A Single-ball-driven Self-Balancing Robot Controller Based on Genetic Algorithm Optimization.

This paper proposes a self-balancing control method based on genetic algorithm optimization in order to further improve the self-balancing ability of robots. Firstly, the robot system is simplified to a 2D single-ball inverted pendulum model, and an Eular-Lagrange dynamic model is established for mechanical analysis. Then, the self-balancing controller is designed by combining the motion control law of the robot and the linear quadratic optimal control algorithm based on genetic algorithm. The normalized quadratic performance index is taken as the optimization standard, and the weight matrix is optimized by genetic algorithm. Finally, the optimized controller is used to simulate the origin self-balancing control. The simulation results show that the optimized controller not only improves the response speed and robustness of the system, but also effectively reduces the power consumption of the controller.