.png){kind=icon}
.png){kind=icon}
Return
ADAPT-planner: Asynchronous distributed adaptive priority-based trajectory planner for heterogeneous AAV swarms
.png){kind=icon}
OA
AI
N
N
N
N
N
DOI：10.1016/j.dt.2026.07.027.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The growing deployment of heterogeneous autonomous aerial vehicles (AAVs) in urban environments increases inter-agent collision risk, posing significant challenges to safe, scalable swarm coordination. To address these challenges, an asynchronous, distributed, adaptive priority-based trajectory planner (ADAPT-Planner) is designed for efficient heterogeneous AAV swarm trajectory generation. A hierarchical asynchronous framework is first established to decouple communication, planning, and execution processes to enable independent parallel planning for individuals without synchronization barriers or data blocking. Second, leveraging differential flatness, we formulate the trajectory planning as a unified, lightweight, and unconstrained optimization problem, thereby guaranteeing efficiency and extensibility for heterogeneous dynamics. Additionally, considering heterogeneous maneuverability and distributed replanning burden, an adaptive priority-based planning strategy is introduced to adjust agent priorities online and reactivate idle agents to rejoin coordinated planning, thereby achieving efficient swarm deconfliction and rapid trajectory convergence. Extensive simulations and real-world tests validate that ADAPT-Planner resolves dense trajectory conflicts for a 32-agent heterogeneous swarm within only 1.73 s, reducing planning runtime by 53.8% compared with the standard distributed asynchronous baseline, and by 26.1% relative to the static sequential-priority strategy, while maintaining a 100% success rate across all tested scales.
Keywords:
Autonomous aerial vehicles (AAVs)
Heterogeneous dynamics
Trajectory planning
Asynchronous distributed optimization
Adaptive priority-based planning
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
5.9
Papers:
1.9K
Citations:
6.4K
Organization
Cited Papers