A combined generalized potential fields / dynamic pathplanning approach to collision avoidance for a mobile autonomous robot operating in a contrained environment

The HASREG collision avoidance path planning system is presented. The algorithm combines elements of the Generalized Potential Fields method with a non-heuristic Dynamic Path Planning algorithm. The vehicle's dynamics and kinematics are accounted for, and goal reachability is determined. The HASREG system consists of four independent, yet communicating, modules: MEAS-STRAT coordinates sensor measurements, BUG generates goal positions, GPF computes avoidance and goal vectors, and SA updates the vehicle state. The system has been implemented on the IPAMAR mobile robot, and results show the feasibility of this approach.