Pushing Through Clutter With Movability Awareness of Blocking Obstacles

-- Navigation Among Movable Obstacles (NAMO) poses a challenge for traditional path-planning methods when obstacles block the path, requiring push actions to reach the goal. We propose a framework that enables movability-aware planning to overcome this challenge without relying on explicit obstacle placement. A physics engine is adopted to simulate the interaction result of the rollouts with the environment, and generate trajectories that minimize contact force. In qualitative and quantitative experiments, SVG-MPPI outperforms the existing paradigm that uses only binary movability for planning, achieving higher success rates with reduced cumulative contact forces. Our code is available at: https://github.com/tud-amr/SVG-MPPI I. INTRODUCTION A fundamental ability of autonomous robots is to navigate towards a goal while avoiding collisions along the way [1]. However, in complex and cluttered environments, such as domestic settings where obstacles like chairs and boxes may obstruct the path to the goal, finding collision-free paths often becomes impractical. In such cases, traditional navigation methods often fail and Navigation Amongst Movable Obstacles (NAMO) becomes essential.