H//sub/spl infin// filter in autonomous robot navigation

A robust H_infin filtering is applied to an autonomous robot navigation system for eliminating the uncertainty noise of sensors. When all the noise statistical characteristic and system model are available, Kalman filtering is an efficient algorithm. Actually, there exist many uncertain factors in real operation process of a robot. The worst is that the filter is not convergence. Simulation results proved H_infin filtering can lower divergence, and then reduce loss of the robot as well as improve on-the-job dependability. Simulation results are given in the paper. A robot equipped with ultrasonic and laser ranger, Web-camera was used for visual tracing. Encoder is used for the dead-reckoning and correction