Embodied at Sea: Imitation Learning for Multi-Task Underwater Manipulation
Overview
This research investigates the use of imitation learning for multi-task underwater manipulation, focusing on overcoming domain shift challenges in marine environments. The study collects and analyzes data from underwater manipulation tasks, trains and evaluates 14 generalist VLA models, and introduces an ACT policy that outperforms a generalist VLA model in specific marine tasks. The BLUEMBODIED KIT, used for data collection, model training, and policy deployment, is open-sourced to support further research in marine field robotics.
From arXiv Robotics — research abstracts
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · gradable on-hardware rollouts: 1260 other · Basis not reported · Differing source assertions
“ta were collected in calm and turbulent waters and were used to train 14 generalist Vision-Language-Action (VLA) models (DynamicVLA) with varying proprioception observation configurations. Across three tasks on 1260 gradable on-hardware rollouts in a pool, 10 models outperformed the baseline, with the highest-performing model achieving a 93.3% overall success rate, including partial successes. Results are also introd”
Exact source · revision 1Source owner not reported
Reported quantity · gradable on-hardware rollouts: 508 other · Basis not reported · Differing source assertions
“uced from a case study evaluating the performance of a single-task Action Chunking Transformer (ACT) and a generalist VLA (SmolVLA) policy for two tasks in marine archaeology and conservation: grasping and mooring. Experiments are performed on an ROV on and around the WWII wreck Coralita X127 and at Manoel Island Marina, Malta. Across 748 collected episodes and 508 gradable on-hardware rollouts, the ACT policy substa”
Exact source · revision 1Source owner not reported
Reported quantity · collected episodes: 748 other · Basis not reported
“uced from a case study evaluating the performance of a single-task Action Chunking Transformer (ACT) and a generalist VLA (SmolVLA) policy for two tasks in marine archaeology and conservation: grasping and mooring. Experiments are performed on an ROV on and around the WWII wreck Coralita X127 and at Manoel Island Marina, Malta. Across 748 collected episodes and 508 gradable on-hardware rollouts, the ACT policy substa”
Exact source · revision 1Source owner not reported
Reported quantity · episodes of underwater manipulation data: 1616 other · Basis not reported
“arXiv:2610.38199v1 Announce Type: new Abstract: Behavior Cloning has demonstrated promising generalization capabilities for robotic manipulation in terrestrial environments. However, marine visual distortions and dynamic currents create a severe domain shift that remains a key bottleneck in adapting generalist architectures to underwater field manipulation. To address this, 1616 episodes of underwater manipulation da”
Exact source · revision 1Source owner not reported
Evidence dependency changes (3)
- 2026-10-30T18:29:48.367Z · evidence updated · source revision 1. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T18:29:33.963Z · evidence updated · source revision 1. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T17:12:00.495Z · evidence updated · source revision 1. Evidence extraction was updated; current source attributions are shown above.
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalEmbodied at Sea: Imitation Learning for Multi-Task Underwater Manipulation
This research investigates the use of imitation learning for multi-task underwater manipulation, focusing on overcoming domain shift challenges in marine environments. The study collects and analyzes data from underwater manipulation tasks, trains and evaluates 14 generalist VLA models, and introduces an ACT policy that outperforms a generalist VLA model in specific marine tasks. The BLUEMBODIED KIT, used for data collection, model training, and policy deployment, is open-sourced to support further research in marine field robotics.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).