Next-Best-View Planning

Follow this topic Ask about this topic

Next-best-view planning studies how a robot or sensing system can choose where to look next to improve a 3D reconstruction or answer a task-specific question. Common approaches score candidate views using information gain, Fisher information, coverage, or uncertainty, while learned policies and receding-horizon planners can account for future observations. Applications include 3D reconstruction, aerial and industrial inspection, grasping, localization, and semantic search, where occlusion, reachability, collision risk, and travel cost constrain useful views. A central tension is whether to optimize general geometric completeness or downstream task performance, while keeping view scoring efficient enough for real-time use.

Latest papers

61 papers9 papers in the last four weeks, down 25% on the four weeks before. 0.1% of all new papers.

Object-Reconstruction-Aware Whole-body Control of Mobile Manipulators
Sep 4, 2025Fatih Dursun, Bruno Vilhena Adorno, Simon Watson +1Next-Best-View Planning Mobile Robotics
Manchester Centre for Robotics and AI, University of Manchester, UK

Sep 4, 2025J/K move · Enter open · S save

Object-Reconstruction-Aware Whole-body Control of Mobile Manipulators

Fatih Dursun, Bruno Vilhena Adorno, Simon Watson, Wei Pan

Manchester Centre for Robotics and AI, University of Manchester, UK

Next-Best-View Planning Mobile Robotics Multi-View 3D Reconstruction Whole-Body Control

Open paper AskPDF arXiv

Object reconstruction and inspection tasks play a crucial role in various robotics applications. Identifying paths that reveal the most unknown areas of the object is paramount in this context, as it directly affects reconstruction efficiency. Current methods often use sampling based path planning techniques, evaluating views along the path to enhance reconstruction performance. However, these methods are computationally expensive as they require evaluating several candidate views on the path. To this end, we propose a computationally efficient solution that relies on calculating a focus point in the most informative region and having the robot maintain this point in the camera field of view along the path. In this way, object reconstruction related information is incorporated into the whole body control of a mobile manipulator employing a visibility constraint without the need for an additional path planner. We conducted comprehensive and realistic simulations using a large dataset of 114 diverse objects of varying sizes from 57 categories to compare our method with a sampling based planning strategy and a strategy that does not employ informative paths using Bayesian data analysis. Furthermore, to demonstrate the applicability and generality of the proposed approach, we conducted real world experiments with an 8 DoF omnidirectional mobile manipulator and a legged manipulator. Our results suggest that, compared to a sampling based strategy, there is no statistically significant difference in object reconstruction entropy, and there is a 52.3% probability that they are practically equivalent in terms of coverage. In contrast, our method is 6.2 to 19.36 times faster in terms of computation time and reduces the total time the robot spends between views by 13.76% to 27.9%, depending on the camera FoV and model resolution.