Controller Design For The Constrained Flexible Robots

In the operation,robots are more likely to contact with people or the environment,and the need for a safe robot system is urgent.This system can be widely used in the robot system when interacting with people or circumstances,and this system can ensure the safety of staff and reduce the peripheral equipment and environmental damage.Compared to the traditional robot,the flexible robot interaction with environment or human normally shows the superior performance of security,robustness and flexibility,the series elastic actuator of a new generation of robots has attracted much attention,but the problems such as nonlinear problems caused by stiffness variable because of the elastic element,the unknown driver dynamic model,make it challenging for the control of such robots.At the same time in the control system,we have higher requirements for the control precision and speed in a variety of precision machining and production equipments.Due to the back-like hysteresis and time-varying delay of advanced sensors and drive technology,it has practical significance to doing research on the robot system with back-like hysteresis and time-varying delay constraints.This paper study various problems that flexible robot system often faces,such as nonlinear problems caused by stiffness variable because of the elastic element,the unknown driver dynamic model,back-like hysteresis and time-varying delay.To solve these problems,this paper uses adaptive control and other intelligent control methods,based on Lyapunov direct method,we propose corresponding controller design method.In order to solve the problem of model uncertainty,adaptive control method and fuzzy logic system(FLS)and neural network(NNs)theory are used in this paper.The disturbance observer can be used to estimate and compensate the synthetic interference terms in the system,that is,the external disturbance,the input nonlinear term and the approximation error of the system model.The corresponding adaptive controller design method for robotic nonlinear system is proposed,which can make the system reach high control precision.The main research contents of this paper can be summarized as:(1)Taking the unknown hysteresis input existing in the MIMO system into account,we do research on a dynamic hysteresis model of this kind of problem,and present the effective method to weaken the hysteresis effect;(2)Considering back-like hysteresis and time-varying delay,we use the adaptive fuzzy control method to study the parameters of multiple input multiple output(MIMO)tracking control system;(3)Considering series elastic actuator and the difficult to obtain robot precise model,we use adaptive neural network control method to do research on the control of such flexible robot;(4)We do experiments on the Baxter robot simulation and experimental platform which is driven by series elastic actuator to verify the control method presented in this paper,analyse the performance of the controller,and compare it to other methods in the literature.