UAV Intelligent obstacle avoidance algorithm based on greedy DDPG
In response to the difficulty of traditional UAV DDPG obstacle avoidance algorithm in addressing intelligent obstacle avoidance for UAVs flying in unknown conditions or complex environments,this paper pro-poses a greedy DDPG-based intelligent obstacle avoidance algorithm.Based on the traditional DDPG algorithm,the dynamic greedy degree adjustment method and Gaussian noise strategy are introduced to enable UAVs to bal-ance the relationship between exploration and utilization more efficiently in time of exploring the environment and developing obstacle avoidance strategies,so as to improve the learning efficiency and obstacle avoidance per-formance.Experimental results demonstrate that the greedy DDPG algorithm outperforms the traditional DDPG algorithm in terms of training efficiency and generalization ability,showing good robustness.
万方数据
维普
