AN IMPROVED RADIAL BASIS FUNCTION NETWORK FOR VISUAL AUTONOMOUS ROAD FOLLOWING

We have developed a radial basis function network (RBFN) for visual au tonomous road following, Preliminary testing of the RBFN was done usin g a driving simulator, and the RBFN was then installed on an actual ve hicle at Carnegie Mellon University for testing in an outdoor road-fol lowing application, In our first attempts, the RBFN had some success, but it experienced some significant problems such as jittery control a nd driving failure, Several improvements have been made to the origina l RBFN architecture to overcome these problems in simulation and more importantly in actual road following, and the improvements are describ ed in this paper.