Mobile robot senses surroundings to navigate safely in changing spaces
Object- and Task-Aware Active Perception for Safe Navigation of Mobile Manipulators in Semi-Static Environments
Robotics
Summary
Mobile robots with arms often need to move safely in cluttered places where things might have changed since their last visit. The authors propose a method where the robot uses its hand camera to actively look around uncertain or potentially changed areas while moving. Their approach plans where to look next by focusing on important objects near its path and adjusts its viewpoint without losing track of its main route. Tests show this method helps the robot maintain better maps and avoid collisions compared to more passive approaches.
What this means in practice
- •For logistics teams: Improve safe navigation of warehouse robots by actively inspecting uncertain obstacles along their routes.
- •For manufacturing operators: Enable mobile robots to maintain updated maps and avoid collisions in semi-static factory environments with changing layouts.
Authors
Jingxing Qian, Xintong Du, Jun Yang, Siqi Zhou, Steven L. Waslander, Angela P. Schoellig
Abstract
Mobile manipulators must navigate safely in cluttered, partially unknown environments across logistics, manufacturing, and service applications. Semi-static changes between robot visits can lead to outdated scene knowledge and increased collision risk. We propose an active perception framework that enables a high degree-of-freedom (DoF) mobile manipulator to inspect uncertain or safety-critical regions with its hand camera during navigation. A probabilistic, object- and task-aware next-best-view (NBV) formulation prioritizes observations near the desired path and around uncertain or potentially changing objects. The perception objective is integrated into hierarchical-task model predictive control (HTMPC), exploiting kinematic redundancy for viewpoint adjustment while preserving the higher priority of base-trajectory tracking. The system unifies navigation, map maintenance, and active perception in a closed loop. Real-world tests show improved map consistency, reduced task-dependent map entropy, and lower navigation risk than passive or non-task-aware baselines. Additional simulation and real-world sequences demonstrate responses to controlled semi-static scene changes.