Shape aware neural fields improve robotic motion and manipulation planning
Shape-Aware Neural Time Fields for Rigid-Body Motion and Manipulation Planning
RoboticsMachine Learning
Summary
Planning how a robot moves objects is tricky because objects can move by shifting position and rotating, and these moves are differently affected by obstacles. The authors created a new way for robots to learn a map of how hard it is to move an object from one place to another, considering both position and rotation. This map helps the robot plan better and faster paths through cluttered spaces. Their method works well in both 2D and 3D tasks and can adapt quickly when the environment changes or if the robot’s moves are disturbed.
What this means in practice
- •For industrial automation teams: Plan efficient and adaptive robot arm paths for moving objects in cluttered factory floors using learned travel-time maps.
- •For warehouse robotics developers: Enable robots to push and reposition objects quickly while handling unexpected disturbances by fast motion replanning.
Authors
Yanming Zhou, Yuchen Liu, Ahmed H. Qureshi
Abstract
Rigid-body motion planning is fundamental to robotic transport and manipulation, where both object position and orientation determine whether a body can move through cluttered environments. This coupling makes planning chal- lenging because translation and rotation can be constrained very differently by surrounding geometry. We propose Shape- Aware Neural Time Fields, a physics-informed neural motion planning framework that learns a reusable travel-time, or cost- to-go, representation for rigid-body motion while explicitly accounting for these distinct geometric restrictions. The learned field captures local motion difficulty in translation and rotation and is paired with a sampling-based planner that balances progress toward the goal with the cost of reaching nearby poses. We evaluate the framework across 2D and 3D rigid- body planning tasks and non-prehensile manipulation through object pushing in diverse environments. Results show substantial improvements in planning success and inference efficiency over physics-informed neural planning baselines, while remaining competitive with classical sampling-based planners at significantly lower planning times. Ablation studies further validate the contributions of both the learned representation and the proposed planning strategy. Finally, closed-loop pushing experiments under execution deviations and external disturbances demonstrate that fast replanning enables reactive and adaptive task execution.