1、 Design of an Intelligent Car Controller Based on Embedded Platform Liu Gang Zhang Qingyu Wang Yanlin (Key Laboratory of Modern Measurement & Control Technology （ Beijing Information Science & Technology University）， Ministry of Education.) (Beijing Information Science & Technology University Beijin
2、g 100192 China) Abstract: The paper presents a design of an intelligent car controller using embedded ARM7 chips as core component. Modular method has been applied in the design of the hardware； the paper focuses on layout of tracking circuit for the car and design ideas for the software. The experi
3、ment result proves that the designed intelligent car is stable in operation and good in tracking performance. Keywords: ARM Intelligent Tracking 1 、 Introduction In the 21st century, with development of science and technology, researches on intelligent car and its correlative techniques have become
4、the focus in this field. Aiming to enhance practical ability， innovation, and teamwork performance of college students across the country, the Education Department sponsored National Undergraduate Intelligent Car Contest. Based on the background, the paper introduces the design of multifunctional in
5、telligent car controller on embedded platform, including design of hardware circuit and software implementation for key functional modules. 2、 Design of hardware By function and application, the controlling platform for the intelligent car is divided into several modules as shown below. 2.1 Design o
6、f core module AT91SAM7S256 microprocessor produced by ATMEL has been adopted for the controller of the car, which a 32-bit low-power RISC microprocessor chip based on ARM7 TDMI-S core, and embedded with 64KB SARM, 256KB high-speed Flash and JTAG port for downloading or debugging of the program. As t
7、he core component of the car, the microprocessor plays a key role in controlling all running statuses of the car. PWM generating module inside it can be change duty cycle of outputted square wave by programming, and thus change the voltage loaded on the DC motor, which is amplified to control the re
8、volution speed of the motor. The ports PA0, PA1, PA2 and PA3 of the PWM module respectively control the DC motor and the steering motor to perform functions such as go forward, retreat and turn etc. 2.2 Design of power module The car is powered by four AA dry batteries. The voltage is outputted thro
9、ugh low-Noise LDO regulator MIC5209-3.3 to supply power for ARM7 chips and peripheral circuits. The working current of MIC5209-3.3 is as high as 500mA. When input voltage is above 3.5V, the module of MIC5209-3.3 can output stable voltage of 3.3V and achieve low power consumption. 2.3 Temperature det
10、ection Module Mono-line digital temperature sensor DS18B20 is used to detect temperature in the car. The measure range is from -55 to +125 , with increment of 0.5 . It is low in power consumption and small in size, occupying only one I/O port. 2.4 Auto tracking module 2.4.1 Principles for tracking o
11、f intelligent car Tracking means that the car goes along the two-centimeter-wide black guide line on the white floor. Infrared acquisition and camera shooting acquisition are commonly used for it. Infrared acquisition: Taking advantage of the feature that infrared light can change its reflective qua
12、lity according to object surface of various colors. During running the car continuously sends infrared light to the ground, which will, either be reflected back by the white floor and received by the receiving tube in the car, or be absorbed by the black guide line and thus missed by the receiving t
13、ube. By this means the black guide line is positioned to identify the path for the car. Camera shooting acquisition: In certain resolution sample the image by interlaced scanning. When scanning a point, image sensor transfers the gray threshold into corresponding voltage which will be outputted via
14、video signal port. As the car achieves auto tracking by recognizing the black guide line on the track, the image processing is a process of extracting the destination guide line. The task of image processing program is to identify the dots in the black-and-white image and filter noise, record dot po
15、sitions relative to the image, and finally, by algorithm of control strategy, realize tracking and turning of the car along the guide line. 2.4.2 Installation of tracking infrared probe Choosing proper detection method and sensor is the important factor to achieve tracking. Here we choose infrared a
16、cquisition. Correct installation of device is also a decisive factor for accomplishment of tracking circuit. In terms of simplicity, easiness, practicality and reliability, four infrared probes need be installed on the front chassis of the car to fulfill two-staged directional correction control to
17、enhance the reliability of tracking. Four tracking sensors have been fixed, all in one line, among which L1 and R1 are primary sensors for direction control； L2 and R2 are secondary sensors. The distance between the two ipsilateral sensors should nor be more than width of the black direction control
18、. When the car is running, the black guide line is always kept right between the two primary sensors L1 and L2. When the car goes off the black line, the primary sensors detect it, and thus the ARM7 chip detect level jump and execute the pre-prepared correction program to navigate the car back onto
19、the track. The secondary sensors are actually a back-up for the primary. Once the car offsets the track for inertia, beyond the reach of detection of the primary probes, the secondary perform to correct the motion of the car, so as to ensure the reliability of tracking. 3 Design of software The soft
20、ware is developed in C language in Keil Uvision3 IDE, debugged and downloaded in J-Link ARM emulator. J-Link is a JTAG emulator which was brought out by SEGGER in USA to support emulation chips with ARM core. It works with IDEs such as IAREWARM, ADS, Keil, WINARM, and RealView, supports all ARM7/ARM
21、9 core chips simulation, and seamlessly connects with various IDE by RDI interface. Easy to operate and to connect to, it is the most practical tool for study and development of ARM. The key for the design of software lies in the control process of tracking. The sensors are equipped with E3F-DS10C4
22、integrated infrared probes with photoelectric switch. There are only three wires（ power wire, ground wire, and signal wire） at the output pin of the module. Connect the signal wire to I/O port of ARM7 chip, and execute enquiry check. Low level will be detected for the black guide line, while high le
23、vel for the white floor. According to the principles stated above, flows of the algorithm for control tracking of the car. Two-stage control method is adopted to ensure the cars adherence to the black guide line, and the effect is satisdied. 4 Debugging of the finished car Based on the design scheme
24、 presented above, finish making of PCB board for hardware circuit of the car, welding of components, and debugging and downloading of the software. Test the car for several times on the track made of white KT board in the middle of which a two-centimeter-wide black guide line is pasted. The results
25、have showed that, the car runs steadily even at a high speed along straight black guide line. When around the curve, if control the speed properly, the car goes smoothly as well. Two pieces of experience as shown below: (1) E3F-DS10C4 photoelectric sensor should be fixed as close to the ground as po
26、ssible to minimize the interference of environmental light to it. Vertical height of the sensor had better be 58mm. Too far distance from the ground causes weak reflective signal and unstable output of up level signal； too close distance may damage the sensor and intensify the effect of diffuse refl
27、ection. (2) Due to common DC motor adopted for it, the control of the car is not accurate and stable enough to perform a break turn unless several same photoelectric sensors are added to the bottom of the car. 5 Conclusion Intelligent car is a front subject which has synthesized many other subjects
28、and has a widely-applied prospect. It particularly helps to develop the present Chinese undergraduates imagination, practical abilities, team awareness, and hi-tech innovation capacity. References 1 Wu Binghua, Huang Weihua, Cheng Lei among others, Systematic Design of Intelligent Car Based on Route
29、 Identification J. Application of Electronic Technique, 2007(3): 80-83. 2 Wang Chaoyi, Wang Yihuai. Design of Control System of Auto Tracking Car Based on Infrared Sensor J； Computer and Automation Techniques, 2008, 34(11):60-62 3 Li Yi, Lu Ren Yi, & Wu Tian. Intelligence Tracking Car J. Electronic Techniques, 2008, 45(1): 39-41
