A FUZZY-LOGIC APPROACH FOR SAFETY AND COLLISION-AVOIDANCE IN ROBOTIC SYSTEMS

A major factor that has limited the application of robots in industria l and human service applications has been the lack of robust sensing a nd control algorithms for detection and prevention of collision condit ions. This paper discusses an approach to the collision avoidance cont rol of robots using a fuzzy logic methodology for the integration of s ensory input data from the robot's environment. The paper presents a f ormulation of the collision avoidance problem using the occupancy grid formulation, and discusses the use of a combination of Dempster-Shafe r inference and fuzzy logic in fusing the sensory information and maki ng robot movement decisions. This hybrid approach utilizes the strengt hs of both systems to provide an effective and computationally tractab le result. (C) 1995 John Wiley & Sons, Inc.