To appear in Autonomous Robots Learning to Perceive Objects for Autonomous Navigation

Current machine perception techniques that typi cally use segmentation followed by object recognition lack the required robustness to cope with the large variety of situations encountered in real world naviga tion Many existing techniques are brittle in the sense that even minor changes in the expected task environ ment e g di erent lighting conditions geometrical distortion etc can severely degrade the performance of the system or even make it fail completely In this paper we present a system that achieves robust perfor mance by using local reinforcement learning to induce a highly adaptive mapping from input images to seg mentation strategies for successful recognition This is accomplished by using the con dence level of model matching as reinforcement to drive learning Local reinforcement learning gives rises to better improve ment in recognition performance The system is veri ed through experiments on a large set of real images of tra c signs