Short-Range Sensing for Intelligent Vehicles: Urban Driving at the DARPA Grand Challenge

This paper presents one of the artificial vision systems developed for the Oshkosh Team's vehicle TerraMax used during the DARPA Urban Challenge 2007. The system is derived from an existing one used as start inhibit and already integrated on Volvo and Iveco prototypes. The aim of the system is to monitor the close proximity of the vehicle up to 10 m. A lane markings detection algorithm has been added to the above mentioned system, which was used for obstacle detection during autonomous urban driving. Laserscanner data fusion developed specifically for the race will be discussed as well. The simplicity of the approach together with the robustness of the technological solution and its reduced cost makes it a surprisingly innovative alternative for short range sensing.