Multi-robot trajectory planning improved with diffusion denoising method
Denoising Multi-Robot Trajectories
RoboticsMultiagent Systems
Summary
Planning paths for many robots at once is very hard because their movements and goals can be complex and conflict with each other. The authors improve a method called D4orm that uses a special sampling approach to quickly and reliably find safe and efficient routes for multiple robots. Their method works well on different types of robots and in various environments, including real-world tests with flying drones and ground robots. This new approach also supports online adjustments and distributed coordination, making it flexible for many uses.
What this means in practice
- •For robotics engineers: Generate safe, feasible multi-robot paths quickly using diffusion denoising on GPUs for complex coordination tasks.
- •For drone fleet operators: Deploy real-time trajectory planning that adapts online for obstacle-rich environments using onboard distributed computation.$Commercial implications: Enables commercial drone fleets to coordinate autonomously and safely with efficient onboard planning, improving operational reliability.
Authors
Yuhao Zhang, Keisuke Okumura, Ajay Shankar, Amanda Prorok
Abstract
Multi-robot trajectory planning is a fundamental problem in multi-robot coordination but remains computationally challenging due to its nonconvex, multimodal, and high-dimensional nature. This work builds upon D4orm, a dynamics-aware diffusion-denoising framework, and develops a family of planning architectures for diverse operational requirements. Unlike conventional numerical optimization methods, D4orm employs sampling-based optimization to generate solution trajectories through massively parallel sampling, leveraging modern computing architectures such as GPUs. Its diffusion-denoising structure iteratively optimizes \textit{deformations} to candidate control trajectories, providing an efficient and versatile paradigm for generating kinodynamically feasible and conflict-free trajectories. Using D4orm as the building block for advanced planners, we present a decoupled planner for improved scalability, an online receding-horizon planner with feedback control, and a distributed planner for resource-constrained settings. Evaluations with differential-drive and holonomic robots in 2D and 3D environments demonstrate that D4orm-based approaches find high-quality solutions faster and more reliably than other sampling-based optimization methods, such as MPPI, as well as a learned diffusion-model-based method. We further demonstrate zero-shot deployment on ten real quadrotors with obstacles, large-scale deconfliction with 100 simulated robots, and fully onboard distributed `lifelong' operation with six ground robots. Overall, these results establish diffusion denoising as a scalable and reliable framework for multi-robot coordination. Code and video: https://github.com/proroklab/d4orm