Validating robotics simulators on real-world impacts

B Acosta, W Yang, M Posa - IEEE Robotics and Automation …, 2022 - ieeexplore.ieee.org
A realistic simulation environment is an essential tool in every roboticist's toolkit, with uses
ranging from planning and control to training policies with reinforcement learning. Despite …

Distributionally adaptive meta reinforcement learning

A Ajay, A Gupta, D Ghosh, S Levine… - Advances in Neural …, 2022 - proceedings.neurips.cc
Meta-reinforcement learning algorithms provide a data-driven way to acquire policies that
quickly adapt to many tasks with varying rewards or dynamics functions. However, learned …

Fundamental challenges in deep learning for stiff contact dynamics

M Parmar, M Halm, M Posa - 2021 IEEE/RSJ International …, 2021 - ieeexplore.ieee.org
Frictional contact has been extensively studied as the core underlying behavior of legged
locomotion and manipulation, and its nearly-discontinuous nature makes planning and …

Data-augmented contact model for rigid body simulation

Y Jiang, J Sun, CK Liu - Learning for dynamics and control …, 2022 - proceedings.mlr.press
Accurately modeling contact behaviors for real-world, near-rigid materials remains a grand
challenge for existing rigid-body physics simulators. This paper introduces a data …

Physics-penalised regularisation for learning dynamics models with contact

G Pizzuto, M Mistry - Learning for Dynamics and Control, 2021 - proceedings.mlr.press
Robotic systems, such as legged robots and manipulators, often handle states which involve
ground impact or interaction with objects present in their surroundings; both of which are …

Addressing Stiffness-Induced Challenges in Modeling and Identification for Rigid-Body Systems With Friction and Impacts

M Halm - 2023 - search.proquest.com
Imperfect, useful dynamical models have enabled significant progress in planning and
controlling robotic locomotion and manipulation. Traditionally, these models have been …

Benchmarking Rigid Body Contact Models

M Guo, Y Jiang, AE Spielberg… - Learning for Dynamics …, 2023 - proceedings.mlr.press
As robots are increasingly deployed in contact-rich tasks, there has been increased interest
in models of contact that are more accurate than those of untuned simulations. These …

Residual model learning for microrobot control

J Gruenstein, T Chen, N Doshi… - 2021 IEEE International …, 2021 - ieeexplore.ieee.org
A majority of microrobots are constructed using compliant materials that are difficult to model
analytically, limiting the utility of traditional model-based controllers. Challenges in data …

Experimental Validation of Nonsmooth Dynamics Simulations for Robotic Tossing involving Friction and Impacts

MJ Jongeneel, L Poort, N van de Wouw, A Saccon - 2023 - hal.science
In this paper, we evaluate the prediction performance of two nonsmooth rigid-body dynamics
simulators on realworld data with spatial impacts in the context of robotic tossing and visual …

[PDF][PDF] Predictive performance of nonsmooth rigid-body collision models for carton box impacts

L Poort - 2020 - research.tue.nl
With an additional boost due to the Covid-19 pandemic, the logistics market is growing faster
than ever before and the potential workforce cannot keep up with demands from industry …