Simultaneous Learning of Contact and Continuous Dynamics

Bibit Bianchini, Mathew Halm, Michael Posa

Keywords: system identification, dynamics learning, contact-rich manipulation

TL;DR: We simultaneously learn contact and continuous dynamics of novel objects through contact-rich trajectories, using model-based structure and residual physics.

Abstract: Robotic manipulation can greatly benefit from the data efficiency, robustness, and predictability of model-based methods if robots can quickly generate models of novel objects they encounter. This is especially difficult when effects like complex joint friction lack clear first-principles models and are usually ignored by physics simulators. Further, numerically-stiff contact dynamics can make common model-building approaches struggle. We propose a method to simultaneously learn contact and continuous dynamics of a novel, possibly multi-link object by observing its motion through contact-rich trajectories. We formulate a system identification process with a loss that infers unmeasured contact forces, penalizing their violation of physical constraints and laws of motion given current model parameters. Our loss is unlike prediction-based losses used in differentiable simulation. Using a new dataset of real articulated object trajectories and an existing cube toss dataset, our method outperforms differentiable simulation and end-to-end alternatives with more data efficiency. See our project page for code, datasets, and media: https://sites.google.com/view/continuous-contact-nets/home

Student First Author: yes

Instructions: I have read the instructions for authors (https://corl2023.org/instructions-for-authors/)

Video: https://youtu.be/uMCLCIzbgJo

Website: https://sites.google.com/view/continuous-contact-nets/home

Code: https://github.com/ebianchi/dair_pll