Object-Centric Task and Motion Planning in Dynamic Environments

TL;DR

This paper introduces a novel task and motion planning algorithm that optimizes over object-relative Cartesian frames, enabling robots to adapt in real-time to dynamic environments with moving objects, perception errors, and control inaccuracies.

Contribution

The proposed TAMP algorithm uniquely maintains plan validity in dynamic environments by optimizing over object-relative frames, allowing reactive execution and real-time adaptation.

Findings

Successfully applied to a torque-controlled robot in simulation and real-world scenarios.

Demonstrated robustness to environment changes, perception inaccuracies, and control imprecisions.

Achieved real-time adaptation in dynamic pick-and-place tasks.

Abstract

We address the problem of applying Task and Motion Planning (TAMP) in real world environments. TAMP combines symbolic and geometric reasoning to produce sequential manipulation plans, typically specified as joint-space trajectories, which are valid only as long as the environment is static and perception and control are highly accurate. In case of any changes in the environment, slow re-planning is required. We propose a TAMP algorithm that optimizes over Cartesian frames defined relative to target objects. The resulting plan then remains valid even if the objects are moving and can be executed by reactive controllers that adapt to these changes in real time. We apply our TAMP framework to a torque-controlled robot in a pick and place setting and demonstrate its ability to adapt to changing environments, inaccurate perception, and imprecise control, both in simulation and the real world.