Robot navigation in a known environment with unknown moving obstacles

The authors propose a type of artificial potential field, a hybrid potential field, to navigate a robot in situations in which the environment is known except for unknown and possibly moving obstacles. The authors show how to compute hybrid potential fields in real time, and use them to control the motions of a real robot. Their method is tested on both a simulated and a real robot. Extensive simulation results are given with up to 50 randomly moving obstacles. A feature matching approach is presented for error position correction. It is validated experimentally with the mobile robot.<
>