Optimization Design for PID Parameter of Mobile Robot Based on Genetic Algorithm

Focusing on the control requirements for position and speed of mobile robot, PID controller parameters of mobile robot based on genetic algorithm are designed and optimized. In this paper, the performance index for the time integral of absolute error value is adopted as the minimum object function to choose parameters, and the global searching capability of genetic algorithm is used to obtain optimal solution and further to reduce difficulty in tuning PID parameters under the no prior knowledge condition. The simulation results show that the optimization design for PID parameters proposed in this paper improve the control precision and robustness of the system.

1. Introduction

The mobile robot is a typical non-holonomic control system with non-holonomic constraints, and it is also a comprehensive system with many functions as environment perceptions, dynamic programming and planning, and behavior control and implementation etc. The mobile robot system requires stricter control for position and speed, thereby adopting close-loop control. PID control is one of the most widely applied strategies. Therefore, conventional robot motion control often adopts PID controller, and many corresponding design methods appear, such as adaptive fuzzy PID control, neural network predictive control, fuzzy neural network control, and so on.