Enhancing Tilt-Integral-Derivative Controller to Motion Control of Holonomic Wheeled Mobile Robot by Using New Hybrid Approach

Abstract

Trajectory tracking of wheeled mobile robot is the most interesting and important subject in robotic systems field. In this paper, a four mecanum wheeled mobile robot (FMWMR) has been considered. This type was considered because it is the most famous holonomic type of wheeled mobile robot (WMR). The main purpose of this work is to design a new hybrid controller for the trajectory tracking of FMWMR based on the kinematic model. The proposed controller consists of Tilt-Integral-Derivative (TID) controller tuned parameters with neural network as well as social spider optimization i.e., (TID-NN-SSO) which is applied on the kinematic model of the FMWMR. The forward and inverse kinematic equations were derived. The TID controller was used for computing the controlled signals which are the angular velocities of each wheel while the neural network (NN) and social spider optimization (SSO) were to compute the parameters of TID controller. MATLAB/Simulink programing was implemented to simulate the results. A comparative study between TID-NN-SSO and TID controller tuned parameters by using particles swarm optimization (TID-PSO) was conducted. The results of the mean square errors (MSE) in (x and y) coordinates as well as the orientation error obtained from TID–NN-SSO are (2.105*10−4 m, 1.025*10−4 m, 0.0815 rad) and they are less than the MSE that was obtained from TID-PSO which are (0.00567 m, 0.0356 m, 1.22 rad/s). This indicates that TID–NN-SSO is more efficient than TID-PSO.