arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1809.07874math.OCcs.ROcs.SYeess.SY
收藏

基于信息瓶颈的任务驱动估计与控制

Task-Driven Estimation and Control via Information Bottlenecks

Princeton University(普林斯顿大学)

Vincent Pacelli, Anirudha Majumdar

AI总结:

本文提出了一种基于信息瓶颈理论的任务驱动估计与控制框架,通过任务相关的变量构建高效表示,并设计了在线算法进行状态估计以实现鲁棒的控制策略。

摘要:

我们的目标是开发一种原理明确且通用的算法框架,用于机器人系统的任务驱动估计与控制。最先进的机器人控制系统通常依赖于准确估计机器人完整状态(例如,运行中的机器人可能估计关节角度和速度、躯干状态以及相对于目标的位置)。然而,完整状态表示通常对于特定任务来说过于丰富,可能导致显著的计算不高效和对状态估计误差的脆弱性。相反,我们提出了一种方法,避免使用这种丰富的表示,并试图创建任务驱动的表示。关键的技术洞察是利用信息瓶颈理论来形式化“任务驱动表示”的概念,以信息理论量度来衡量表示的最小性。我们提出了新的迭代算法,用于自动合成(离线)任务驱动表示(以一组任务相关变量(TRVs)给出)和一个以TRVs为函数的高效控制策略。我们还提出了在线算法来估计TRVs,以便应用控制策略。我们证明了我们的方法在理论和彻底的模拟实验(包括向目标位置奔跑的弹簧加载倒立摆)中都对未建模的测量不确定性具有显著的鲁棒性。

英文摘要:

Our goal is to develop a principled and general algorithmic framework for task-driven estimation and control for robotic systems. State-of-the-art approaches for controlling robotic systems typically rely heavily on accurately estimating the full state of the robot (e.g., a running robot might estimate joint angles and velocities, torso state, and position relative to a goal). However, full state representations are often excessively rich for the specific task at hand and can lead to significant computational inefficiency and brittleness to errors in state estimation. In contrast, we present an approach that eschews such rich representations and seeks to create task-driven representations. The key technical insight is to leverage the theory of information bottlenecks}to formalize the notion of a "task-driven representation" in terms of information theoretic quantities that measure the minimality of a representation. We propose novel iterative algorithms for automatically synthesizing (offline) a task-driven representation (given in terms of a set of task-relevant variables (TRVs)) and a performant control policy that is a function of the TRVs. We present online algorithms for estimating the TRVs in order to apply the control policy. We demonstrate that our approach results in significant robustness to unmodeled measurement uncertainty both theoretically and via thorough simulation experiments including a spring-loaded inverted pendulum running to a goal location.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