🤖 AI Summary
Redundant manipulator inverse kinematics (IK) poses challenges due to non-convex optimization induced by solution multiplicity, while conventional supervised learning suffers from poor generalization and low data efficiency. To address these issues, this paper proposes an embodied self-supervised learning framework featuring a novel sampling-training co-adaptation mechanism: online data generation and model updating operate in a closed loop to resolve solution ambiguity. We further design batched inference and parallel sampling strategies to accelerate training convergence, and introduce a lightweight model adaptation method enabling configuration transfer within minutes. Experiments demonstrate significant improvements in IK solution accuracy and convergence speed; data sampling efficiency increases by 3.2×; and rapid generalization to unseen manipulator configurations is achieved.
📝 Abstract
Forward and inverse kinematic models are fundamental to robot arms, serving as the basis for the robot arm's operational tasks. However, in model learning of robot arms, especially in the presence of redundant degrees of freedom, inverse kinematic model learning is more challenging than forward kinematic model learning due to the non-convex problem caused by multiple solutions. Besides, Current learning-based methods often segregate data sampling from model training, potentially leading to suboptimal data utilization and restricted model adaptability. In this paper, we introduce the concept of “Embodiment” and propose a framework for autonomous learning of the robot arm inverse kinematic model based on embodied self-supervised learning (EMSSL) with sampling and training coordination, effectively solving the non-convex problem of the inverse kinematic model and significantly enhancing data sampling efficiency. Concurrently, we investigate batch inference and parallel computation strategies for data sampling to expedite model learning. Additionally, we develop two approaches for the fast adaptation of the robot arm models. A series of experimental evaluations attest to the efficacy of our proposed method.
