Dynamic path planning by using dual-layer fuzzy controllers

雷艳敏,朱齐丹,关秀丽,冯志彬
DOI: https://doi.org/10.13245/j.hust.2011.09.014
2011-01-01
Abstract:To solve path planning problem of multi-robot systems with no communication in unknown dynamic environment, a dual-layer fuzzy controller was proposed. Danger degree fuzzy and velocity fuzzy controllers were designed in this controller. The velocity information of the moving obstacles was fully considered in the danger fuzzy controller, and it was more appropriate for dynamic environ- ment. Based on the time and distance factors of the collision, the danger degree of collision denoted the possibility that the robot collided with obstacles. The velocity fuzzy controller took full account of target azimuth, obstacle azimuth and danger degree of collision and designed fuzzy rules using behav- ior ideas. Fuzzy rules reflected the move-to-goal behavior, avoid-obstacle behavior and follow-obstacle behavior, and made velocity fuzzy controller suit to the complicated environment. Simulation results show that the proposed method is feasible and valid.
What problem does this paper attempt to address?