Differentiable Contact Dynamics for Stable Object Placement Under Geometric Uncertainties

TL;DR

This paper introduces a differentiable contact dynamics model that enables robots to achieve stable object placement despite geometric uncertainties by estimating and compensating for pose and shape inaccuracies through gradient-based methods.

Contribution

It presents a novel differentiable simulation approach that relates force-torque data to geometric uncertainties, improving stability in robotic object placement under uncertainty.

Findings

Effective uncertainty estimation via gradient descent

Successful application on a Franka robot with various objects

Robust handling of pose, shape, and environment uncertainties

Abstract

From serving a cup of coffee to positioning mechanical parts during assembly, stable object placement is a crucial skill for future robots. It becomes particularly challenging under geometric uncertainties, e.g., when the object pose or shape is not known accurately. This work leverages a differentiable simulation model of contact dynamics to tackle this challenge. We derive a novel gradient that relates force-torque sensor readings to geometric uncertainties, thus enabling uncertainty estimation by minimizing discrepancies between sensor data and model predictions via gradient descent. Gradient-based methods are sensitive to initialization. To mitigate this effect, we maintain a belief over multiple estimates and choose the robot action based on the current belief at each timestep. In experiments on a Franka robot arm, our method achieved promising results on multiple objects under various geometric uncertainties, including the in-hand pose uncertainty of a grasped object, the object shape uncertainty, and the environment uncertainty.