DDPG-Based Improved Seeker Optimization Algorithm for Robot Path Planning

Abstract

Robot path planning is one of the core technologies for mobile robot applications. In path planning, the traditional seeker optimization algorithm (SOA) suffers from insufficient intelligence, slow convergence, and poor solving ability in robot path planning. Therefore, this paper proposes a deep deterministic policy gradient (DDPG)-based improved seeker optimization algorithm (DDPGSOA). The DDPGSOA is successfully used in the robot path planning problem. First, this paper introduces robot map modeling and path modeling. Then, the shortcomings of the traditional SOA are analyzed and improved using the DDPG algorithm to produce the DDPGSOA algorithm. Finally, the path planning comparison experiments were conducted in MATLAB with a genetic algorithm, particle swarm optimization algorithm, gray wolf algorithm, and original seeker optimization algorithm. The experimental results show that DDPGSOA has more advantages in convergence performance and robot path solving ability.