Cross-Entropy Optimization of Physically Grounded Task and Motion Plans

TL;DR

This paper introduces a GPU-accelerated physics-based planning method that explicitly considers dynamics and contacts, optimizing control parameters via cross-entropy to generate feasible, executable plans for robotic tasks involving object manipulation.

Contribution

It presents a novel approach combining physics simulation and cross-entropy optimization to produce realistic, low-cost plans that can be directly executed by robots, improving over prior abstraction-based methods.

Findings

Successfully plans with physics-based realism

Plans are directly executable on real robots

Demonstrates environment-aware object manipulation

Abstract

Autonomously performing tasks often requires robots to plan high-level discrete actions and continuous low-level motions to realize them. Previous TAMP algorithms have focused mainly on computational performance, completeness, or optimality by making the problem tractable through simplifications and abstractions. However, this comes at the cost of the resulting plans potentially failing to account for the dynamics or complex contacts necessary to reliably perform the task when object manipulation is required. Additionally, approaches that ignore effects of the low-level controllers may not obtain optimal or feasible plan realizations for the real system. We investigate the use of a GPU-parallelized physics simulator to compute realizations of plans with motion controllers, explicitly accounting for dynamics, and considering contacts with the environment. Using cross-entropy optimization, we sample the parameters of the controllers, or actions, to obtain low-cost solutions. Since our approach uses the same controllers as the real system, the robot can directly execute the computed plans. We demonstrate our approach for a set of tasks where the robot is able to exploit the environment's geometry to move an object. Website and code: https://andreumatoses.github.io/research/parallel-realization