KAI: A Kinematic-Aware Interface for Data-Efficient Articulated Object Manipulation
2026-07-27 • Robotics
Robotics
AI summaryⓘ
The authors created KAI, a tool that helps robots understand how parts of objects move together, making it easier to learn how to manipulate them. By using KAI, robots can learn effective motion control from fewer examples, especially when data is limited. Their approach also works well in new or messy environments and can improve by using videos of people interacting with objects. Overall, this helps robots handle articulated objects more efficiently and reliably.
articulated object manipulationkinematic structurepolicy learninginductive biassample efficiencysimulation tasksgeneralizationvisual distractorshuman interaction videostransfer learning
Authors
Yaping Li, Zhaxizhuoma, Qiaojun Yu, Jia Zeng, Dahua Lin, Jiangmiao Pang
Abstract
Articulated object manipulation requires an understanding of kinematic structure that is difficult and costly to learn from robot demonstrations alone. We introduce the Kinematic-Aware Articulation Interface (KAI), a structured intermediate representation that captures the kinematic structure of articulated objects. By embedding interpretable geometric and kinematic priors into policy learning, KAI provides a strong inductive bias aligned with the underlying structure of articulated motion. This design effectively improves sample efficiency, with gains particularly pronounced in low-data regimes: across six simulation tasks, our method achieves an average success rate of 82.9%, matching or surpassing baseline performance while using only half the demonstration data. Our method also exhibits robust generalization to unseen backgrounds and visual distractors, transferring from a single clean training environment to cluttered real-world scenes. KAI's action-agnostic design further enables co-training with human interaction videos to enhance real-world robustness: under diverse visual distractions, our method with video co-training achieves over 70% average success rate.