To improve vehicle handling stability and ride comfort, and enhance the integration of automotive chassis control, this paper conducts integrated control research on the semi-active suspension system (SASS) and electric power steering (EPS) systems in the chassis. An integrated controller for SASS and EPS based on the embedded system ARM S3C44B0X was designed and prototyped, and bench tests were conducted. The results show that the developed integrated controller performs well and can be used for the development and research of automotive chassis integrated control systems. Vehicle chassis control improves the dynamic characteristics of the chassis through electronic control systems, enhancing active safety and driving comfort. This is currently a hot topic in automotive research and represents the future development direction of chassis systems. At present, domestic automotive chassis integration mainly focuses on the integration of anti-lock braking systems (ABS), traction control systems (TCS), anti-slip systems (ASR), and adaptive cruise control (ACC) [1-4]. Reports on the integrated control of vehicle suspension and steering systems are still relatively rare. This paper designs an integrated controller for a semi-active suspension and electric power steering system. An embedded system, SAMSUNG S3C44B0X, is used, employing fuzzy control and PID algorithms to control the SASS and EPS systems respectively. An integrated controller based on the embedded system is successfully fabricated. Bench tests show that the integrated controller achieves good results and meets the design requirements. 1. Integrated Control Scheme Design The suspension system studied is an adjustable damping semi-active suspension (generally referred to as active suspension, ASS), and the steering system is an electric power steering system. Figure 1 shows a schematic diagram of the ASS/EPS integrated control. The active suspension system and electric power steering are considered as a whole, taking into account the coupling of some state variables. An integrated controller is designed to coordinate the control of the semi-active damper stepper motor and the electric power steering DC motor, changing the adjustable damper damping and providing steering assistance to improve vehicle attitude changes during steering and to reconcile the contradiction between handling stability and ride comfort. Based on the integrated model and considering the difficulty of software programming, this paper adopts a fuzzy + PID control strategy, as shown in Figure 2 [5]. The boost voltage U of EPS is controlled by PID to correct the boost, improve the response of yaw rate, and improve steering sensitivity. The fuzzy controller controls the ASS system according to the feedback state variables to improve the vertical acceleration of the center of gravity and the dynamic deflection response of the suspension, and improve the driving smoothness of the vehicle. 2. Hardware design of the control system The hardware design of the controller is shown in Figure 3. It mainly includes the input signal acquisition and conditioning module, the microprocessor interface module, and the output control module for the actuator. When the vehicle is driving normally, the sensor collects the body vertical vibration acceleration, steering shaft torque, vehicle speed and other state signals outside the control system. After conditioning, the signals are transmitted to the ECU of the controller. The ECU analyzes and calculates the signals and generates control signals to transmit to the actuator. The actuator drives the stepper motor of the suspension damper and the DC motor of the steering system according to the control requirements, changes the damping of the damper, and provides steering assistance at the same time, realizing the coordinated integrated control of ASS and EPS. The controller is the core of the integrated system, and the microprocessor MCU is the core of the controller. Considering the speed of the MCU, the integrated resources, the input and output ports and its development environment, this paper selects SAMSUNG's S3C44B0X as the microprocessor chip of the controller. The S3C44B0X microprocessor integrates the ARM7TDMI core [6] on-chip and is manufactured using 0.25um CMOS technology. It integrates rich peripheral function modules on the basis of the basic functions of the ARM7TDMI core, which is convenient for low-cost design of application systems. Before the input signal enters the MCU, it needs to be converted into analog and digital and the signal conditioning such as level matching. The steering wheel signal is provided by the torque sensor. The torque sensor consists of a slider, a steel ball, a ring and a potentiometer. It is used to obtain the magnitude and direction signal of the steering wheel operation force and convert it into a voltage value and transmit it to the AIN0 and AIN1 pins of the MCU. The MCU receives the main and auxiliary symmetrical signals. Only one circuit is needed when sampling. The input signal amplitude is 0 to 5V. The input voltage range of the S3C44B0X A/D converter is 0 to 2.5V. Therefore, filtering and voltage division processing are required, as shown in Figure 4. The sampling filter is a second-order low-pass active circuit. R1 and R2, with equal resistance values, first divide the input signal, halving its amplitude. Then, R1 and C1 form a first-order low-pass filter circuit. R3 and C2 form a second-order first-order low-pass filter. The operational amplifier acts as a voltage follower. The accelerometer, based on