Title :
Monte Carlo road safety reasoning
Author :
Broadhurst, Adrian ; Baker, Simon ; Kanade, Takeo
Author_Institution :
Robotics Inst., Carnegie Mellon Univ., Pittsburgh, PA, USA
Abstract :
This paper presents a framework for reasoning about the future motion of multiple objects in a road scene. Unlike previous approaches, we do not look for known dangerous configurations of objects, but rather we reason about the future paths of all objects in the scene, and assess their danger. Monte Carlo path planning is used to generate a probability distribution for the possible future motion of every car in the scene. This framework can be used to either control the car, or to display warnings for the driver.
Keywords :
Monte Carlo methods; object detection; path planning; probability; road safety; road traffic; road vehicles; Monte Carlo path planning; Monte Carlo road safety reasoning; car control; probability distribution; Humans; Laser radar; Layout; Monte Carlo methods; Path planning; Probability distribution; Road accidents; Road safety; Robots; Uncertainty;
Conference_Titel :
Intelligent Vehicles Symposium, 2005. Proceedings. IEEE
Print_ISBN :
0-7803-8961-1
DOI :
10.1109/IVS.2005.1505122