Research On New Hydraulic Servo Rotary Drive Joint And Its Control Technology

With the continuous development of robotics intelligence and automation technology,different kinds of robots have been gradually applied in many fields.Among them,hydraulically driven robots have many advantages such as large carrying capacity,strong adaptability and rapid response and can largely replace human beings to complete various complex tasks.As the basic movement component of robots,driving joints directly affect the overall performance of hydraulic robots.Therefore,it is necessary to develop hydraulic joints with a compact structure,large output torque,fast response and precise control.In the study,aiming at the problems and shortcomings in the joints of foot hydraulic robots,a new hydraulic joint was developed for rotary drive.In addition,the necessary structural optimization design,bearing capacity test and precise position control was explored to further improve the new joint.The main contents of the thesis are summarized below.From the perspectives of human anatomy,kinematics and anthropometry,the correlation between the motion characteristics of the lower limb hip joint and the walking cycle was analyzed to provide a theoretical basis for the structural design and corner design of the servo rotary hydraulic joint.The structural components of the servo rotary hydraulic joint were introduced,including the split cylinder,curved piston rod and special seals.The working principle of the designed hydraulic joint was analyzed based on the working characteristics of hydraulic power.The stress conditions of the cylinder and piston rod were analyzed and the necessary strength and stiffness were checked by finite element analysis software.Based on the physical composition and working principle of the valve-controlled rotary hydraulic joint position servo system,the balance equation of the output and load torque of the rotary hydraulic cylinder,the flow continuity equation of the hydraulic power component and the flow equation of the electro-hydraulic servo valve were analyzed.By linearizing the load flow equation of the electro-hydraulic servo valve and approximating the pressures of two chambers of the hydraulic cylinder,the mathematical model of the valve-controlled hydraulic servo rotary joint system was established and the transfer function and state space description of the electro-hydraulic position servo system were proposed.In addition,the electro-hydraulic position servo control system was simulated in AMESim software.The experimental platform was established to test the necessary bandwidth and bearing capacity of the designed joint.Aiming at the hydraulic position servo system with unknown parameters,the sliding mode control method was combined with the adaptive algorithm and fuzzy control theory to design the controller.Through the combination of the advantages of various control modes,the satisfactory control effect was achieved.First,the sliding mode control method based on the reaching law and PID control method were used to track the position of the hydraulic rotary joint under no-load and load conditions.The comparison results showed that the sliding mode control method based on the exponential approach law allowed the good dynamic characteristics and strong robustness.Then,a parameter adaptive estimation law was constructed based on Lyapunov theory to estimate the generalized uncertain parameters and the discontinuous projection algorithm was used to guarantee the boundedness of parameter estimation.Finally,by introducing the fuzzy control theory,the approach control coefficient in the sliding mode control was subjected to fuzzy processing and the control signal was smoothed with the saturation function.The experimental results showed that the proposed adaptive fuzzy sliding mode control method had the better control performance in reducing the tracking error of a target joint.Sliding mode control combined with neural network algorithm was applied in precise position control of the hydraulic position servo system with model uncertainty and high nonlinearity.Firstly,the dynamic equation of the hydraulic position servo control system was described as a new form of the general third-order system and the parameters in the dynamic equation of the system were adjusted with the stability filter.Then,a new high-order neural network estimator was proposed and based on the Lyapunov stability theory,it was proved that the online learning law of the neural network weights could guarantee the stability and convergence of the whole closed-loop system.Then,based on the neural network identification,the sliding mode controller was designed to improve the robustness in the control of the target system.Finally,in order to verify the proposed sliding mode controller based on the adaptive neural network,the control simulation and actual motion control of the hydraulic servo rotary joint position were carried out.The experimental results showed that the control method could quickly identify the structure of the hydraulic position servo system,enhance the adaptability to the controlled system,reduce the system error and improve the position tracking accuracy of the hydraulic servo rotary drive joint.