BriefGPT.xyz
Dec, 2018
从演示中学习限制
Learning Constraints from Demonstrations
HTML
PDF
Glen Chou, Dmitry Berenson, Necmiye Ozay
TL;DR
本论文提出了一种通过hit-and-run抽样方法，利用任务演示、成本函数以及系统动力学和控制约束知识，学习跨任务共享的未知约束的方法。该方法可以学习系统动力学的不确定性，学习一定量的约束，并且适用于线性和非线性系统的控制。
Abstract
We extend the
learning from demonstration
paradigm by providing a method for learning
unknown constraints
shared across tasks, using demonstrations of the tasks, their cost functions, and knowledge of the
→