2018/05/02 by Gilad Francis, Francis, Gilad, Lionel Ott +3
Computer Science · Engineering · #Advanced Image and Video Retrieval Techniques #FOS: Computer and information sciences #Robotic Path Planning Algorithms #Robotics (cs.RO) #Robotics and Sensor-Based Localization
paper · pdf · doi:10.48550/arxiv.1805.01079
openalex publication_date 2018/05/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Autonomous exploration is a complex task where the robot moves through an\nunknown environment with the goal of mapping it. The desired output of such a\nprocess is a sequence of paths that efficiently and safely minimise the\nuncertainty of the resulting map. However, optimising over the entire space of\npossible paths is computationally intractable. Therefore, most exploration\nmethods relax the general problem by optimising a simpler one, for example\nfinding the single next best view. In this work, we formulate exploration as a\nvariational problem which allows us to directly optimise in the space of\ntrajectories using functional gradient methods, searching for the Next Best\nPath (NBP). We take advantage of the recently introduced Hilbert maps to devise\nan information-based functional that can be computed in closed-form. The\nresulting trajectories are continuous and maximise safety as well as mutual\ninformation. In experiments we verify the ability of the proposed method to\nfind smooth and safe paths and compare these results with other exploration\nmethods.\n