A two-stage intelligent rendezvous algorithm based on KM and PPO is proposed for the problem of multi-UAV rendezvous trajectory planning. Firstly, based on the Dubins curve and the iterative KM algorithm, the cost distance matrix and the UAV target assignment result are constructed. Secondly, based on the flight cost constraint, a set of comprehensive rewards is designed for the algorithm, and a training method of shared parameters is introduced. Finally, the simulation results reveal that the algorithm improves the convergence speed of training by 75% and the convergence effect by 5.1%. Compared with other methods, it is more efficient in completing the trajectory planning task.
Multi-UAV Rendezvous Trajectory Planning Based on PPO Algorithm
Lect. Notes Electrical Eng.
International Conference on Guidance, Navigation and Control ; 2024 ; Changsha, China August 09, 2024 - August 11, 2024
02.03.2025
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Rendezvous Guidance Trajectory Shaping
IEEE | 1987
|Trajectory shaping rendezvous guidance
NTRS | 1986
|Trajectory shaping rendezvous guidance
Tema Archiv | 1986
|