Extension of Multi-Object-Tracking supporting Cooperative Perception

• Applicant: Professor Dr.-Ing. Klaus Dietmayer
• Subject Area: Automation, Mechatronics, Control Systems, Intelligent Technical Systems, Robotics ; Traffic and Transport Systems, Intelligent and Automated Traffic
• Term: from 2015 to 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 272101568

In order to realize highly automated driving functions in future vehicles, a cooperative perception of the vehicles environment is required. During the last years, several approaches for cooperative perception have been presented. However, all of them assume the road users to be statistically independent. Especially in situations with short drop-outs of measurements or transmitted information, this assumption may lead to physically impossible or not meaningful constellations of the multi-object state. In this project, the recently proposed Labeled Multi-Bernoulli Filter of the research group will be enhanced to facilitate the integration of object interactions in a mathematically well-founded way. Further, suitable approaches to model object interactions in traffic scenarios have to be derived and the performance gain of modelling statistical dependencies will be evaluated.

• DFG Programme: Priority Programmes
• Subproject of: SPP 1835: Kooperativ interagierende Automobile