# Tri-Manual Visuomotor Imitation Learning of Robot Policies

arXiv: https://arxiv.org/abs/2607.25731 · HTML: https://mlsanity.com/paper/2607.25731/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 20 of its 34 references that are in the map.
- First posted on arXiv 28 Jul 2026.

## Placed in

- Offline RL and imitation learning from data / [Robot manipulation imitation learning](https://mlsanity.com/offlinerl/robot-manipulation-imitation-learning/)

## Builds on (most cited first, 20 of 20)

- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset](https://mlsanity.com/paper/2403.12945/) (2024)
- [Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots](https://mlsanity.com/paper/2402.10329/) (2024)
- [MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations](https://mlsanity.com/paper/2310.17596/) (2023)
- [ALOHA Unleashed: A Simple Recipe for Robot Dexterity](https://mlsanity.com/paper/2410.13126/) (2024)
- [TACO: Learning Task Decomposition via Temporal Alignment for Control](https://mlsanity.com/paper/1803.01840/) (2018)
- [Learning Multi-Arm Manipulation Through Collaborative Teleoperation](https://mlsanity.com/paper/2012.06738/) (2020)
- [Instructing Robots by Sketching: Learning from Demonstration via Probabilistic Diagrammatic Teaching](https://mlsanity.com/paper/2309.03835/) (2023)
- [DexImit: Learning Bimanual Dexterous Manipulation from Monocular Human Videos](https://mlsanity.com/paper/2602.10105/) (2026)
- [Cooperative Task and Motion Planning for Multi-Arm Assembly Systems](https://mlsanity.com/paper/2203.02475/) (2022)
- [Proleptic Temporal Ensemble for Improving the Speed of Robot Tasks Generated by Imitation Learning](https://mlsanity.com/paper/2410.16981/) (2024)
- [A Constraint Programming Approach to Simultaneous Task Allocation and Motion Scheduling for Industrial Dual-Arm Manipulation Tasks](https://mlsanity.com/paper/1901.07914/) (2019)
- [Learning Symbolic and Subsymbolic Temporal Task Constraints from Bimanual Human Demonstrations](https://mlsanity.com/paper/2403.16953/) (2024)
- [Joint Flow Trajectory Optimization For Feasible Robot Motion Generation from Video Demonstrations](https://mlsanity.com/paper/2509.20703/) (2025)
- [Cross-Modal Instructions for Robot Motion Generation](https://mlsanity.com/paper/2509.21107/) (2025)
- [Unified Learning of Temporal Task Structure and Action Timing for Bimanual Robot Manipulation](https://mlsanity.com/paper/2603.06538/) (2026)
- [Robots that Collaborate: Sequential Asymmetric Imitation for Learning Coupled Robot Policies](https://mlsanity.com/paper/2606.16490/) (2026)
- [HATS: A Human-Agent Teleoperation System for Multi-Arm Data Collection](https://mlsanity.com/paper/2606.16491/) (2026)

## Abstract (arXiv)

Bimanual teleoperation provides an effective way to collect robot demonstrations, but it assumes that the operator and robot have matching numbers of simultaneous control channels. This assumption breaks for tri-manual systems: the robot can coordinate three arms concurrently, whereas a single operator can continuously control only two. Pairwise mode switching may therefore record otherwise independent motions sequentially, causing behaviour cloning to reproduce delays imposed by the interface rather than required by the task. We present TriManPolicy, a tri-manual imitation learning system that allows one operator to demonstrate behaviours for three arms. Its central component is Dependency-Aware Tri-Arm Scheduling (DATS). The key idea is to preserve the demonstrated arm motions while reconsidering when they occur. DATS retimes demonstrations offline by preserving local sensorimotor segments of fixed duration and repositioning them according to constraints on task order and arm usage that are reviewed by a human. The resulting data train a single synchronous policy for all three arms, while deployment requires neither the dependency graph nor the scheduler. Across six challenging tasks performed in the real world, policies trained on demonstrations retimed by DATS exhibit more efficient coordination while maintaining comparable observed task success. Offline analysis further shows that DATS changes the supervision across arms rather than merely removing idle periods. Project videos and additional material are available at https://aus.bot/trimanpolicy/.