Validating Robotics Simulators on Real-World Impacts

Brian Acosta*, William Yang*, Michael Posa - RA-L 2022

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 the centrality of simulation in modern robotics, little work has been done to compare the performance of robotics simulators against real-world data, especially for scenarios involving dynamic motions with high speed impact events. Handling dynamic contact is the computational bottleneck for most simulations, and thus the modeling and algorithmic choices surrounding impacts and friction form the largest distinctions between popular tools. Here, we evaluate the ability of several simulators to reproduce real-world trajectories involving impacts. Using experimental data, we identify system-specific contact parameters of popular simulators Drake, MuJoCo, and Bullet, analyzing the effects of modeling choices around these parameters. For the simple example of a cube tossed onto a table, simulators capture inelastic impacts well while failing to capture elastic impacts. For the higher-dimensional case of a Cassie biped landing from a jump, the simulators capture the bulk motion well but the accuracy is limited by numerous model differences between the real robot and the simulators.

paper - code - dataset

Impact Invariant Control with Applications to Bipedal Locomotion

William Yang, Michael Posa - IROS 2021

When legged robots impact their environment, they undergo large changes in their velocities in a small amount of time. Measuring and applying feedback to these velocities is challenging, and is further complicated due to uncertainty in the impact model and impact timing. This work proposes a general framework for adapting feedback control during impact by projecting the control objectives to a subspace that is invariant to the impact event. The resultant controller is robust to uncertainties in the impact event while maintaining maximum control authority over the impact invariant subspace. We demonstrate the utility of the projection on a walking controller for a planar five-link-biped and on a jumping controller for a compliant 3D bipedal robot, Cassie. The effectiveness of our method is shown to translate well on hardware.

paper - code

Impact-Invariant Control: Maximizing Control Authority During Impacts

William Yang, Michael Posa - Submitted to TRO 2023

(Journal Version of "Impact-invariant Control with Applications to Bipedal Locomtion")

When legged robots impact their environment, they undergo large changes in their velocities in a short amount of time. Measuring and applying feedback to these velocities is challenging, further complicated by uncertainty in the impact model and impact timing. This work proposes a general framework for adapting feedback control during impact by projecting the control objectives to a subspace that is invariant to the impact event. The resultant controller is robust to uncertainties in the impact event while maintaining maximum control authority over the impact-invariant subspace. We demonstrate the improved performance of the projection over other commonly used heuristics on a walking controller for a planar five-link-biped. The projection is also applied to jumping, box jumping on to a platform 0.4 m tall, and running controllers for the compliant 3D bipedal robot, Cassie. The modification is easily applied to these various controllers and is a critical component to deploying on the physical robot.

paper - code

Spine morphology and energetics: how principles from nature apply to robotics

Yevgeniy Yesilevskiy, William Yang, and C. David Remy - Bioinspiration & biomimetics 2018

Inspired by the locomotive advantages that an articulated spine enables in quadrupedal animals, we explore and quantify the energetic effect that an articulated spine has in legged robots. We compare two model instances of a conceptual planar quadruped: one with a traditional rigid main body and one with an articulated main body with an actuated spinal joint. Both models feature four distinct legs, series elastic actuation, distributed mass in all body segments, and limits on actuator torque and speed. Using optimal control to find the energetically optimal joint trajectories, actuator inputs, and footfall timing, we examine and compare the positive mechanical work cost of transport of both models across multiple gaits and speeds. Our results show that an articulated spine increases the maximum possible speed and improves the locomotor economy at higher velocities, especially for asymmetrical gaits. The driving factors for these improvements are the same mechanistic effects that facilitate asymmetrical gaits in nature: improved leg recirculation, elastic energy storage in the spine, and enlarged stride lengths.