application in mininggps system and application in mining
gps system and application in mining
 
IRJET- Automated Guided Vehicle using Servo Motor in Indoor Positioning S...
IRJET-  	  Automated Guided Vehicle using Servo Motor in Indoor Positioning S...IRJET-  	  Automated Guided Vehicle using Servo Motor in Indoor Positioning S...
IRJET- Automated Guided Vehicle using Servo Motor in Indoor Positioning S...
 
LANE CHANGE DETECTION AND TRACKING FOR A SAFE-LANE APPROACH IN REAL TIME VISI...
LANE CHANGE DETECTION AND TRACKING FOR A SAFE-LANE APPROACH IN REAL TIME VISI...LANE CHANGE DETECTION AND TRACKING FOR A SAFE-LANE APPROACH IN REAL TIME VISI...
LANE CHANGE DETECTION AND TRACKING FOR A SAFE-LANE APPROACH IN REAL TIME VISI...
 
Traffic Signals Safety and Efficiency Project – Stage Two Implementation
Traffic Signals Safety and Efficiency Project – Stage Two ImplementationTraffic Signals Safety and Efficiency Project – Stage Two Implementation
Traffic Signals Safety and Efficiency Project – Stage Two Implementation
 
Application of GPS in mining
Application of GPS in miningApplication of GPS in mining
Application of GPS in mining
 
booysen_vehicle_paper automotive 2015.pdf
booysen_vehicle_paper automotive 2015.pdfbooysen_vehicle_paper automotive 2015.pdf
booysen_vehicle_paper automotive 2015.pdf
 
Design of automatic navigation control system for agricultural vehicle
Design of automatic navigation control system for agricultural vehicleDesign of automatic navigation control system for agricultural vehicle
Design of automatic navigation control system for agricultural vehicle
 
Spot speed study, transport planning
Spot speed study, transport planningSpot speed study, transport planning
Spot speed study, transport planning
 
Ic engine
Ic engineIc engine
Ic engine
 
Lane Detection and Obstacle Aviodance
Lane Detection and Obstacle AviodanceLane Detection and Obstacle Aviodance
Lane Detection and Obstacle Aviodance
 
Lane Detection and Obstacle Aviodance Revised
Lane Detection and Obstacle Aviodance RevisedLane Detection and Obstacle Aviodance Revised
Lane Detection and Obstacle Aviodance Revised
 
paper Presentation
paper Presentationpaper Presentation
paper Presentation
 
COORDINATION OF ACTUATED SIGNALS FOR A CORRIDOR
COORDINATION OF ACTUATED SIGNALS FOR A CORRIDORCOORDINATION OF ACTUATED SIGNALS FOR A CORRIDOR
COORDINATION OF ACTUATED SIGNALS FOR A CORRIDOR
 
Safer Cycle
Safer CycleSafer Cycle
Safer Cycle
 
Ijett v5 n1p108
Ijett v5 n1p108Ijett v5 n1p108
Ijett v5 n1p108
 
AGV Presentation (2) (1).ppt
AGV Presentation (2) (1).pptAGV Presentation (2) (1).ppt
AGV Presentation (2) (1).ppt
 
4. traffic engineering
4. traffic engineering4. traffic engineering
4. traffic engineering
 
An IoT based Dynamic Traffic Signal Control
An IoT based Dynamic Traffic Signal ControlAn IoT based Dynamic Traffic Signal Control
An IoT based Dynamic Traffic Signal Control
 
文献紹介2改
文献紹介2改文献紹介2改
文献紹介2改
 
Global positioning system( copy1
Global positioning system(   copy1Global positioning system(   copy1
Global positioning system( copy1
 

Dernier

Data governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationData governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationKnoldus Inc.
 
Microservices, Docker deploy and Microservices source code in C#
Microservices, Docker deploy and Microservices source code in C#Microservices, Docker deploy and Microservices source code in C#
Microservices, Docker deploy and Microservices source code in C#Karmanjay Verma
 
Digital Identity is Under Attack: FIDO Paris Seminar.pptx
Digital Identity is Under Attack: FIDO Paris Seminar.pptxDigital Identity is Under Attack: FIDO Paris Seminar.pptx
Digital Identity is Under Attack: FIDO Paris Seminar.pptxLoriGlavin3
 
