An environmental perception system to autonomous off-road navigation by using multi-sensor data fusion

Environmental perception is one of the most difficult problems on the research of off-road autonomous vehicles. This paper describes a multi-sensor data fusion based environmental perception system for off-road autonomous navigation. The system is composed of one camera, four laser range finders, one microwave radar, and several ultrasonic sensors. A hierarchical structure is used to organize the sensors from feature level to high fusion level. A 2D fusion map consisting of six classifications is proposed to describe the surrounding environment. Dempster-Shafer evidence theory is adopted to decide the classification of each grid in the fusion map. A weighted evidence combination rule is proposed and implemented to improve the decision results under high conflicting circumstance. The experimental results show that the proposed system is reliable and is able to fulfil the requirements from off-road navigation.