cs.ROApr 9, 2026

A Unified Multi-Layer Framework for Skill Acquisition from Imperfect Human Demonstrations

Authors: Zi-Qi Yang, Mehrdad R. Kermani

Organizations: Department of Electrical and Computer Engineering, Western University, London, ON N6A 5B9, Canada.

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Current Human-Robot Interaction (HRI) systems for skill teaching are fragmented, and existing approaches in the literature do not offer a cohesive framework that is simultaneously efficient, intuitive, and universally safe. This paper presents a novel, layered control framework that addresses this fundamental gap by enabling robust, compliant Learning from Demonstration (LfD) built upon a foundation of universal robot compliance. The proposed approach is structured in three progressive and interconnected stages. First, we introduce a real-time LfD method that learns both the trajectory and variable impedance from a single demonstration, significantly improving efficiency and reproduction fidelity. To ensure high-quality and intuitive {kinesthetic teaching}, we then present a null-space optimization strategy that proactively manages singularities and provides a consistent interaction feel during human demonstration. Finally, to ensure generalized safety, we introduce a foundational null-space compliance method that enables the entire robot body to compliantly adapt to post-learning external interactions without compromising main task performance. This final contribution transforms the system into a versatile HRI platform, moving beyond end-effector (EE)-specific applications. We validate the complete framework through comprehensive comparative experiments on a 7-DOF KUKA LWR robot. The results demonstrate a safer, more intuitive, and more efficient unified system for a wide range of human-robot collaborative tasks.

Figures & tables

Fig. 1 : Framework to achieve intuitive and efficient skill teaching on a KUKA robot in a physical HRI setup: (i) learning from demonstration and path reproduction under perturbation (solid red: demonstration; dashed green: perturbed reproduction; green dot: start; red dot: end) (ii) null space optimization for pose adjustment (white arrow) and singularity avoidance, enabling high-quality single-handed demonstration of writing “2025”, which is challenging to execute precisely with one arm; and (iii) null space compliance that allows active user interactions (green arrows) without interfering the main task.
Fig. 2 : Experiment A: (a,i) 3D FDM learned path (with a 2D inset for easier visualization), (a.ii–a.iv) estimation error of the learned path, reproduction error, and reproduction under perturbation for a trapezoid-shaped path; (b.i–b.iv) corresponding results for a ‘W’-shaped path. Light blue shaded areas indicate the period of interaction, with the corresponding impact highlighted in circles. All values are expressed in centimeters, and all paths start at zero height ( zz -axis).
Fig. 3 : Experiment B: (a) Measured velocity and force for an L-shaped path using the proposed method (i-ii) and consistency comparison with the GC method, expressed in isotropic representation (iii-iv). (b) Singularity avoidance test showing xx -axis position (i) and interaction force (ii); shaded areas indicate interaction, with release causing backlash (circled).
Fig. 4 : Experiment C: (i) tracking desired path in the XY plane, with external interactions’ impacts labelled with I and II; (ii) joint deviations due to external interactions. (iii) position variation along the zz -axis. Light blue shaded areas indicate the period of interaction.
More in this topic

Compliant Robot Control

See all 15 papers

Explore similar work

CardsList

  1. User-Driven Learning from Demonstration: A Trajectory and Impedance Learning Method

    Jul 18, 2026Zi-Qi Yang, Mehrdad R. KermaniRobotic ControlRobot Motion Generation

    Department of Electrical and Computer Engineering, Western University, London, ON N6A 5B9, Canada.

  2. Data-Efficient Learning from Demonstration for Humanoid Robots via Feature-Level Perception and Prompt-Based Object Detection

    Oct 19, 2026Keyhan Rayati, AliReza Beigy, Anoosheh Saadati +4Robot Skill LearningLearning from Demonstration

    Human and Robot Interaction Lab, University of Tehran, Tehran, Iran · Intelligent Robotics Laboratory, Oakland University, Rochester, Michigan, USA

  3. Human-to-Robot Interaction: Learning from Video Demonstration for Robot Imitation

    Feb 22, 2026Thanh Nguyen Canh, Thanh-Tuan Tran, Haolan Zhang +3Learning from DemonstrationRobot Imitation Learning

    School of Information Science, Japan Advanced Institute of Science and Technology, Nomi, 923-1211, Ishikawa, Japan. · University of Engineering and Technology, Vietnam National University, 10000, Hanoi, Vietnam. · Department of Robotics, Hanyang University, Ansan, 15588, Gyeonggi, Korea.