Autonomous driving in complex urban environments, including traffic merge, four-ways stop, overtaking, etc., requires a very wide range sensorial capabilities, both in angle and distance. This paper presents a vision system, designed to help merging into traffic on two-ways intersections, and able to provide a long detection distance (over 100m) for incoming vehicles. The system is made of two high resolution wide angle cameras, each one looking laterally (70 degrees) with respect of the moving direction, performing a specific background subtraction based technique, along with tracking and speed estimation. The system works when the vehicle is stopped at intersections, and is triggered by the high-level vehicle manager. The system has been developed and tested on the Oshkosh Team’s vehicle TerraMax™, one of the 11 robots admitted to the DARPA Urban Challenge 2007 Final Event.
Lateral vehicles detection using monocular high resolution cameras on TerraMax™
2008 IEEE Intelligent Vehicles Symposium ; 1143-1148
01.06.2008
327267 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Lateral Vehicles Detection Using Monocular High Resolution Cameras on TerraMaxTM
British Library Conference Proceedings | 2008
|The TerraMax Autonomous Vehicle
Springer Verlag | 2007
|The TerraMax autonomous vehicle
British Library Online Contents | 2006
|Speed Estimation of Vehicles Using Monocular Cameras
IEEE | 2023
|TerraMax Vision at the Urban Challenge 2007
IEEE | 2010
|