aiqbench analysis
AquaBEV-Nav addresses a critical limitation in underwater robotics by replacing unreliable monocular depth estimation with direct bird's-eye-view occupancy prediction, enhancing safety in scattering-rich environments. This approach integrates learned polar representations into hierarchical exploration frameworks to enable robust collision avoidance and trajectory planning without intermediate geometric unprojection. The method demonstrates significant performance gains in simulated reef scenarios, offering a viable path forward for autonomous underwater vehicles operating under degraded visual conditions.
AquaBEV-Nav: Learned BEV Occupancy for Underwater Navigation and Exploration
arXiv:2609.32156v1 Announce Type: new Abstract: Safe underwater exploration requires a robot to understand where surrounding structures are located and which regions are available for motion. Existing vision-based underwater exploration systems commonly obtain this information indirectly by estimating monocular depth, unprojecting the geometry into 3D space, and accumulating it into a 2D bird's-eye-view occupancy map. This reliance on intermediate depth estimation is particularly problematic underwater, where scattering and wavelength-dependent attenuation degrade visual cues and limit the reliability of monocular depth estimates. We introduce AquaBEV-Nav, an underwater exploration framework that bypasses explicit monocular depth estimation through direct bird's-eye-view occupancy prediction. Built upon the CORAL hierarchical exploration framework, AquaBEV-Nav replaces its depth-based perception front end with AquaBEV. Given a single RGB frame, AquaBEV maps visual features into a learned polar representation, performs causal reasoning along the range dimension, and reconstructs local Cartesian occupancy without relying on intermediate depth prediction. The resulting occupancy map is accumulated into CORAL's persistent spatial memory, providing spatial context for VLM-based high-level planning and collision constraints for dynamics-aware local trajectory generation. Across ten simulated reef environments and six occupancy backbones evaluated under a single protocol, AquaBEV-Nav reaches 37.48 structure IoU and 53.2 target IoU, 88.95% closed-loop coverage with zero collisions.
Abstract:Safe underwater exploration requires a robot to understand where surrounding structures are located and which regions are available for motion. Existing vision-based underwater exploration systems commonly obtain this information indirectly by estimating monocular depth, unprojecting the geometry into 3D space, and accumulating it into a 2D bird's-eye-view occupancy map. This reliance on intermediate depth estimation is particularly problematic underwater, where scattering and wavelength-dependent attenuation degrade visual cues and limit the reliability of monocular depth estimates. We introduce AquaBEV-Nav, an underwater exploration framework that bypasses explicit monocular depth estimation through direct bird's-eye-view occupancy prediction. Built upon the CORAL hierarchical exploration framework, AquaBEV-Nav replaces its depth-based perception front end with AquaBEV. Given a single RGB frame, AquaBEV maps visual features into a learned polar representation, performs causal reasoning along the range dimension, and reconstructs local Cartesian occupancy without relying on intermediate depth prediction. The resulting occupancy map is accumulated into CORAL's persistent spatial memory, providing spatial context for VLM-based high-level planning and collision constraints for dynamics-aware local trajectory generation. Across ten simulated reef environments and six occupancy backbones evaluated under a single protocol, AquaBEV-Nav reaches 37.48 structure IoU and 53.2 target IoU, 88.95% closed-loop coverage with zero collisions.
Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2609.32156 [cs.RO]
(or arXiv:2609.32156v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.32156 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 295 words
Related coverage
AquaBEV: Monocular Underwater BEV Occupancy with 3D Sonar Supervision
arXiv · Sep 7, 2026
Unified Planning-Learning Framework for Robust UUV Navigation Under Partial Observability
arXiv · Aug 7, 2026
Towards End to End Motion Planning and Execution for Autonomous Underwater Vehicles Using Reinforcement Learning
arXiv · Jun 9, 2026
ODG-NoMaD: Overhead-Camera Direction-Guided NoMaD
arXiv · Aug 25, 2026