用于从专家演示中推断非线性连续约束函数的正样本-无标签约束学习
Positive-Unlabeled Constraint Learning for Inferring Nonlinear Continuous Constraints Functions from Expert Demonstrations
AI总结：
针对机器人任务约束未知或难指定的问题，提出两步式PUCL算法，通过生成无标签不可行轨迹、识别可靠不可行数据来学习约束函数，在多环境中验证其精度与安全性优于基线。
摘要：
为各类现实世界机器人任务规划需要知晓并编写所有约束条件，但存在约束未知或难以准确指定的情况，一种可行的解决方案是从专家演示中推断未知约束。本文提出了一种新颖的两步式正样本-无标签约束学习（Positive-Unlabeled Constraint Learning，PUCL）算法，可从演示中推断连续约束函数，无需像现有研究那样预先掌握真实约束的参数化形式或环境模型。我们将演示中的所有数据视为正（可行）样本，并学习一个控制策略以生成潜在不可行的轨迹，将其作为无标签数据。所提出的两步学习框架首先通过距离度量识别可靠的不可行数据，随后利用可行演示数据与可靠的不可行数据学习一个二分类可行性分类器（即约束函数）。该方法能够灵活学习复杂形状的约束边界，且不会像以往方法那样将演示数据误分类为不可行。研究在四种带约束的环境中，分别采用网络化策略和动态系统策略验证了所提方法的有效性，结果表明该方法成功推断出连续非线性约束，在约束精度和策略安全性方面均优于其他基线方法。本研究已发表于《IEEE Robotics and Automation Letters (RA-L)》，最终版本详见https://doi.org/10.1109/LRA.2024.3522756
英文摘要：
Planning for diverse real-world robotic tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. This paper presents a novel two-step Positive-Unlabeled Constraint Learning (PUCL) algorithm to infer a continuous constraint function from demonstrations, without requiring prior knowledge of the true constraint parameterization or environmental model as existing works. We treat all data in demonstrations as positive (feasible) data, and learn a control policy to generate potentially infeasible trajectories, which serve as unlabeled data. The proposed two-step learning framework first identifies reliable infeasible data using a distance metric, and secondly learns a binary feasibility classifier (i.e., constraint function) from the feasible demonstrations and reliable infeasible data. The proposed method is flexible to learn complex-shaped constraint boundary and will not mistakenly classify demonstrations as infeasible as previous methods. The effectiveness of the proposed method is verified in four constrained environments, using a networked policy or a dynamical system policy. It successfully infers the continuous nonlinear constraints and outperforms other baseline methods in terms of constraint accuracy and policy safety. This work has been published in IEEE Robotics and Automation Letters (RA-L). Please refer to the final version at https://doi.org/10.1109/LRA.2024.3522756
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。
