The Surface Edge Explorer (SEE++) is a Next Best View (NBV) planning approach that uses a novel unstructured density representation to efficiently obtain complete scene observations. SEE++ uses point-based computations to identify boundaries in a scene observation, propose new views to extend the observation and select a next best view while proactively considering occlusions and the attainable improvement in scene coverage. Experimental results show that SEE++ outperforms NBV planning approaches with volumetric representations in terms of the computational cost, travel distance and number of views required to obtain complete scene observations.
- The International Journal of Robotics Research (IJRR)
- Submitted, Manuscript #IJR-22-4541
- Proactive estimation of occlusions and scene coverage for planning next best views in an unstructured representation
- Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
- Proceedings of the IEEE International Conference on Robotics and Automation (ICRA)
- Brisbane, Australia