Joint Optimization of High-level Planning and Low-level Control Based on Contact State Recognition

Determine how to optimally design and jointly tune the high-level planning module that generates desired forces and moments from geometric and environmental constraints and the low-level controller (e.g., position-based force control or impedance control), using contact state recognition, for robotic peg-in-hole assembly.

Background

The contact model-based paradigm decomposes assembly into contact-state recognition and compliant control, further split into a high-level planner (force/moment targets) and a low-level controller. The literature often optimizes these components in isolation, overlooking their interdependence.

The paper explicitly states that it remains unclear how to best optimize these modules jointly based on contact-state recognition to achieve robust and adaptive assembly, highlighting a key gap in integrated system design.

References

Furthermore, it remains unclear how best to optimize the high-level planning module and the low-level controller according to the contact model recognition.

— Compare Contact Model-based Control and Contact Model-free Learning: A Survey of Robotic Peg-in-hole Assembly Strategies (1904.05240 - Xu et al., 2019) in Section 3, Compliant control, High-level planning module