Trajectory Tracking of a Mobile Robot Using a PID Controller Combined with Neural Networks

Abstract

In this document, a cascade control strategy is proposed for trajectory tracking of a mobile robot. In the proposed scheme, the objective of the outer loop is to reduce position error by using a model-based kinematic controller; and, in the other hand, the inner loop is responsible for controlling the linear and angular velocity of the robot using a neural controller combined with a PID to correct possible changes dynamics of the robot. The neural networks that are applied to dynamic system, undergoes a parameter identification step through neural network. This controller is updated at every sample time and collaborates with a PID controller to train network. Finally, a comparative analysis is carried out between a classical PID and proposed controller, both qualitatively and quantitatively, showing that the proposed approach significantly improves the performance of the system in the task of trajectory tracking.