Control Algorithm And Experimental Platform Of Space Robot Modular Joint

The Robot modular joint has always been interested in robot research, as a key component of the robot, its performance will have a direct impact on the performance of robots. Modular joint is a multi-variable, strong coupling, nonlinear and time-varying system. Robot joint has different load in different position, with friction and other nonlinear-interference factors, if using a group of fixed PID parameters on the implementation of robot controlling, it's difficult to achieve a good dynamic and static performance.Firstly, based on the research work that has been done home and abroad on the modular joint, this thesis establishes a Matlab model of the modular joint based on mathematical theory. Use fuzzy adaptive PID control method to track the error and error rate change of the input and use fuzzy ratiocination to achieve the PID parameters self-optimization. Simulation and experimental results show that this strategy can effectively improve the system performance, the system response fast, overshoot small, robust and adaptive.Secondly, this thesis design and implement a joint performance test platform based on virtual instrument. The joint's servo control performance has been tested and demarcated. Many experiments have been done on improving speed and position performance when the joint under different load. The experimental results have shown that high speed performance is satisfactory while speed performance is poor in the low region. Ripple of speed is serious in low speed.Low speed control performance has much more influence on the precision of servo systems. To improve the speed response in low speed region is very useful to modular joint's servo systems. So Kalman Filter is applied to low speed control of modular joint servo systems in order to identify the information of low speed instead of incremental encoder. In high speed region, encoder is still used to acquire speed information. The simulation and experimental of modular joint servo system with Kalman Filter has been carried out under Matlab and test platform. The results have shown that detecting low speed with Kalman Filter has good static and dynamic performance. Especially in low speed, this method has good position tracking performance and speed tracking performance. So with this method, the static and dynamic performance can be guaranteed in the whole speed range.