Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.14159 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913125080498176 |
|---|---|
| author | Dong, Dayi Bhatt, Maulik Choi, Seoyeon Mehr, Negar |
| author_facet | Dong, Dayi Bhatt, Maulik Choi, Seoyeon Mehr, Negar |
| contents | As robots become more integrated in society, their ability to coordinate with other robots and humans on multi-modal tasks (those with multiple valid solutions) is crucial. Such behaviors can be learned from expert demonstrations via imitation learning (IL), but when expert demonstrations are multi-modal, standard IL approaches usually average across modes or collapse to a single mode, preventing effective coordination. Being inspired by diffusion models' ability to capture complex multi-modal trajectory distributions in single-agent settings, we develop a diffusion-based framework for coordinated multi-modal behavior in multi-agent systems. However, existing multi-agent diffusion approaches typically require a centralized planner or explicit communication among agents. This assumption can fail in real-world scenarios where robots must operate independently or with agents like humans that they cannot directly communicate with. Therefore, we propose MIMIC-D, a joint training with decentralized execution paradigm for multi-modal multi-agent IL via diffusion. We jointly train all agents' policies with only local information to achieve implicit coordination. In simulation and hardware experiments, our method exhibits robust multi-modal coordination behavior in various tasks and environments, improving upon state-of-the-art baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14159 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MIMIC-D: Multi-modal Imitation for MultI-agent Coordination with Decentralized Diffusion Policies Dong, Dayi Bhatt, Maulik Choi, Seoyeon Mehr, Negar Robotics As robots become more integrated in society, their ability to coordinate with other robots and humans on multi-modal tasks (those with multiple valid solutions) is crucial. Such behaviors can be learned from expert demonstrations via imitation learning (IL), but when expert demonstrations are multi-modal, standard IL approaches usually average across modes or collapse to a single mode, preventing effective coordination. Being inspired by diffusion models' ability to capture complex multi-modal trajectory distributions in single-agent settings, we develop a diffusion-based framework for coordinated multi-modal behavior in multi-agent systems. However, existing multi-agent diffusion approaches typically require a centralized planner or explicit communication among agents. This assumption can fail in real-world scenarios where robots must operate independently or with agents like humans that they cannot directly communicate with. Therefore, we propose MIMIC-D, a joint training with decentralized execution paradigm for multi-modal multi-agent IL via diffusion. We jointly train all agents' policies with only local information to achieve implicit coordination. In simulation and hardware experiments, our method exhibits robust multi-modal coordination behavior in various tasks and environments, improving upon state-of-the-art baselines. |
| title | MIMIC-D: Multi-modal Imitation for MultI-agent Coordination with Decentralized Diffusion Policies |
| topic | Robotics |
| url | https://arxiv.org/abs/2509.14159 |