Fuzzy-PID hybrid controller for mobile robot using point cloud and low cost depth sensor

In mobile robots, motion control systems play an important role to assume trajectory planning and obstacle avoidance. Proportional-Integral-Derivative (PID) controllers are the most popular controller used in industrial control systems including mobile robots. The PID controller is developed based on the linear control theory but it gives inconsistent performance for different condition. In order to overcome this problem, we propose a Fuzzy-tuned PID controller in which the PID parameters are learned, adapted and changed thanks to the fuzzy system. The PID inputs are given by the Kinect sensor after being processed by the point cloud library. The effectiveness of this method is evaluated experimentally in real time using the mobile robot iRobot Create.