FATHOM: World-Model-Guided Diffusion Action-Trajectory Selection for Sparse-Marker AUV Tracking
FATHOM: World-Model-Guided Diffusion Action-Trajectory Selection for Sparse-Marker AUV Tracking
This research introduces FATHOM, a framework designed for reliable relative tracking of autonomous underwater vehicles (AUVs) using sparse active light markers. The system combines structured marker observations, inertial data, and action history within a multimodal recurrent state-space model to form a predictive tracking belief. A diffusion action prior generates multiple candidate trajectories, which are then evaluated against predicted future consequences via learned latent dynamics. FATHOM selects the optimal trajectory using a receding-horizon objective and updates decisions based on onboard observations. Extensive offline analyses and real-AUV experiments demonstrate robust autonomous tracking under nominal and challenging conditions, including partial marker visibility and dynamic motion, all while maintaining real-time operation without external motion-capture feedback.
The paper presents FATHOM, a world-model-guided diffusion framework for sparse-marker AUV tracking that integrates multimodal observations with latent dynamics prediction to select optimal thruster trajectories under partial visibility and nonlinear vehicle dynamics.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Control
- Reported autonomousOpen source S5
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ive offline analyses verify the learned predictive belief, multi-step world-model prediction, diffusion trajectory generation, and predictive selection. More importantly, comprehensive real-AUV experiments, including strong pose-based comparisons and component ablations, demonstrate robust autonomous tracking across nominal and challenging motions with naturally occurring partial marker visibility, while maintaining
Open source S5
arXiv:2611.08898v1 Announce Type: new Abstract: Reliable relative tracking is fundamental to cooperative autonomous underwater vehicles (AUVs), yet remains challenging under nonlinear vehicle dynamics and degraded underwater visual conditions. Active light markers improve robustness to appearance variation, but provide deliberately sparse observations that can become incomplete or corrupted by occlusion, reflections,
Open source S2
recurrent state-space model. A diffusion action prior then generates multiple expert-like multi-thruster trajectories, whose future tracking consequences are predicted through the learned latent dynamics. Rather than directly executing a generated trajectory, FATHOM selects the most favorable candidate under an explicit receding-horizon objective and continually updates the decision from onboard observations. Extens
Open source S4
real-time onboard operation without motion-capture feedback.
Open source S6
FATHOM: World-Model-Guided Diffusion Action-Trajectory Selection for Sparse-Marker AUV Tracking
Open source S1
Source：arXiv Robotics — research abstracts · arxiv.org