ABSTRACT
This paper investigates the real-time optimal obstacle avoidance of a micro ground vehicle (MGV) using the modified Carathéodory-π solutions. The optimal control problem is sequentially solved by the pseudospectral optimal control theory as a minimum-time point-to-point maneuver. The MGV uses the generated optimal control law and optimal trajectory for its maneuver. When the vehicle approaches an obstacle, the cost function’s value increases. The proposed approach is experimentally validated by controlling the real MGV in the indoor facility. Experimental results demonstrate the feasibility of commanding the MGV with the optimal solutions generated by this method.
