Research On Optimal Fault-tolerant Control For Multi-fault Reconfigurable Manipulators With Input Constraints

In recent years,with the rapid development of intelligent control,China has made significant breakthroughs in technologies such as aerospace,deep-sea exploration,and intelligent medical diagnosis.For example,reconfigurable robotic arms assist in precise docking of aerospace orbits,detecting submarine cable laying,and assisting in limb rehabilitation training,all of which intuitively demonstrate the important role of reconfigurable robotic arms in the development of intelligent control.The flexible configuration of reconfigurable robotic arms benefits from their standardized physical electrical interface joints and linkage structures.Based on this,reconfigurable robotic arms can be applied in scenarios with diverse tasks and complex working environments,thereby completing various types of technical tasks.However,in practical production and life,highintensity work tasks and unknown complex external environments may lead to certain system failures.Data statistics show that sensors and actuators are the two devices with the highest failure rate.If system failures are not adjusted and repaired in a timely manner,it will cause significant safety hazards.In order to prevent accidents before they occur,faulttolerant control is designed for reconfigurable robotic arm systems that fail,which can better ensure the safety,stability,and risk resistance of system control performance.The anti saturation fault-tolerant control of reconfigurable robotic arms with multiple faults can help achieve efficient production and has broad practical significance in industrial production.Considering the reality,the output capacity of the actuator cannot be infinitely large.In the process of fault tolerant control,the control torque of the system will increase greatly due to the fault compensation effect.Once the maximum output value of the actuator is exceeded,the components will be damaged,leading to actuator failure and system paralysis.In order to avoid the problem of excessive control torque caused by fault compensation,it is particularly important to design anti saturation fault-tolerant controllers for reconfigurable robotic arm systems that experience faults.At the same time,reconfigurable robotic arms are widely used,which means that the working environment in which the system operates is diverse,which may hide many uncertain risks and unstable factors.Based on such situations,it is necessary to incorporate environmental uncertainty constraints into the consideration of actuator saturation,in order to comprehensively design the system controller.Combining external environment and one’s own capabilities is of great significance in comprehensively improving system control accuracy and improving energy utilization efficiency.Nowadays,the reconstructed robotic arm system with multiple faults has many excellent results based on optimal control,but at the same time considering that the saturation and actual environmental uncertainty of the actuator is relatively small.This article proposes a multi-fault-restructured robot anti-saturation fault tolerance control method based on adaptive dynamic planning,rectification,rectification.The articles are summarized as follows:(1)Combined with the Newton-Euler iteration algorithm,the traditional dynamic model of reconstructed robotic arm is obtained through the positive and reverse iterative algorithms.At the same time,the reconstructed robotic arm state space expression of sensor failure and performer failure is obtained.(2)For sensors and actuators,complications can reconstruct the robotic arm system,combined with the principle of micro-division and embryo,and the sensor faults are converted into pseudo-executors.Introduce the hyperbolic positive cut function to solve the saturation problem,use multi-fault observer to achieve real-time improvement of performance indicators,and solve the problem of fault-tolerant control through solving the optimal strategy.Using the performance index function of the neural network approaching the improvement of the improvement of the neural network,you can get the system’s optimal fault tolerance control law,combined with the principle of Lyapunov and the simulation experiments to prove the rationality of the algorithm.(3)Under the circumstances of considering the uncertain environment,multiple faulty reconstructed robotic arm systems add environmental uncertain items to the model,so as to obtain a state space expression with an external restraint system.Combined with adaptive fault observed,the improved cost function is obtained and obtained the equation of Hamilton Jacobi Bellman.Combined with the self-learning optimization algorithm of neural networks,the system has an ambient-constrained anti-saturation fault tolerance control strategy with environmental constraints.The reasonable and effective effects of this algorithm are proved by the robotic arm of different structures.