Real-time path optimization of mobile robots based on improved genetic algorithm

Abstract

The emergence of intelligent mobile robots has liberated the human labor to a certain extent, especially their abilities to work in harsh environments in place of humans. For intelligent mobile robots, how to achieve fast path optimization is an important issue. In this article, the model establishment method of environmental information collected by robot sensors and the genetic algorithm for real-time optimization of running paths are briefly introduced first, the crossover, mutation probability, and fitness function are improved based on the shortcomings of the traditional genetic algorithm, and then the simulation analysis of the two algorithms is carried out using matrix laboratory (MATLAB) software. The results show that the improved algorithm obtains a smaller length of optimal path, fewer inflection points, and a smaller turning angle, which also converges faster and has a greater degree of fitness. It takes 0.053 s for the traditional algorithm to calculate the optimal path, while the improved algorithm needs 0.013 s. In summary, the improved genetic algorithm can quickly and efficiently calculate the optimal path, which is suitable for real-time path optimization of mobile robots.