The Role of FIDO in a Cyber Secure Netherlands: FIDO Paris Seminar.pptx
The Role of FIDO in a Cyber Secure Netherlands: FIDO Paris Seminar.pptxThe Role of FIDO in a Cyber Secure Netherlands: FIDO Paris Seminar.pptx
The Role of FIDO in a Cyber Secure Netherlands: FIDO Paris Seminar.pptxLoriGlavin3
 
Potential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsPotential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsRavi Sanghani
 
2024 April Patch Tuesday
2024 April Patch Tuesday2024 April Patch Tuesday
2024 April Patch TuesdayIvanti
 
The State of Passkeys with FIDO Alliance.pptx
The State of Passkeys with FIDO Alliance.pptxThe State of Passkeys with FIDO Alliance.pptx
The State of Passkeys with FIDO Alliance.pptxLoriGlavin3
 
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxThe Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxLoriGlavin3
 
Glenn Lazarus- Why Your Observability Strategy Needs Security Observability
Glenn Lazarus- Why Your Observability Strategy Needs Security ObservabilityGlenn Lazarus- Why Your Observability Strategy Needs Security Observability
Glenn Lazarus- Why Your Observability Strategy Needs Security Observabilityitnewsafrica
 
UiPath Community: Communication Mining from Zero to Hero
UiPath Community: Communication Mining from Zero to HeroUiPath Community: Communication Mining from Zero to Hero
UiPath Community: Communication Mining from Zero to HeroUiPathCommunity
 
Scale your database traffic with Read & Write split using MySQL Router
Scale your database traffic with Read & Write split using MySQL RouterScale your database traffic with Read & Write split using MySQL Router
Scale your database traffic with Read & Write split using MySQL RouterMydbops
 
Bridging Between CAD & GIS: 6 Ways to Automate Your Data Integration
Bridging Between CAD & GIS:  6 Ways to Automate Your Data IntegrationBridging Between CAD & GIS:  6 Ways to Automate Your Data Integration
Bridging Between CAD & GIS: 6 Ways to Automate Your Data Integrationmarketing932765
 
Tampa BSides - The No BS SOC (slides from April 6, 2024 talk)
Tampa BSides - The No BS SOC (slides from April 6, 2024 talk)Tampa BSides - The No BS SOC (slides from April 6, 2024 talk)
Tampa BSides - The No BS SOC (slides from April 6, 2024 talk)Mark Simos
 
A Glance At The Java Performance Toolbox
A Glance At The Java Performance ToolboxA Glance At The Java Performance Toolbox
A Glance At The Java Performance ToolboxAna-Maria Mihalceanu
 
A Framework for Development in the AI Age
A Framework for Development in the AI AgeA Framework for Development in the AI Age
A Framework for Development in the AI AgeCprime
 
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...BookNet Canada
 
Landscape Catalogue 2024 Australia-1.pdf
Landscape Catalogue 2024 Australia-1.pdfLandscape Catalogue 2024 Australia-1.pdf
Landscape Catalogue 2024 Australia-1.pdfAarwolf Industries LLC
 
Top 10 Hubspot Development Companies in 2024
Top 10 Hubspot Development Companies in 2024Top 10 Hubspot Development Companies in 2024
Top 10 Hubspot Development Companies in 2024TopCSSGallery
 
React Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkReact Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkPixlogix Infotech
 
Infrared simulation and processing on Nvidia platforms
Infrared simulation and processing on Nvidia platformsInfrared simulation and processing on Nvidia platforms
Infrared simulation and processing on Nvidia platformsYoss Cohen
 

Dernier (20)

Data governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationData governance with Unity Catalog Presentation
Data governance with Unity Catalog Presentation
 
Microservices, Docker deploy and Microservices source code in C#
Microservices, Docker deploy and Microservices source code in C#Microservices, Docker deploy and Microservices source code in C#
Microservices, Docker deploy and Microservices source code in C#
 
Digital Identity is Under Attack: FIDO Paris Seminar.pptx
Digital Identity is Under Attack: FIDO Paris Seminar.pptxDigital Identity is Under Attack: FIDO Paris Seminar.pptx
Digital Identity is Under Attack: FIDO Paris Seminar.pptx
 
The Role of FIDO in a Cyber Secure Netherlands: FIDO Paris Seminar.pptx
The Role of FIDO in a Cyber Secure Netherlands: FIDO Paris Seminar.pptxThe Role of FIDO in a Cyber Secure Netherlands: FIDO Paris Seminar.pptx
The Role of FIDO in a Cyber Secure Netherlands: FIDO Paris Seminar.pptx
 
