Abstract
Intermittent visual loss disrupts target-relative feedback during underwater orbiting, making it difficult to maintain coordinated motion and reacquire a moving target. We present AquaOrbit, a reinforcement-learning controller with a recovery module for underwater target orbiting under interrupted visual feedback. During detection loss, the recovery module uses latched line-of-sight, roll, and depth references to support stabilization and target reacquisition. We train the controller in Isaac Sim with dynamics, observation, and vision-loss randomization. Evaluated without retraining in Gazebo/ROS2 under a different physics engine and perception perturbations, AquaOrbit completes 20/20 orbiting trials in each of the static- and moving-target conditions on an unseen variable-depth 3-D trajectory. In the moving-target condition, it reduces mean line-of-sight error by approximately 46% relative to a PID-based visual servoing controller with recovery while maintaining comparable path-tracking accuracy; removing the recovery module reduces completion to 9/20. Zero-shot physical deployment with fully onboard perception and control demonstrates elliptical, figure-eight, and variable-depth circular trajectories, including the latter two trajectory types absent from training. The robot maintains attitude stability during manual occlusions lasting up to 8s and reacquires the target within 2.5s in the reported attitude-induced field-of-view loss events.
Keywords
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Yao, K., Leng, J., Zhang, H., Sun, Z., & Li, X. (2026). AquaOrbit: Sim-to-Real Reinforcement Learning for Underwater Target Orbiting under Intermittent Visual Feedback. https://omanscience.com/en/articles/aquaorbit-sim-to-real-reinforcement-learning-for-underwater-target-orbiting-under-intermittent-visual-feedback
MLA 9
Yao, Kanzhong, et al. "AquaOrbit: Sim-to-Real Reinforcement Learning for Underwater Target Orbiting under Intermittent Visual Feedback." https://omanscience.com/en/articles/aquaorbit-sim-to-real-reinforcement-learning-for-underwater-target-orbiting-under-intermittent-visual-feedback.
Chicago (author–date)
Yao, Kanzhong, Jinyi Leng, Hao Zhang, Zhe Sun, and Xuelong Li. 2026. "AquaOrbit: Sim-to-Real Reinforcement Learning for Underwater Target Orbiting under Intermittent Visual Feedback." https://omanscience.com/en/articles/aquaorbit-sim-to-real-reinforcement-learning-for-underwater-target-orbiting-under-intermittent-visual-feedback.
Harvard
Yao, K., Leng, J., Zhang, H., Sun, Z. and Li, X. (2026) 'AquaOrbit: Sim-to-Real Reinforcement Learning for Underwater Target Orbiting under Intermittent Visual Feedback', Available at: https://omanscience.com/en/articles/aquaorbit-sim-to-real-reinforcement-learning-for-underwater-target-orbiting-under-intermittent-visual-feedback.
Vancouver
Yao K, Leng J, Zhang H, Sun Z, Li X. AquaOrbit: Sim-to-Real Reinforcement Learning for Underwater Target Orbiting under Intermittent Visual Feedback. https://omanscience.com/en/articles/aquaorbit-sim-to-real-reinforcement-learning-for-underwater-target-orbiting-under-intermittent-visual-feedback
IEEE
K. Yao, J. Leng, H. Zhang, Z. Sun, and X. Li, "AquaOrbit: Sim-to-Real Reinforcement Learning for Underwater Target Orbiting under Intermittent Visual Feedback," https://omanscience.com/en/articles/aquaorbit-sim-to-real-reinforcement-learning-for-underwater-target-orbiting-under-intermittent-visual-feedback.