RAT iLQR: A Risk Auto-Tuning Controller to Optimally Account for Stochastic Model Mismatch

RAT iLQR: A Risk Auto-Tuning Controller to Optimally Account for Stochastic Model Mismatch
复制标题

RAT iLQR:一种风险自动调整控制器,可最佳地解决随机模型不匹配的问题

DOI:
10.1109/lra.2020.3048660
复制
发表时间:
2021
影响因子:
5.2
通讯作者:
Schwager, Mac
Schwager, Mac
中科院分区:
计算机科学2区
文献类型:
--
作者:
Nishimura, Haruki;Mehr, Negar;Gaidon, Adrien;Schwager, Mac

文献摘要

参考文献

被引文献

相似文献

机器人在随机环境中的成功操作依赖于对潜在概率分布的准确描述,但由于知识有限,这往往是不完美的。这项工作提出了一种能够处理这种分布失配的控制算法。具体地说,我们提出了一种新的用于分布鲁棒控制的非线性预测控制,它针对给定的KL发散界和高斯分布的最坏情况分布规划局部最优反馈策略。利用分布式稳健控制和风险敏感最优控制之间的数学等价性,我们的框架还提供了一种算法来在线动态调整风险敏感控制的风险敏感水平。在人体运动的预测分布是错误的动态碰撞避免场景中,展示了分布稳健性以及自动风险敏感性调整的好处。
Successful robotic operation in stochastic environments relies on accurate characterization of the underlying probability distributions, yet this is often imperfect due to limited knowledge. This work presents a control algorithm that is capable of handling such distributional mismatches. Specifically, we propose a novel nonlinear MPC for distributionally robust control, which plans locally optimal feedback policies against a worst-case distribution within a given KL divergence bound from a Gaussian distribution. Leveraging mathematical equivalence between distributionally robust control and risk-sensitive optimal control, our framework also provides an algorithm to dynamically adjust the risk-sensitivity level online for risk-sensitive control. The benefits of the distributional robustness as well as the automatic risk-sensitivity adjustment are demonstrated in a dynamic collision avoidance scenario where the predictive distribution of human motion is erroneous.
DOI: 10.1109/cdc.2016.7799215
发表时间: 2016-12
期刊: 2016 IEEE 55th Conference on Decision and Control (CDC)
影响因子: --
作者:
Ioannis Exarchos;Evangelos A. Theodorou;P. Tsiotras
通讯作者: Ioannis Exarchos;Evangelos A. Theodorou;P. Tsiotras
DOI: --
发表时间: 2020-09
期刊: ArXiv
影响因子: --
作者:
B. Ivanovic;Amine Elhafsi;G. Rosman;Adrien Gaidon;M. Pavone
通讯作者: B. Ivanovic;Amine Elhafsi;G. Rosman;Adrien Gaidon;M. Pavone
DOI: 10.1109/ccta.2017.8062463
发表时间: 2017
期刊: 2017 IEEE Conference on Control Technology and Applications (CCTA
影响因子: --
作者:
Samuelson, Samantha;Yang, Insoon
通讯作者: Yang, Insoon
一种分布式稳健的黑盒优化方法
DOI: --
发表时间: 2018
期刊:
影响因子: --
作者:
A. Philpott;Michael G. Kapteyn;K. Willcox
通讯作者: K. Willcox
Wasserstein 使用条件风险值进行具有安全约束的分布式鲁棒运动规划和控制
DOI: 10.1109/icra40945.2020.9196857
发表时间: 2020
期刊: 2020 IEEE International Conference on Robotics and Automation (ICRA)
影响因子: --
作者:
A. Hakobyan;Insoon Yang
通讯作者: Insoon Yang