Advanced Learning Control in Physical Interaction Tasks

Robotics are increasingly and urgently expected to acquire human-like dexterous manipulation skills in physical interaction environments. Due to this, the loop between the high-level action policy and the low-level motion execution needs to be closed by developing advanced data-driven or model-based learning and control approaches. Although recent studies have been shown to demonstrate promising results and advances of the closed-loop learning control algorithms, several key issues remain quite changeling.

Typically, three problems in this field of research are yet to be solved:

1) how to integrate learning and control models seamlessly for more dexterous manipulation and interaction performances

2) how to compute learning and control policies from multimodal/cross modal data

And 3) how to make use of advances of both data-driven and model-based models for compliant and flexible interactions

The goal of this Research Topic is to bring together the newest theoretical findings and experimental results in advanced learning control applied to robot-environment physical interaction systems.

This Research Topic encourages paper contributions, which are related to but not limited to:
- Learning control from demonstrations, especially from multimodal demonstrations
- Probabilistic and statistical methods for learning control
- Reinforcement learning based adaptive control for contact-rich tasks, especially for long-term tasks
- Impedance/ admittance/force control for compliant manipulation
- Advanced control methods (e.g., iterative learning control) applied to robot-environment interaction scenarios
- Applications of learning control in physical interactions, e.g, industrial, medical, and rehabilitation tasks