Impedance Controller With Adaptive Stiffness For The Dual- Arm Cooperative Manipulators Handling Varying Payload

Human could complete many tasks using their two arms, such as grasping objects for assembly or polish under unstructured environments. The traditional manipulators have been widely used in industrial application to replace the human for simple and tedious task. However, they could not complete such work like human does. Imitation of biological mechanism of human arms cooperation and the development of dual-arm cooperative robot will be the developing direction of the next generation and be the key of industrial robots. Tackling and solving the problem caused by varying payload is a key to figure out the issues for the Dual-arm cooperative robot in practical application, which have crucial theoretical and practical value.This work depend on the project of "Multifunctional Redundant Industrial Robot with two Arms", which is supported by science and technology support plan key projects of Jiangsu province. We have conducted an in-depth research on the dual-arm cooperative robot, have developed the robot operating and hardware system for the redundant dual-arm cooperative robot, have designed the nonlinear system based on the robot dynamic model, and have done a research on dual object impedance control with variable object stiffness for the dual- arm cooperative manipulators. In addition, the effectiveness of the proposed strategy and theoretical discussion are proved by simulation and experiment results. The main work and Innovation can be listed as follows;1) An Adaptive Fuzzy Sliding Mode Controller with Nonlinear Observer (AFSMCO) for the redundant robotic manipulator is proposed. A suitable observer using the recursive algorithm is presented for an accurate estimation of external disturbances caused by the varying external force. The adaptive fuzzy logic is introduced to update switching gain of the sliding mode controller (AFSMC) to avoid chattering in real time. It is shown that the control law of the SMC designed based on Lyapunov theory could guarantee stability of the close-loop. This approach could achieve a precise trajectory tracking for a robot manipulator in the presence of uncertainties and disturbances.2) The dynamic decomposition optimization approach of the internal force is proposed based on the minimal energy cost. The grasp matrix and the kinematic and dynamic constrain among the dual-arm cooperative robot and objective are designed and analyzed respectively, via establishing the couple dynamic model of the dual-arm cooperative robot and objective object. The Minimum Norm and Zero space method are employed to decompose the resultant force of the objective into internal and external forces respectively and the minimal energy consumption is adopted as the cost function to optimize again the internal force, such that projection for the resultant force of the object to the operation space of the robots is obtained. Indeed, the theoretical models are used for force control in the next chapter.3) Two-loop impedance controllers with variable stiffness (DIVS) is proposed. Inner-loop and outer-loop impedance control controllers are used for the internal and external forces respectively, by introducing reference external force as the input for the outer-loop controller; in so doing, the stable interaction between object and environment is fulfilled. Based on the property of human’s arm. impedance control strategy with variable stiffness based on PD algorithm is proposed and is characterized by strong robustness. In order to prevent the internal force from destroying the object, the similar strategy is introduced to inner-loop impedance control. In addition, the out-loop stability, inter-loop stability, and completed closed-loop stability are proved by using Lyapunov theory. Generally, two-loop impedance control strategy guarantees closed-loop stability for the cooperative grasping task and the interaction with the external environment.4) Embedded operating system of the dual-arm cooperative robot is developed.In order to deal with the variable task and complex algorithm, the open source Linux+RTAI is selected to serve as operating system of the dual-arm cooperative robotwith sensory perceptual system such as, visual perception, force perception, current measurement built and master control system with capability of fast computation and data processing developed. Communication mode which is based on bus could dramatically improve the speed and accuracy of information exchange among each control nodes and total master control system.5) The experiment platform is developed and the effectiveness of the algorithm is verified. In order to solve the issues caused by the varying payload to the dual-arm cooperative robot, the experiments of the manipulator with varying payload, dual-arm cooperative robot with varying external force interacting with environment, and dual-arm cooperative robot implementing assembly task, are conducted to test the proposed algorithm on dual-arm cooperative robot and tested on a dual-arm cooperative robot platform. Overall, the effectiveness of the proposed strategy and theoretical discussion are proved by experiment results.