In this paper, the results from a series of experiments using the virtual bumper collision avoidance algorithm implemented on a Navistar tractor cab are described. The virtual bumper combines longitudinal and lateral collision avoidance capabilities to control a vehicle in normal and emergency situations. A programmable boundary, the virtual bumper, defines a personal space around the host vehicle. A radar and a laser range sensor were used to sense the location of vehicles in the region in front of the truck. Incursions into the personal space by target vehicles impose a virtual 'force' on the host using impedance control, which in turn modifies the vehicle's trajectory in order to avoid collisions with objects in the field of view. The virtual bumper longitudinal controller was tested under several driving situations and at several speeds. Several scenarios were evaluated. The results from two experiments are described. These involve the truck performing a critical stop when the target vehicle ahead is stopped, and a situation in which the host must maintain a safe headway in a major traffic slow down. The algorithm demonstrated robustness to sensor noise and the ability to maintain a safe headway for both normal and emergency driving scenarios. (A*)
Abstract