Potential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsPotential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and Insights
 
2024 April Patch Tuesday
2024 April Patch Tuesday2024 April Patch Tuesday
2024 April Patch Tuesday
 
The State of Passkeys with FIDO Alliance.pptx
The State of Passkeys with FIDO Alliance.pptxThe State of Passkeys with FIDO Alliance.pptx
The State of Passkeys with FIDO Alliance.pptx
 
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxThe Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
 
Glenn Lazarus- Why Your Observability Strategy Needs Security Observability
Glenn Lazarus- Why Your Observability Strategy Needs Security ObservabilityGlenn Lazarus- Why Your Observability Strategy Needs Security Observability
Glenn Lazarus- Why Your Observability Strategy Needs Security Observability
 
UiPath Community: Communication Mining from Zero to Hero
UiPath Community: Communication Mining from Zero to HeroUiPath Community: Communication Mining from Zero to Hero
UiPath Community: Communication Mining from Zero to Hero
 
Scale your database traffic with Read & Write split using MySQL Router
Scale your database traffic with Read & Write split using MySQL RouterScale your database traffic with Read & Write split using MySQL Router
Scale your database traffic with Read & Write split using MySQL Router
 
Bridging Between CAD & GIS: 6 Ways to Automate Your Data Integration
Bridging Between CAD & GIS:  6 Ways to Automate Your Data IntegrationBridging Between CAD & GIS:  6 Ways to Automate Your Data Integration
Bridging Between CAD & GIS: 6 Ways to Automate Your Data Integration
 
Tampa BSides - The No BS SOC (slides from April 6, 2024 talk)
Tampa BSides - The No BS SOC (slides from April 6, 2024 talk)Tampa BSides - The No BS SOC (slides from April 6, 2024 talk)
Tampa BSides - The No BS SOC (slides from April 6, 2024 talk)
 
A Glance At The Java Performance Toolbox
A Glance At The Java Performance ToolboxA Glance At The Java Performance Toolbox
A Glance At The Java Performance Toolbox
 
A Framework for Development in the AI Age
A Framework for Development in the AI AgeA Framework for Development in the AI Age
A Framework for Development in the AI Age
 
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
 
Landscape Catalogue 2024 Australia-1.pdf
Landscape Catalogue 2024 Australia-1.pdfLandscape Catalogue 2024 Australia-1.pdf
Landscape Catalogue 2024 Australia-1.pdf
 
Top 10 Hubspot Development Companies in 2024
Top 10 Hubspot Development Companies in 2024Top 10 Hubspot Development Companies in 2024
Top 10 Hubspot Development Companies in 2024
 
React Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkReact Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App Framework
 
Infrared simulation and processing on Nvidia platforms
Infrared simulation and processing on Nvidia platformsInfrared simulation and processing on Nvidia platforms
Infrared simulation and processing on Nvidia platforms
 

