Tulipp collaboration Workshop - Advanced Computing and CPS - June 2016
This project has received funding from the European Union’s Horizon 20 20 research and innovation programme under grant agreement No 688403
Diana Göhringer – Ruhr-University Bochum
Advanced Computing and CPS
TULIPP - Towards Ubiquitous Low-power Image Processing Platforms - is a research project that creates reference design rules to guide and facilitate the development of power efficient image processing platforms, where platform means hardware, operating system and toolchain together with standardized interfaces.
The technical and commercial criteria for selecting a processor for TULIPP were difficult. Each individual of the consortium had preferences and ideas for what was not only the best candidate based on prediction what the future will bring in terms of semiconductor technology, but also had to be commercial available (or at least visible) today. It has to be available to anybody in any low-volume “Use Cases” that are not even covered during the project. The choice(s) is overwhelming as below picture can illustrate. They are the best (not all) of the semiconductor manufactures that offers processors today that would be a candidate for being a TULIPP processor
The main objective of Tulipp is to provide at the end of the project a reference platform through associated guidelines so that image processing board manufacturers could develop a low-power high performance computing system with the expected performances. The project will concentrate on interfaces between the components of the platform as well as design and implementation rules. Following the rules, anyone will be able to produce a compliant platform and benefit from the technological advances generated by the project. The interfaces will be defined, so as to enable any industrial to produce any sub-part of the platform and plug it to other existing platform so as to create a new Tulipp platform instance.
Bring intelligence to the dronesThe TULIPP solution aims to fill this processing gap by using its good performance to weight and power consumption to weight figures.In this UAV use case we plan to use computer vision algorithms such as depth estimation and obstacle detection to evaluate the surroundings and make the UAV more intelligent.
Safer driving experienceADAS vision systems require real time, low latency processing, at high to very high computational load. They need to be robust and reliable, and will often be treated as safety critical systems. The Tulipp project addresses all these questions – by offering a toolset and standardization, it will help the designers to focus on the image processing application rather than platform details.
Reduce radiation dose by 75%In the Tulipp project our aim is to try and reduce the level of radiation by 75%. As a result of this, more powerful image processing will be required in order to still be able to see many of the small details in the human body that are crucial during surgery. Since most operating rooms are small, the device needs to be small and mobile. A system that integrates the processing close to the sensor would be ideal to help reduce extraneous wires
The Advisory Board Members will be involved in the TULIPP project on a non-exclusive and volunteering basis with no financial contributions towards cost of time, efforts and travel by any partners. The Advisory Board Member has no liabilities and/or obligations towards the TULIPP consortium. Confidential information will be exchanged between the Advisory Board Member and the TULIPP consortium once an appropriate non-disclosure agreement is signed. The TULIPP consortium selects the Advisory Board Member based on ‘Best-Fit-Approach’ to support the development of the platform and is limited to 20 participants. The Advisory Board Member may leave at any time and the TULIPP consortium may ask the Advisory Board Member to leave without any financial obligations on any parties.
The Advisory Board Member will recommend the TULIPP partners on ways to improve and help the TULIPP project towards normalization and standardization of its reference platform guideline for the power-efficient image processing domain. The Advisory Board Member will be kept informed about progress and consulted at every step of the project to review and evaluate the reference platform guideline in respect to pre-standardization and help in promoting the reference platform as a standard.