ProFusion

For several years research and development (R&D) activities have been keeping on developing and improving preventive safety equipment and functions. More recent advances have led to an ‘agreement’ on the utilisation of common sets of sensors for distinct safety applications and to sensor data fusion for the improvement of performance and robustness of functions.

As IP PReVENT addresses all preventive safety issues and develops functions that widely rely on multi-sensor approaches and sensor data fusion, it is planned to have cross-functional activities that ensure and urge use of multi-sensor systems beyond the current state-of-the-art. 

This is the role of ProFusion subproject, which follows a two-step procedure: during a short initial phase of 6 months, the subproject ProFusion I was in close connection to all vertical subprojects (VSPs) in PReVENT that lead activities related to sensors and sensor data fusion. At the end of this preparatory phase, recommendations for a proposal on sensors and sensor data fusion concepts were given. In a second phase, the R&D-oriented project ProFusion II is focusing on research work of common interest in the field of sensors and sensor data fusion. 

Objectives

Key objectives of ProFusion1 have been:

• Create and lead a forum on sensors and sensor data fusion within PReVENT. This forum will be the centre of expertise and focal point for exchanges relating to these topics, with information flowing from and to vertical application oriented subprojects;
• Gather information on the state-of-the-art, analyse the abilities, limitations and orientations chosen by vertical subprojects, concerning sensors and low/high level sensor data fusion; complement information collected with surveys, reviews, concepts from research, synthesize results and issue reports on state-of-the-art;
• Identify and define needs for original resarch work(s) of common interest for the Preventive Safety community in the field of sensors and sensor data fusion, issue guidelines and recommendations to be used by ProFusion II.      

Key concepts

The major objective of ProFusion is to push the sensor data fusion techniques used for driver assistance systems beyond the current state-of-the-art. This will be done by setting up a modular and interoperable fusion architecture for multi sensor systems, which integrates diverse approaches (e.g. low and high sensor data fusion, algorithms for situation refinement, etc.). Below, the figure sketches roughly the concept (ProFusion I) of the development work (ProFusion II):

• The system will be based on an extended environment model. Objectives are to deliver an extended view of the environment to applications (in terms of more detailed representations) and to define data structures for each processing step during the fusion process (e.g. for raw sensor data, signal features, objects detected in the environment);
• For object refinement, ProFusion will develop existing and new algorithms for filtering, data association and object classification. Especially, will analyse chances for deducing confidence information from complementary and redundant sensor data for each processing result. Additionally, the research work will be focused on new motion models;
• Research topics on situation refinement will include trajectory classification and prediction for objects on the road, trajectory estimation of the ego-vehicle based on multi source information (e.g. detected lanes, ego-vehicle dynamics and map data), as well as on generic decision components for the  prediction of the driver’s intention.

The newly developed perception system, and especially the fusion approaches, will be tested extensively based on existing demonstrators taken from vertical subprojects. The results will be compared to the performance of environment perception achieved in the VSPs to show benefits and drawbacks of the new approaches.

Results

ProFusion I has contributed to the IP PReVENT results with three reports on the requirements and state-of-the-art of sensors and sensor data fusion, as well as recommendations for further research in the field of SDF.

Four main reports have been published by ProFusion:

• Requirements from PReVENT subprojects for sensor and sensor data fusion
• Compendium on state-of-the-art of Sensors
• Compendium on state-of-the-art of Sensor data fusion
• Recommendations for further innovative research

Contact Information

ProFusion1

SP Leader
Olivier Aycard
INRIA Rhone-Alpes

Tel : +33 4 76 61 55 14

e-mail : olivier.aycard@inrialpes.fr

Consortium Partners

DaimlerChrysler AG (ProFusion1 & 2)			BMW Forschung Und Technik GmbH (ProFusion1)
Universität Passau (ProFusion1 & 2)			Institute of Communication and Computer Systems (ProFusion2)
INRIA (ProFusion1 & 2)			Peugeot Citroën Automobiles SA (ProFusion1)
Centro Ricerche Fiat (ProFusion1 & 2)			Volvo Technology Corporation (ProFusion2)
Delphi (ProFusion1 & 2)			Ibeo Automobile Sensor GmbH (ProFusion2)
SAGEM			Technische Universitaet Chemnitz (ProFusion2)

To the top

You are here: www.prevent-ip.org HORIZONTAL ACTIVITIES ProFusion1