Identifying Mechanical Models through Differentiable Simulations

TL;DR

This paper introduces a differentiable physics-based method to identify unknown mechanical properties of objects through manipulation, enabling real-time property estimation with a robotic system.

Contribution

It extends differentiable physics engines to 3D forces and develops a gradient-based approach for real-time mechanical property identification during object manipulation.

Findings

Successfully identified mechanical properties of objects in real-time

Extended differentiable physics engine to 3D force modeling

Demonstrated effectiveness with real robot experiments

Abstract

This paper proposes a new method for manipulating unknown objects through a sequence of non-prehensile actions that displace an object from its initial configuration to a given goal configuration on a flat surface. The proposed method leverages recent progress in differentiable physics models to identify unknown mechanical properties of manipulated objects, such as inertia matrix, friction coefficients and external forces acting on the object. To this end, a recently proposed differentiable physics engine for two-dimensional objects is adopted in this work and extended to deal forces in the three-dimensional space. The proposed model identification technique analytically computes the gradient of the distance between forecasted poses of objects and their actual observed poses and utilizes that gradient to search for values of the mechanical properties that reduce the reality gap. Experiments with real objects using a real robot to gather data show that the proposed approach can identify the mechanical properties of heterogeneous objects on the fly.