Design and Implementation of Neural Network Based Non Linear Control System (LQR) for Target Tracking Mobile Robots

This research work involves the speed control of a mobile robot where stabilization is provided by Non-linear controller. A Mobile Robot capable of following human target, assisted by various sensors that detects the clearance distance between the target and the robot to avoid collision or to maintain constant distance needs a robust control system. A robot in order to follow a human or target which has variable speed must have decision capabilities to decide when to speed up or slow down according to the situation. This decision can be obtained by the use of Artificial Neural Network (ANN) which reduces the burden on the controller reducing the complexity of calculation involved to control the speed of the motors. All these above sensors and actuators provide values as input to the Neural Network database for learning stage providing classification into several actions. The classes and weights obtained by the Neural Network learning phase is then used to vary the speed of the Robot. The speed however being unstable and providing several unnecessary movements will then be stabilized by using Linear Quadratic Regulator control algorithm. Thus, providing a complete solution to setting a fluent and stable Mobile Robot that can track and follow and target under any conditions.