Neural-learning enhanced admittance control of a robot manipulator with input saturation

We propose a cascaded impedance control algorithm based on a virtual dynamics model (VDM) to achieve robust and effective mechanical impedance for a robot interacting with unknown environments. This cascaded controller consists of an internal loop of virtual impedance control based on a VDM and an external loop of impedance reference control. The VDM-based virtual impedance control can achieve the same effect as the conventional admittance control; its intermediate output of force/torque serves as the input for the external loop reference impedance control. Therefore, this cascaded controller shows superior performance by combining the advantages of admittance control and impedance control. We evaluate the controller in multiple-contact experiments on a six-degrees of freedom (6-DOF) industrial robot manipulator. The result shows that under various contact situations such as soft and rigid surfaces and free space, the proposed method can rapidly track the target and effectively maintain stability. In the experiments conducted on the robot in contact with various environments, the proposed control method reduced the steady-state error by more than 20% compared with the conventional admittance control.

Download Full-text

Disturbance observer-based composite neural learning path following control of underactuated ships subject to input saturation

Ocean Engineering ◽

10.1016/j.oceaneng.2020.108033 ◽

2020 ◽

Vol 216 ◽

pp. 108033 ◽

Cited By ~ 1

Author(s):

Guoqing Zhang ◽

Chenliang Zhang ◽

Tingting Yang ◽

Weidong Zhang

Keyword(s):

Disturbance Observer ◽

Input Saturation ◽

Path Following ◽

Neural Learning ◽

Learning Path ◽

Path Following Control

Download Full-text

Safe and intuitive manual guidance of a robot manipulator using adaptive admittance control towards robot agility

Robotics and Computer-Integrated Manufacturing ◽

10.1016/j.rcim.2021.102127 ◽

2021 ◽

Vol 70 ◽

pp. 102127

Author(s):

Daniel Reyes-Uquillas ◽

Tesheng Hsiao

Keyword(s):

Robot Manipulator ◽

Admittance Control

Download Full-text

The Windup Phenomenon and Anti-windup Illustrated

Modern Anti-windup Synthesis ◽

10.23943/princeton/9780691147321.003.0001 ◽

2011 ◽

Author(s):

Luca Zaccarian ◽

Andrew R. Teel

Keyword(s):

Disturbance Rejection ◽

Feedback Loop ◽

Robot Manipulator ◽

Input Saturation ◽

Positioning System ◽

System A ◽

Highly Oscillatory ◽

Oscillatory Response ◽

Spring System ◽

Mass Spring

This chapter uses examples where windup occurs to motivate anti-windup synthesis. In particular, it discusses a SISO academic example, a MIMO academic example, the longitudinal dynamics of an F8 aircraft, a servo-positioning system, the damped mass-spring system, the experimental spring-gantry system, a robot manipulator, and a disturbance rejection problem. The examples illustrate the windup phenomenon, which may arise due to input saturation in a feedback loop, and the capabilities of anti-windup synthesis in mitigating the windup phenomenon. They show that windup can manifest itself as a sluggish response, a highly oscillatory response, or a diverging response. In each example, alternatives to anti-windup synthesis include investing in actuators with more capabilities, or redesigning the controller from scratch to account for input saturation directly. These strategies should be considered when the control system's actuators are continuously trying to act beyond their limits.

Download Full-text