Accelerometer-Based Transportation Mode Detection

  • 1. Accelerometer-Based Transportation Mode Detection on Smartphones Alice Valentini alice.valentini7@studio.unibo.it 1
  • 2. Transportation mode detection The capability to capture transportation behaviour on smartphones would have an impact on: ❏ Urban planning ❏ Monitoring and addressing the spread of diseases ❏ Localization and positioning algorithms ❏ Calculating CO2 footprint or level of physical activity 2
  • 3. Transportation mode detection Previous studies on transportation mode detection used GPS. ❏ Accurate when signal’s available but not energy-efficient. ❏ Unavailable when view to satellites is obstructed. ❏ Can not make accurate distinction between motorised transportation. 3
  • 4. Transportation mode detection Variation in GSM and WiFi signal could be an alternative to GPS. ❏ More energy-efficient. ❏ Susceptible to varying of WiFi access point density and GSM cell size. ❏ Not reliable outside urban areas. 4
  • 5. Transport mode detection This study focuses on the use of accelerometer for transport mode detection because of its properties: ❏ Very low power consumption. ❏ Measures user’s movements directly and does not depend on external sources ❏ Highly detailed information about phone movement. 5
  • 6. Transportation mode detection ❏ This method uses horizontal and vertical acceleration data. ❏ Different type of vehicles can be identified from their acceleration and braking periods. 6
  • 7. How does an accelerometer work? ❏ An accelerometer is a sensor that measures proper acceleration. ❏ They are incorporated in smartphones for detecting the device orientation. ❏ It behaves as a mass on a spring. 7
  • 8. Preprocessing and gravity estimation ❏ Gravity affects accelerometer signals. ❏ Preprocessing include calculating gravity eliminated projections of vertical and horizontal acceleration. 8
  • 9. Preprocessing and gravity estimation ❏ The algorithm used for estimating the gravity component uses a threshold for detecting periods where the device is stationary. ❏ The threshold is dynamically adjusted according to the movement patterns as during activities like walking or bicycling the measurements contain large variations for a sustained period. ❏ To reduce the influence of orientation changes, the estimate of the gravity component is reseted when a large shift in orientation is observed. 9
  • 10. Feature extraction Features are extracted on three levels of granularity: ❏ Frame-based ❏ Peak-based ❏ Segment-based 10
  • 11. ❏ The frame-based features are able to capture characteristics of high-frequency motion, e.g., physical movements during pedestrian activity. ❏ From each frame, 27 features are extracted from both vertical and horizontal representations. Frame-based features 11
  • 12. Peak-based features ❏ They capture movements with lower frequencies such as acceleration and braking periods of motorised vehicles. ❏ These are essential for distinguishing between different motorised transportation modalities. ❏ To extract these features, peak-areas that correspond to acceleration and braking periods are identified in the horizontal acceleration projection. 12
  • 13. Peak-based features ❏ The beginning of a peak-area is identified by first detecting significant changes in the horizontal acceleration. ❏ The end of a peak-area corresponds in a significant decrease in the horizontal acceleration. ❏ Once the boundaries have been identified 10 peak features are extracted. 13
  • 14. Peak-based features Peak areas detected from gravity eliminated horizontal acceleration during a metro ride. 14
  • 15. Segment-based features ❏ Segment-based features characterize patterns of acceleration and deceleration periods. ❏ Features are: ❏ frequency of acceleration and breaking periods, ❏ frequency and duration of the intermittent stationary periods, ❏ In total 14 segment-based features are considered. 15
  • 17. Segment-based features Peak and segment-based features describe the movement patterns of vehicles instead of those of the users. 17
  • 18. Classification 18 Full list of the 78 features considered.
  • 19. Classification 19 The transportation mode detection is decomposed hierarchically into 3 classifiers:
  • 20. Classification For classifying a frame classifiers use the machine learning algorithm AdaBoost. It combines more weak classifiers into one strong classifier. 20
  • 21. Segment-based classification ❏ It is assumed that within a non-pedestrian segment the transportation modality remains unchanged. ❏ Therefore, transition between consecutive stationary or motorised transportation modes requires pedestrian activity. ❏ The transportation modality of the segment is predicted based on its history using frame and peak features observed. 21
  • 22. Kinematic Motion classifier ❏ Uses frame-based accelerometer features to distinguish between pedestrian and other modalities. ❏ The most effective features are variance for both horizontal and vertical representations. ❏ Accuracy is over 99%. 22
  • 23. Stationary Classifier ❏ Uses peak and frame-based features for distinguishing between stationary and motorised periods. ❏ Its precision is over 95%. 23
  • 24. Motorised Classifier ❏ Distinguishes between transportation modalities. ❏ The correct modality is detected with approximately 80% precision. 24
  • 26. Evaluation ❏ 67 hours of transportation data were collected among two scenarios. ❏ The mean precision is over 80%. The variance is relatively small. 26
  • 27. Evaluation ❏ Power consumption Continuous transportation mode detection requires minimal power consumption. The average power consumption of the system is 85 mW. 27
  • 28. Discussion and improvements ❏ Detection latency is the main limitation and can be reduced using also other sensors (GPS, GSM or WiFi). ❏ The application could be switched off during extended stationary behaviour. ❏ The system is susceptible to interference from extraneous kinematic events (user interaction, orientation changes…). 28
  • 29. References [1] Hemminki, Samuli, Petteri Nurmi, and Sasu Tarkoma. "Accelerometer-based transportation mode detection on smartphones." Proceedings of the 11th ACM Conference on Embedded Networked Sensor Systems. ACM, 2013. [2] Mizell, David. "Using gravity to estimate accelerometer orientation." Proc. 7th IEEE Int. Symposium on Wearable Computers (ISWC 2003). Vol. 252. 2003. [3] Figo, Davide, et al. "Preprocessing techniques for context recognition from accelerometer data." Personal and Ubiquitous Computing 14.7 (2010): 645-662. 29