任务感知动态运动基元用于接触丰富操作中的故障检测与恢复
Task aware Dynamic Movement Primitives for failure detection and recovery in contact rich manipulation
浏览论文内容
中文总结 AI 辅助
本文提出任务感知故障检测与恢复框架，结合DMP轨迹生成与QDA阶段分类，利用马氏距离检测异常并触发螺旋搜索恢复，在轴孔装配中实现95%分类准确率及3毫米错位下的可靠恢复。
中文摘要 AI 辅助
在存在紧密公差和复杂接触交互的情况下，装配仍然是具有挑战性的机器人操作任务。虽然像动态运动基元（DMPs）这样的从演示中学习（LfD）框架能够有效地从单个演示中编码轨迹，但它们对初始抓取配置的变化和外部接触力高度敏感。这种变化通常会导致在接触丰富阶段的任务失败。本文提出了一种任务感知的故障检测与恢复框架，该框架将基于DMP的轨迹生成与实时阶段分类相结合。利用在多模态传感器数据上训练的二次判别分析（QDA），该框架将轴孔装配（PiH）操作执行分割为接近、对齐和插入阶段。通过使用目标相对位置数据作为特征，这种分类能够泛化到未见过的目标位置，而无需重新训练，与DMPs固有的泛化能力相匹配。异常检测通过计算力特征上的马氏距离度量在线进行，将接触引起的故障与名义轨迹执行区分开来。一旦检测到故障，触发螺旋搜索恢复策略，在接触下主动重新对齐销钉，然后恢复学习的DMP插入。所提出的方法在实验装置上进行了评估，实现了95%的阶段分类准确率，并展示了在仅使用单个演示的情况下，在高达3毫米的横向错位下可靠的故障恢复能力。
英文摘要
Assembly remains a challenging robotic manipulation task in presence of tight tolerances and complex contact interactions. While Learning from Demonstration (LfD) frameworks like Dynamic Movement Primitives (DMPs) can effectively encode trajectories from a single demonstration, they are highly sensitive to variations in initial grasp configurations and external contact forces. Such variations often lead to task failures during the contact rich phases. This paper presents a task aware failure detection and recovery framework that integrates DMP based trajectory generation with real time stage classification. Utilizing Quadratic Discriminant Analysis (QDA) trained on multimodal sensor data, the framework segments execution into approach, alignment, and insertion stages for a Peg in Hole (PiH) assembly operation. By using goal relative position data as features, this classification generalizes to unseen goal positions without requiring retraining, matching the inherent generalization capability of DMPs. Anomaly detection is performed online using a Mahalanobis distance metric computed over force features, isolating contact induced failures from nominal trajectory execution. Upon failure detection, a spiral search recovery policy is triggered to actively realign the peg under contact before resuming the learned DMP insertion. The proposed approach is evaluated on an experimental setup achieving 95% stage classification accuracy, and demonstrates reliable failure recovery under lateral misalignments of up to 3 mm using only a single demonstration.
发表机构
- Atomberg Technologies Limited（阿通伯格科技有限公司）
- Indian Institute of Technology Madras (IITM)（印度理工学院马德拉斯分校）
- Appian Corporation（阿皮安公司）
机构由 AI 辅助整理，请以论文原文为准。
