A high precision servo controller is designed based on the chip of TMS320F28335. The hardware composition and working principle of the photoelectric tracking servo system are described. The hardware design of the servo controller is emphatically introduced, and the schematic diagram of the control circuit is given. The difficulty of tracking control of the system is discussed and the software solution is given. Practice shows that the system has the characteristics of good stability, high precision, fast responsiveness and easy debugging, and achieves the expected goal, and has practical value. Servo system, also known as servo system, is an important part of photoelectric tracking system. The working principle of the photoelectric tracking servo system is that when the TV or infrared camera recognizes the moving target through image processing and transmits the angle deviation data between the target and the center point in the field of view to the servo system, the servo controller processes the data, and the servo driver drives the photoelectric load through power conversion control to make the servo system work. The optical axis of TV or infrared camera always aims at the target, and synchronizes with the target to achieve automatic tracking. The accuracy, stability and fast responsiveness of servo system are important indexes of dynamic performance and measurement accuracy of photoelectric tracking system.
The difficulty of tracking servo system design is how to achieve high accuracy, high stability and good fast response. Due to the resource limitation on chip, thermostatic element the microprocessor microcontroller which is commonly used can hardly meet the performance requirements of the servo system in terms of high operation rate and high real-time performance of data processing. In the automatic control system, high-performance DSP has become the mainstream control chip, gradually replacing the single chip. This design adopts the DSP chip TMS320F28335 produced by Texas Instruments Company of the United States. The 32-bit floating-point DSP chip has a high-speed processing capacity of up to 150 MHz.
Its single instruction execution time is only about 6.67 ns. The chip integrates a large number of related circuit resources of motor control. Compared with the previous DSP chip, the 32-bit floating-point DSP chip has more high-speed digital signal processing capability and can complete complex control algorithms in real time. It integrates abundant motor control peripheral components and circuits, simplifies the hardware design of the control circuit, and is a DSP chip specially designed for motor control. The servo system is mainly composed of controller, driver, servo motor set, photoelectric encoder, power supply and control protection circuit, etc. [1].
The system composition block diagram is shown in Figure 1. The three-loop cascade control scheme of position loop, speed switch loop and current loop is adopted in the design of servo system loop. The principle block diagram of the system loop is shown in Fig. 2. The servo controller is mainly responsible for the closed-loop adjustment of the position loop, and the servo driver is mainly responsible for the closed-loop adjustment of the current loop and the speed loop. The three-loop closed-loop system can better ensure the high accuracy, high stability and fast response of the tracking servo system. The system adopts modular design and special control chip, which greatly reduces the weight and volume of the tracking servo system, and improves the accuracy and stability of the servo system. Servo motor adopts brush DC torque motor and sleeve shaft. The brush DC torque motor has the advantages of low speed, large torque, small torque fluctuation, large output torque and good linearity [2]. According to the requirement of system accuracy, the system uses 16-bit photoelectric encoder as position detector, and adopts coaxial installation to reduce transmission error. The encoder uses RS422 serial port of 115.2 Kbps baud rate to communicate with SCI inside the DSP. Its angular resolution is 360/216=0.00549. The controller is composed of the core circuit of special control microprocessor TMS320F28335, A/D conversion circuit, level conversion circuit, D/A output circuit, fault signal detection circuit, protection circuit and other peripheral circuits. The structure of the controller is shown in Fig. 3. Its characteristics are that the input signals are digitally sampled, which is convenient for software tuning parameters; complex control algorithm can be realized based on 32-bit floating-point processing unit; reusability and versatility of the software are greatly improved by using modular design. Its main function is to receive the control commands of the host computer, realize the switch of various working conditions and return the fault self-check data of the tracking servo system. CAN bus is a serial data communication protocol developed by German Bosch Company in the early 1980s to solve the data exchange between many control and test instruments in automobiles.
It is a multi-bus with a communication rate of up to 1 Mbps. N2.0B standard [5].
The eCAN module has 32 mailboxes, which can be flexibly configured to receive or send mailboxes according to the design requirements. A single frame can transfer 8 bytes of data. This design uses SN65HVD230 interface chip as CAN transceiver. The data rate is set to 500K and 11-bit identification code is used. The system has the function of manual assistant tracking.
It can control the rotation of azimuth and pitch axis to track the target through an analog handle. The analog voltage is converted to digital signal by A/D conversion circuit. Because of the inconsistency of bus level, it is necessary to convert the level of SN74LVC4245A to be used by DSP. Its voltage input range is ( 10V), which can meet the system speed regulation requirements. The A/D conversion circuit is shown in Figure 4. The speed of the servo system required by the system can be adjusted from 0.05 degree/s to 50 degree/s, and the speed ratio of the system is 50/0.05=1000. The A D7836 is a 14-bit D/A converter chip. Its output speed ratio is 213=8192, which can meet the requirements of the speed range of the system.
The chip integrates four 14-bit DACs, and only two of them are used in this design. Using 5V reference voltage, the full range output voltage range is 10V, which is convenient to connect with the servo driver. The D/A conversion circuit is shown in Fig.
5. Servo control software uses CCS 3.3 integrated development environment and C language. The software adopts modular design method to facilitate debugging. The system software flow is shown in Figure 6. Its working process is: first, the system initializes the DSP, then the system self-checks, if the self-checks are abnormal, the system fault processing, and then re-checks. Up to the self-checking fault-free signal, the system receives the instructions from the host computer and enters the corresponding control mode to collect and process data, output data and communicate with the serial port. This design adopts manual tracking and automatic tracking. Manual tracking is that the operator uses the handle to control the servo rotation for preliminary tracking until the target enters the field of view of TV or infrared camera. After the optical sensor catches the target, it converts the miss distance data from the real-time output of the image differentiator into automatic tracking. During the switching process, the optical axis of the optical sensor should always aim at the target. The sketch of maneuvering target tracking in the air is shown in Fig. 7. Because of the small field of view of the optical sensor mounted on the servo system, the dynamic tracking accuracy of the servo system is required to be less than 0.5 bits, so the difficulty of the system design is to track the fast response of switching, small overshoot and high precision of stable tracking. In the process of target acquisition, if there is no effective control strategy, it will cause the problem that the target is easy to lose its tracking state when switching to the automatic tracking state at the edge of the field of view. To solve this problem, we design a dynamic PID tracking control method based on the basic principle of automatic control. The dynamic PID tracking controller is shown in Figure 8.
It is a two-dimensional fuzzy control model with deviation E and the ratio of deviation to time e as input. According to e and e of constant frequency input and dynamic PID tracking control algorithm, the values of three important parameters Kp, Ki and Kd of the PID controller are calculated dynamically and output to the PID controller for output adjustment. This dynamic PID algorithm can solve the synchronization problem of servo tracking acceleration and optoelectronic imaging processor very well. The test results show that the automatic tracking accuracy is less than 0.
5 bit, the maximum azimuth acceleration is 110 degree/s2, and the manual tracking speed is adjustable from 0.05 degree/s to 50 degree/s. The technical indicators meet the design requirements. This paper describes in detail the design of servo controller based on Texas Instrument’s DSP chip TMS320F28335 as the core processor. The model products with this design have been produced in batches, and the tracking effect is good. Practice shows that the system has the characteristics of good stability, high accuracy, fast responsiveness and easy debugging, and achieves the expected goal, and has strong engineering practical value.