Multi-UAV Formation Obstacles Avoidance Path Planning Based on PPO Algorithm

The path planning of multi-UAV formations plays a crucial role in various scenarios, including rescue operations, reconnaissance missions, and target strikes. This paper investigates a method for multi-UAV formation obstacle avoidance route planning based on deep reinforcement learning. The proposed...

Full description

Saved in:
Bibliographic Details
Published in2023 9th International Conference on Big Data and Information Analytics (BigDIA) pp. 55 - 62
Main Authors Wan, Yu, Zhao, Zipeng, Tang, Jun, Chen, Xi, Qi, Jingtao
Format Conference Proceeding
LanguageEnglish
Published IEEE 15.12.2023
Subjects
Collision avoidance
deep reinforcement learning
Drones
Heuristic algorithms
multi-UAV
Optimization
Path planning
PPO
Surveillance
Task analysis
Online AccessGet full text

Cover

More Information
Summary:The path planning of multi-UAV formations plays a crucial role in various scenarios, including rescue operations, reconnaissance missions, and target strikes. This paper investigates a method for multi-UAV formation obstacle avoidance route planning based on deep reinforcement learning. The proposed method allows UAVs to efficiently plan routes in complex and dynamic three-dimensional obstacle environments. The autonomous decision-making of multiple UAVs is realized by using the proximal policy optimization (PPO) algorithm. In addition, in order to achieve formation obstacle avoidance, we design a reward function to assess the advantages and disadvantages of UAV strategies. Finally, an environment with dynamic obstacles is constructed and trained using the method. Experimental results show that the method is feasible and generalizable.
ISSN:2771-6902
DOI:10.1109/BigDIA60676.2023.10429413