Adaptive dynamic feedback control of parallel robots with unknown kinematic and dynamic properties

Area of Science:

• Robotics
• Control Systems Engineering
• Mechatronics

Background:

• Uncertainties in dynamic and kinematic parameters are inherent challenges in robotic manipulator control.
• Traditional calibration methods are time-consuming and unsuitable for robots with slowly changing parameters, like deployable cable-driven robots.

Purpose of the Study:

• To develop an adaptive dynamic feedback controller to address parameter uncertainties in robotic manipulators.
• To enable effective control for deployable cable-driven robots where calibration is impractical.

Main Methods:

• An adaptive dynamic feedback controller with novel adaptation laws and states was designed.
• Jacobian matrix and its determinant were formulated in regressor form.
• A non-singular sliding surface was employed for trajectory tracking error.
• Lyapunov direct method was used to ensure fast finite-time trajectory tracking.

Main Results:

• The proposed controller effectively manages unknown parameters and external disturbances.
• Fast finite-time feasible trajectory tracking was achieved.
• The controller was validated on a 4RPR redundant rigid body and a 3-DOF cable-driven robot.

Conclusions:

• The adaptive controller overcomes limitations of traditional calibration for robotic manipulators.
• The method is suitable for deployable cable-driven robots and demonstrates superior performance compared to state-of-the-art techniques.