A Study On Dsp-based Multi-axis Motion Controller

As the core of open NC system, motion controller is very important and attracts increasing attention. It has been widely applied to mobile robot, aerospace, medical equipment, test and measurement, semiconductor manufacturing and material processing etc. Our country has some distance compared with America, Europe, Japan and other developed countries in the field of motion controller, so it is of practical significance to develop an open motion controller. Combining with the motion control requirements of the 2-DOF mobile robot, this thesis made an exploration on DSP-based multi-axis motion controller. This thesis organized as follows:On the basis of the architecture of mainstream motion controller at the market, an overall scheme of DSP-based 2-axis motion controller is proposed. The detail about how to design the hardware and the software of motion controller is discussed. The hardware of motion controller includes DSP hardware circuit and Complex Programmable Logic Device (CPLD). TMS320LF2407A from TI was the core of the control system, while CPLD EMP7128, which belongs to the MAX series product of ALTERA, was used to extend the function of motion controller. The circuit diagram of each module of hardware is provided. The software of motion controller includes initializing program of DSP system and control algorithm. The DSP initializing functions and the flow chart of control algorithm are provided. In the end, the motion controller designed in this study has been applied to a 2-DOF mobile robot MTR.A FPGA-based 4-axis motion controller is designed for the applications of more DOF mobile robot, robot manipulator, etc. System structure of the 4-axis motion controller and the software and hardware solution are given. Each logic module of FPGA with Quartus II platform is analyzed and simulated. This motion controller adopts TMS320F2812 from TI and FPGA EP2C5Q, which belongs to the Cyclone II series product of ALTERA, as the core of control system. The digital signal processing capability of DSP and logic processing power of FPGA are brought into full play in this structure.A compensating control strategy based on Fuzzy Cerebellar Model Articulation Controller (FCMAC) is proposed, for it might be impossible for PID controller to achieve a high precision control. Combining the compensating controller based on FCMAC with model reference control, the proposed control strategy can compensate parameter variation and disturbance in real time. The simulation results have shown that the proposed strategy could effectively reduce the tracking error of the velocity and displacement of the system, even in the extreme case that the system is affected by persistent sinusoidal disturbance.