Adaptive neurons based control system design for mobile robots

A novel tracking control scheme for nonholonomic mobile robots is developed, which integrates two adaptive neurons or non-adaptive neurons, backstepping technique and sliding mode. The proposed control system includes a velocity controller embedded with two biological neurons and a torque controller based on sliding mode. An adaptive neuron model is presented to extend the functions of the neuron model. Two adaptive neurons are embedded into the backstepping-based velocity controller to improve the tracking performance of the proposed control scheme. The proposed systems can completely eliminate the sharp speed jumps existing commonly in the mobile robot systems and is able to ensure the mobile robot to navigate safely under any circumstance. The simulation results demonstrate the effectiveness and efficiency of the proposed control scheme.