RoCUS: Robot Controller Understanding via Sampling

Yilun Zhou, Serena Booth, Nadia Figueroa, Julie Shah
Proceedings of the 5th Conference on Robot Learning, PMLR 164:850-860, 2022.

Abstract

As robots are deployed in complex situations, engineers and end users must develop a holistic understanding of their behaviors, capabilities, and limitations. Some behaviors are directly optimized by the objective function. They often include success rate, completion time or energy consumption. Other behaviors – e.g., collision avoidance, trajectory smoothness or motion legibility – are typically emergent but equally important for safe and trustworthy deployment. Designing an objective which optimizes every aspect of robot behavior is hard. In this paper, we advocate for systematic analysis of a wide array of behaviors for holistic understanding of robot controllers and, to this end, propose a framework, RoCUS, which uses Bayesian posterior sampling to find situations where the robot controller exhibits user-specified behaviors, such as highly jerky motions. We use RoCUS to analyze three controller classes (deep learning models, rapidly exploring random trees and dynamical system formulations) on two domains (2D navigation and a 7 degree-of-freedom arm reaching), and uncover insights to further our understanding of these controllers and ultimately improve their designs.

Cite this Paper

BibTeX
@InProceedings{pmlr-v164-zhou22a, title = {RoCUS: Robot Controller Understanding via Sampling}, author = {Zhou, Yilun and Booth, Serena and Figueroa, Nadia and Shah, Julie}, booktitle = {Proceedings of the 5th Conference on Robot Learning}, pages = {850--860}, year = {2022}, editor = {Faust, Aleksandra and Hsu, David and Neumann, Gerhard}, volume = {164}, series = {Proceedings of Machine Learning Research}, month = {08--11 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v164/zhou22a/zhou22a.pdf}, url = {https://proceedings.mlr.press/v164/zhou22a.html}, abstract = {As robots are deployed in complex situations, engineers and end users must develop a holistic understanding of their behaviors, capabilities, and limitations. Some behaviors are directly optimized by the objective function. They often include success rate, completion time or energy consumption. Other behaviors – e.g., collision avoidance, trajectory smoothness or motion legibility – are typically emergent but equally important for safe and trustworthy deployment. Designing an objective which optimizes every aspect of robot behavior is hard. In this paper, we advocate for systematic analysis of a wide array of behaviors for holistic understanding of robot controllers and, to this end, propose a framework, RoCUS, which uses Bayesian posterior sampling to find situations where the robot controller exhibits user-specified behaviors, such as highly jerky motions. We use RoCUS to analyze three controller classes (deep learning models, rapidly exploring random trees and dynamical system formulations) on two domains (2D navigation and a 7 degree-of-freedom arm reaching), and uncover insights to further our understanding of these controllers and ultimately improve their designs. } }
Endnote
%0 Conference Paper %T RoCUS: Robot Controller Understanding via Sampling %A Yilun Zhou %A Serena Booth %A Nadia Figueroa %A Julie Shah %B Proceedings of the 5th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2022 %E Aleksandra Faust %E David Hsu %E Gerhard Neumann %F pmlr-v164-zhou22a %I PMLR %P 850--860 %U https://proceedings.mlr.press/v164/zhou22a.html %V 164 %X As robots are deployed in complex situations, engineers and end users must develop a holistic understanding of their behaviors, capabilities, and limitations. Some behaviors are directly optimized by the objective function. They often include success rate, completion time or energy consumption. Other behaviors – e.g., collision avoidance, trajectory smoothness or motion legibility – are typically emergent but equally important for safe and trustworthy deployment. Designing an objective which optimizes every aspect of robot behavior is hard. In this paper, we advocate for systematic analysis of a wide array of behaviors for holistic understanding of robot controllers and, to this end, propose a framework, RoCUS, which uses Bayesian posterior sampling to find situations where the robot controller exhibits user-specified behaviors, such as highly jerky motions. We use RoCUS to analyze three controller classes (deep learning models, rapidly exploring random trees and dynamical system formulations) on two domains (2D navigation and a 7 degree-of-freedom arm reaching), and uncover insights to further our understanding of these controllers and ultimately improve their designs.
APA
Zhou, Y., Booth, S., Figueroa, N. & Shah, J.. (2022). RoCUS: Robot Controller Understanding via Sampling. Proceedings of the 5th Conference on Robot Learning, in Proceedings of Machine Learning Research 164:850-860 Available from https://proceedings.mlr.press/v164/zhou22a.html.

Related Material