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 2406.11984cs.ROcs.AI
收藏

基于主动推理的复杂任务在线帕累托最优决策

Online Pareto-Optimal Decision-Making for Complex Tasks using Active Inference

University of Colorado Boulder(科罗拉多大学博尔德分校)

Peter Amorese, Shohei Wakayama, Nisar Ahmed, Morteza Lahijanian

AI总结:

针对不确定环境下机器人复杂任务多目标平衡与安全执行问题,提出含规划层与选择层的主动推理多目标强化学习框架,可学习最优权衡、对齐用户偏好且性能优于现有方法。

摘要:

当机器人自主执行复杂任务时,通常需要在保障安全性的同时平衡多个相互冲突的目标,在具有随机结果的不确定环境中这一问题会更具挑战性。提升机器人行为的透明度并使其与用户偏好对齐也至关重要。本文提出了一种新型多目标强化学习框架,可保障任务安全执行、优化多目标间的权衡并遵循用户偏好。该框架包含两个核心层级:多目标任务规划器与高层选择器。规划层生成一组可保证满足时序逻辑任务要求的最优权衡方案;选择器利用active inference(主动推理)判定生成的方案中最契合用户偏好的选项,并辅助学习过程。框架以迭代方式运行,基于收集到的数据更新参数化学习模型。在操作机器人与移动机器人上开展的案例研究与基准测试表明,该框架性能优于其他方法,且具备三点特性:(i)可学习多种最优权衡方案;(ii)能够遵循用户偏好;(iii)允许用户调整(i)与(ii)之间的平衡。

英文摘要:

When a robot autonomously performs a complex task, it frequently must balance competing objectives while maintaining safety. This becomes more difficult in uncertain environments with stochastic outcomes. Enhancing transparency in the robot's behavior and aligning with user preferences are also crucial. This paper introduces a novel framework for multi-objective reinforcement learning that ensures safe task execution, optimizes trade-offs between objectives, and adheres to user preferences. The framework has two main layers: a multi-objective task planner and a high-level selector. The planning layer generates a set of optimal trade-off plans that guarantee satisfaction of a temporal logic task. The selector uses active inference to decide which generated plan best complies with user preferences and aids learning. Operating iteratively, the framework updates a parameterized learning model based on collected data. Case studies and benchmarks on both manipulation and mobile robots show that our framework outperforms other methods and (i) learns multiple optimal trade-offs, (ii) adheres to a user preference, and (iii) allows the user to adjust the balance between (i) and (ii).

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。

↑