Research And Design Of Space Motion Tracking System For Robot Teaching

Robot teaching is one of the main ways to realize the robot's track reproduction.There are many shortcomings in the traditional teaching methods including complex process,high operational requirement,low efficiency and security risks and so on.The use of human action capture technology can make up for the traditional teaching methods of the defects,simplify the teaching process and improve the efficiency to ensure the safety of teaching process.This paper uses low-cost MEMS inertial sensor for human arm motion capture technology.The nine-axis inertial sensor is used to measure the relevant parameters in the process of arm movement.The improved algorithm is designed based on the study of SINS theory,and the validity of algorithm is verified in simulation.The position and trajectory of the end of the arm in space are obtained by using the spatial position algorithm,and the effect of robot teaching using motion capture technology is tested by the motion experiment.The result shows that the use of MEMS inertial sensors for human arm motion capture can effectively reduce costs and ensure a certain degree of capture accuracy.It can be used for robot teaching.Due to the poor design and low measurement accuracy of inertial sensors,Firstly,this paper calibrates the initial error of the accelerometer,magnetometer and gyroscope included in the nine-axis inertial sensor.The error model is established by analyzing its working principle,and the error model parameters are calculated by different calibration methods according to their working characteristics to improve the accuracy of the initial data in the practical application.After that,this paper introduces the direction cosine method,Euler angle method and quaternion method commonly used in attitude calculation and complementary filter and kalman filter used in data fusion,analyzes and compares their advantages and disadvantages.Aiming at the shortcomings including low stability and lack of accuracy of the traditional attitude calculation method,a posture solution method based on combined sensor and fused filter is designed.Two sets of MEMS inertial sensors were used in measurement.The Euler angle method and the quaternion method are combined.The complementary filter and Kalman filter are combined.The fastness of fuzzy adjustment and the accuracy of PI adjustment can effectively suppress the error and drift of attitude angle in order to achieve the ideal effect.And then the spatial position calculation method and the spatial position superposition algorithm are used to obtain the trajectory of the arm end in space.In the experiment including horizontal movement of the lower arm,arm reciprocating motion and the complex motion test,the wrist movement trajectory in the space is obtained.Experiments show that the human body motion capture technique using MEMS inertial sensor can correctly express the trajectory of the arm end,and the improved attitude calculation method can reduce the error of the trajectory.It can be used in robot teaching.