Gray wolf optimization algorithm has performance defects such as slow convergence speed, easy to fall into local optimum and high processing complexity. In this paper, an improved gray wolf optimization algorithm (IGWO) is proposed with reference to the improvement idea of whale optimization algorithm. Aiming at the shortcomings of the uneven distribution of individuals in the population initialization of the grey wolf optimization algorithm, a chaotic mapping is introduced to initialize the population, and based on the randomness and regularity of the chaotic mapping, the global search process of the algorithm in this paper is optimized. In addition, an exponential convergence factor is proposed to update the control parameters of the algorithm. Then, a nonlinear convergence parameter is introduced to change the updating method of the iterative formula to optimize the algorithm, and it is applied to artificial intelligence. The research results show that the clustering accuracy of the improved Gray Wolf Optimization Algorithm for a variety of classical functions is maintained at more than 90%, the convergence accuracy and performance are better than the comparison algorithm, and it is able to complete the multi-intelligent path planning under different experimental environments and different number of intelligences control conditions, which verifies the feasibility of the algorithm proposed in this paper in solving the optimization problem of the Gray Wolf Optimization Algorithm.
