Clonal selection algorithm based control for two-wheeled self-balancing mobile robot

Abstract

In this study, we propose adaptive control approaches based on optimization for a two-wheeled mobile robot. The mathematical model of the self-balancing vehicle based on the inverted pendulum principle, is derived from the Lagrange method and the vehicle model is simulated within MATLAB/Simulink environment. Five different optimization-based controller models (Model 1–5) are developed based on common control methods, which are the Pole Placement Design Method, PID, and LQR. Besides the stability control of the robot, speed control and trajectory tracking control are also provided with these models. While improving the optimization-based controller models, the CSA, an artificial immune optimization methods, is adapted for use with the system. The benefits of the study are given as follows: 1) The balancing control and trajectory tracking control performance of the two-wheeled vehicle is improved using the proposed optimization-based control approach; 2) A comprehensive study is introduced with different control models based on Pole Placement Design Method, PID, and LQR Method; 3) Since parameter tuning of multiple PID controllers is challenging for a two-wheeled vehicle constructed by using the inverted pendulum principle, one of the proposed control models is developed using multiple PIDs. To alleviate the difficulties of parameter tuning for multiple PIDs, the CSA method is adapted for a self-balancing two-wheeled vehicle for the first time; 4) To exhibit the optimization effect of the CSA, the simulation results are compared with five other optimization methods, namely PSO, ABC, DE, GoldSa-II, and CS; 5) The effectiveness of the proposed optimization-based control approaches and optimization method are demonstrated through analyses and simulation studies. In this concept, it has been also analyzed effects of the external disturbances and uncertain parameters on the proposed control method.