Research On Highly Integrated Joint Controller Based On EtherCAT Fieldbus

Light weight robotic arms are designed especially for mobility and interaction with humans and priori unknown environments,these robots have broad application and development pospects is the fields of service robots,flexible production lines and humanoid mobile platforms.Compared with traditional industrial robots,the design of light weight manipulators is oriented to a high degree of flexibility and maneuverability.Therefore,the structure of light weight manipulators is more compact,with higher load-to-weight ratio,lighter weight,and lower power consumption.The characteristics of light-weight manipulators require that their joint controllers meet high integration and high efficiency,but also need to have more powerful computing power to be able to handle complex control algorithms.The working environment of light-weight manipulators is usually unknown,and sometimes it is necessary to complete the interaction with humans,which means more complex control strategies.In order to ensure the control performance of the system,it is often required that the system can achieve a shorter control cycle.The communication bandwidth,real-time and reliability of the arm control system put forward extremely high requirements.Therefore,the highly integrated joint controller is of great significance to the design of light-weight manipulators.Aiming at the characteristics of light weight robotic arms,this paper designs and develops a highly integrated joint controller based on Ether CAT fieldbus.The joint controller uses FPGA as the core logic control device.Based on SOPC technology,the NIOS II soft core processor is embedded in the FPGA.It uses hardware to implement high-real-time tasks such as sensor data acquisition and processing,and PMSM vector control.The Ether CAT slave stack code runs on the soft core processor in the form of software.This way of arrangement ensures the integration,and can realize the high-performance and easy-to-maintain joint controller design at the same time.Secondly,in order to further improve the integration of the light weight robotic arm,the joint position acquisition and the motor rotor position acquisition are integrated on the same PCB board.This design scheme uses the MEMS magnetic encoder chip to collect the raw position datas of the master track and the nonius track of the magnetic code disk.The absolute position is calculated using the vernier principle.Due to the installation error of the encoder and magnetic code disk,it will cause errors in the absolute position calculation when calculating the magnetic pole where the current position is located.To solve this problem,a calibration algorithm using interpolation method is proposed.Finally,the FPGA-based hardware current loop for PMSM is implemented.The functional module in the current loop is developed using Verilog HDL,and their correctness is verified by simulation and experiment.The current loop PI parameters are designed using internal model control strategy.By using this strategy,previously four PI controller parameters are simplified into two parameters which indicate the designed current loop bandwidth,and the maximum bandwidth of the current loop is limited by the delays of control algorithm.Finally,the correctness and performance of the current loop is verified by the simulation using Matlab/Simulink and experiments.