Research On Remote Condition Monitoring And Fault Diagnosis System Of Quadruped Robot Joints

Quadruped robot has the characteristics of fast moving speed and strong load capacity,which can be used as a mobile platform for home and business services,scientific detection,emergency search and rescue,material transportation,reconnaissance and patrol operations.Walking speed,flexibility and stability are the main core mobile performance of quadruped robot application.Effective state monitoring and fault diagnosis of quadruped robot is a major research focus to improve the core mobile performance of quadruped robot,and the state monitoring and fault diagnosis of quadruped robot joints is particularly critical.At present,the remote state monitoring and fault diagnosis of quadruped robot joints still need further in-depth and systematic research.Therefore,this paper focuses on the quadruped robot joints and systematically studies the remote state monitoring and fault diagnosis system of joints.The specific contents are as follows:Aiming at the fault problem of quadruped robot’s leg joints,a state signal acquisition module of quadruped robot’s leg joints based on ECDS is designed.The interference source and preprocessing technology of joint state signal of quadruped robot are analyzed.Based on ECDS,a joint state monitoring unit of quadruped robot is designed,and a joint state signal decomposition method based on C-EEMD is proposed.The research results show that C-EEMD has better performance of modal spectrum separation and calculation,and the required number of screening iterations is far lower than EEMD.A fault diagnosis method for quadruped robot based on improved C-EEMD entropy feature and PSO-SVM technology is proposed.According to the characteristics of knee joint state information data of quadruped robot when lifting and dropping legs,the parameter optimization method in the calculation process of energy entropy,singular entropy,arrangement entropy and approximate entropy is studied.Combined with PSO-SVM technology,the fault diagnosis and feature extraction methods of knee joint state of quadruped robot when lifting and dropping legs are studied.The research results show that the improved C-EEMD approximate entropy of the quadruped robot knee joint vibration data has the best classification ability,the energy entropy and the arrangement entropy also have more than90% of the recognition ability,and the singular entropy effect is the last,with each recognition degree less than 60%.The safety research on the quadruped robot joint state monitoring and fault diagnosis system is carried out,and the relevant functions of the quadruped robot joint remote state monitoring and fault diagnosis system designed in this paper are verified by using the master and foreign key constraints to control the data flow among different types of users and in the laboratory environment.The test results show that the system can meet the design requirements of state monitoring and fault diagnosis of quadruped robot joints,and is competent for fault classification and remote information query of typical fault signals of quadruped robot joints.Through the above research,this research will provide an important theoretical basis and technical support for the improvement of quadruped robot’s state monitoring and its motion control performance,and will also have a positive impact on the improvement of quadruped robot’s motion performance and its promotion and application.