Advances in Learning-Based Robotic Interaction Control: Impedance Control, Teleoperation, and Intelligent Human-Robot Collaboration

Historically, impedance control and teleoperation have provided the groundwork for compliant interaction between robots and humans, enabling systems to manage the dynamic interplay between force and motion. However, classical control methods often struggle with unstructured environments, unmodeled uncertainties, and varying human behaviors. The recent convergence of robotics and artificial intelligence has introduced transformative capabilities: learning algorithms, predictive modeling, and adaptive control strategies now allow robots to anticipate human actions, compensate for nonlinearities, and personalize their interaction behaviors. This integration is driving a new generation of intelligent robotic systems that are not only reactive but proactively adapt to dynamic and uncertain environments, enhancing precision, safety, and responsiveness. Within this broader context, applications span industrial automation, service robots, remote manipulation, and biomedical robotics—where AI-driven adaptive control enables personalized rehabilitation, assistive devices, and real-time health monitoring—demonstrating the diversity and impact of intelligent interaction control. Understanding and advancing this integration is critical for the next leap in human-centered robotics.

This Research Topic seeks to highlight cutting-edge advances in enhanced interaction control strategies that address the challenges of real-time interaction, autonomy, and adaptability in human-robot systems. Classical methods such as impedance control and teleoperation are increasingly being complemented by machine learning, reinforcement learning, and predictive modeling—enabling robots to better interpret human intent, handle nonlinear and uncertain dynamics, and facilitate more natural and effective physical collaboration.

We welcome contributions that demonstrate how learning-based impedance control, intelligent teleoperation, and adaptive human-robot collaboration can drive progress not only in therapeutic and assistive robotics—through personalized rehabilitation, real-time health monitoring, and user-specific adaptation—but also in broader applications of automation and physical interaction. These include industrial co-robots, remote manipulation systems, and interactive service robots.

While a strong emphasis is placed on biomedical applications due to their growing significance, we equally encourage interdisciplinary research targeting interaction control algorithms that are safer, smarter, and more robust across diverse real-world environments.

This Research Topic encourages high-quality contributions at the intersection of artificial intelligence and interaction control including human-robot interaction, learning-based control, and adaptive physical collaboration. Topics of interest include, but are not limited to:

1. Impedance and admittance control strategies for compliant and responsive physical interaction.
2. Teleoperation systems enhanced with adaptive control, predictive modeling, or user-in-the-loop feedback.
3. Reinforcement and imitation learning applied to dynamic and interactive control tasks.
4. Human intent recognition and intent-driven adaptation for personalized interaction.
5. Adaptive and intelligent human-robot interaction strategies integrating control and AI techniques.
6. Shared autonomy and semi-autonomous control in assistive, therapeutic, and remote operation scenarios.
7. Stability, robustness, and safety assurance in learning-based control architectures.
8. Biomedical robotics applications, including rehabilitation devices, assistive exoskeletons, and prosthetics.
9. Hybrid control systems combining classical methods with AI for effective human-robot collaboration.
10. Physical interaction and collaboration in unstructured and dynamic environments.

We particularly encourage submissions demonstrating experimental validation, real-world deployments, or interdisciplinary approaches combining control systems, robotics, biomedical engineering, and AI-based modeling.

Topic editor Hamza Khan is employed by Ronfic Co. Ltd. The other Topic Editors report no competing interests related to this Research Topic.