Nonlinear Model Predictive Impedance Control of a Fully Actuated Hexarotor for Physical Interaction

Area of Science:

• Robotics
• Control Systems
• Aerospace Engineering

Background:

• Hexarotor drones are increasingly used for physical interaction tasks.
• Controlling compliant behavior and system constraints simultaneously is challenging.
• Existing impedance control methods may not fully address complex interaction dynamics.

Purpose of the Study:

• To propose a Nonlinear Model Predictive Impedance Control (NMPIC) for hexarotor physical interaction.
• To enable simultaneous handling of constraints and compliant behavior.
• To improve interaction force regulation and system stability.

Main Methods:

• Developed NMPIC by combining nonlinear model predictive control and impedance control.
• Integrated a disturbance observer to estimate and compensate for external forces.
• Implemented a weight adaptive strategy for online tuning of the cost function's weighting matrix.

Main Results:

• NMPIC effectively manages constraints while maintaining compliant behavior during hexarotor interaction.
• The disturbance observer accurately estimates external wrenches for improved model compensation.
• Simulations demonstrate superior performance and stability compared to general impedance controllers.

Conclusions:

• The proposed NMPIC method offers a robust solution for hexarotor physical interaction.
• The adaptive strategy enhances control performance and stability in dynamic scenarios.
• This research presents a novel approach for precise interaction force regulation in aerial robotics.