Knowledge-Based Navigation for Autonomous Road Vehicles

This paper presents a computer vision system for an autonomous road vehicle (ARV) that is capable of negotiating complex road networks including road junctions in real time. The ultimate aim of the system is to enable the vehicle to drive automatically along a given complex road network whose geometric description is known. This computer vision system includes three main techniques which are necessary for an ARV: a) road following, b) road junction detection, c) manoeuvring at the road junction. The road following algorithm presents a method of executing a number of algorithms using different methods concurrently, fusing their outputs together into an accurate road model. A model-based object approach is used for detecting the road junctions in images. In this approach two sequence processes are performed. They are to find a boundary between the candidate road junction surface and the current road surface, and to locate the road junction surface. A multi-camera vision-based re-orientation mode was used to guide the ARV during the manoeuvring process at the road junction. In the re-orientation mode, the position of the ARV with respect to the road junction is determined by the “bootstrap” process. The results are presented for real road stretches and intersection images and performed on the experimental autonomous road vehicle in real time.