MA-SafeDiffuser: Safe Multi-Agent Planning with Diffusion Probabilistic Models

A new arXiv paper (2611.05524v1) introduces MA-SafeDiffuser, a multi-agent extension of SafeDiffuser that gives diffusion-based trajectory planners finite-time diffusion invariance guarantees for joint-agent safety specifications. The work formalizes joint safe sets as intersections of per-agent and pairwise barrier sets, derives centralized and decentralized communication-aware constrained denoising procedures with provable invariance under mild assumptions, and addresses local-trap and deadlock phenomena via time-varying specifications and liveness control barrier functions. The authors report that MA-SafeDiffuser reduces violation counts relative to unconstrained diffusion baselines while retaining planning quality, and they release algorithms, proofs, a reference implementation, and a benchmarking suite including a multi-agent Maze2D domain.

arXiv:2611.05524v1 Announce Type: new Abstract: We propose MA-SafeDiffuser, a multi-agent extension of SafeDiffuser that equips diffusion-based trajectory planners with finite-time diffusion invariance guarantees for joint-agent safety specifications. Building on the single-agent construction that embeds control barrier function CBF constraints into reverse-diffusion updates, we: i formalize joint safe sets as intersections of per-agent and pairwise barrier sets; ii derive centralized and decentralized communication-aware constrained denoising procedures with provable invariance under mild assumptions; iii address local-trap and deadlock phenomena via time-varying specifications and liveness CBFs; and iv develop a lightweight benchmarking suite including a multi-agent Maze2D domain. Empirically, MA-SafeDiffuser reduces violation counts relative to unconstrained diffusion baselines, while retaining planning quality. We provide algorithms, proofs, and reference implementations.