A NEURAL-NETWORK APPROACH FOR SAFETY AND COLLISION-AVOIDANCE IN ROBOTIC SYSTEMS

A major factor which has limited the application of robots in industri al and human service applications has been the lack of robust sensing and control algorithms for detection and prevention of collision condi tions. This paper discusses an approach to the collision avoidance con trol of robots using a neural network methodology for the integration of sensory input data from the robot's environment. The paper presents a formulation of the collision avoidance problem using the occupancy grid formulation, and discusses the use of a combination of Dempster-S hafer inference and neural networks in fusing the sensory information and making robot movement decisions. Initial studies have shown this a pproach to be both robust and computationally tractable in providing e nhanced safety capabilities. (C) 1996 Elsevier Science Limited.