the piezoelectric effect, causes crystal deformation due to acceleration, resulting in a change in charge. After amplification and filtering by a charge amplifier, the signal passes through a second-order low-pass active filter circuit (as shown in Figure 4) before entering the analog-to-digital converter (ADC) port. The MCU further calculates based on the difference and rate of change between the vehicle's vertical vibration acceleration and the wheel vibration acceleration. The vehicle speed sensor, located on the gearbox, generates a proportional signal based on the vehicle speed. This signal, derived from the speedometer, is a unipolar pulse signal with a voltage above 9.5V. The ARM processor can only handle signals up to 2.5V, so the vehicle speed signal conditioning mainly involves signal level matching. Optocouplers are used in this design (see Figure 5). The vehicle speed signal DI is converted into a 5V pulse signal by an optocoupler, and after being divided by resistors R2 and R3 of the same resistance value, it is input to the counter of the ARM. The corresponding vehicle speed is calculated by the program. The output control of the actuator includes the control of the EPS DC motor and the damper stepper motor. DC motor control can be divided into two methods: excitation control and armature control. Here, the switching control method is used to drive the power field effect transistor, and the armature voltage is controlled by pulse width modulation (PWM) to achieve speed control. Figure 6 shows a schematic diagram of DC motor control [7]. The ARM ports PE3, PE4, PE5, and PE6 are defined to output DC motor control signals, which are applied to the control terminals of four MOS switches Q1, Q2, Q3, and Q4 respectively through four drive optocouplers. When the motor is required to rotate forward, Q1 is controlled by the PWM signal, and Q4 is turned on by applying a high level, while Q2 and Q3 are turned off by applying a low level; when the motor is required to rotate in reverse, Q3 is controlled by the PWM signal, Q2 is turned on, and Q1 and Q4 are turned off, which facilitates the direction control and speed control of the motor. 3. Software Design The implementation of hardware functions requires software support. The integrated development environment used here is ADS1.2 (ARM Developer Suite), a comprehensive software provided by ARM specifically for the development and debugging of ARM-related applications. Users can use its CodeWarrior IDE to develop, compile, and debug programs written in C, C++, and ARM assembly languages. The integrated control software flow is shown in Figure 8. After initialization, the program enters a loop, waits for an interrupt, and upon response, enters an interrupt subroutine to return the acquired signal. ARM performs PID control and fuzzy processing on this signal, outputting control signals to the actuators and simultaneously returning output signals for feedback comparison. The two key aspects of the program are the implementation of the PID algorithm and the fuzzy algorithm. To facilitate the implementation of PID control in a computer, when the sampled signal is sufficiently small, summation is used instead of integration, and backward difference is used instead of differentiation. The PID control equation is discretized into a difference equation to obtain the digital PID control equation: Where Kp, Ki, and Kd are the proportional coefficient, integral coefficient, and derivative coefficient, respectively, and these three parameters are tuned through simulation. In the fuzzy control subroutine, a fuzzy control rule table is first defined. Then, the difference in acceleration between the vehicle body and wheels and its rate of change are fuzzified. After limiting the saturation value of the universe of discourse, fuzzy inference is performed. Finally, defuzzification is performed to obtain the direction and pulse number of the stepper motor, thereby controlling the motor's steering and step angle. For ease of programming, trigonometric membership functions are used for fuzzification, the most commonly used if-then rule is used for fuzzy rules, and the centroid method is used for defuzzification. 4. Experiment and Result Analysis The hardware and software design was completed. The on-chip bootloader of the ARM development board was started. Using a hyperterminal terminal, the compiled and debugged integrated controller program binary file was downloaded via USB, overwritten, and burned into the ARM flash. A bench test of the controller was conducted, as shown in Figure 9. The power assist characteristics and smoothness of the integrated controller under a certain steering state were tested, as shown in Table 1. Bench tests show that the peak value and standard deviation of the vertical vibration acceleration of the vehicle body, which characterizes ride comfort, are smaller than those obtained by controlling the suspension alone without integration. The motor current response was rapid during the tests, providing noticeable assistance and basically meeting design requirements, thus verifying the ease of steering. The yaw rate, which characterizes handling stability, could not be detected on the bench tests. Currently, the accuracy and reliability of the integrated controller are being tested in real vehicles. References [1] Cui Haifeng, Liu Zhaodu, Wu Lijun, et al. Automobile hill start assist device based on ABS/ASR integrated control system [J]. Agricultural Mechanization Research, 2006, 8: 193-195. [2] Zhang Jingbo, Liu Zhaodu, Qi Zhiquan, et al. Simulation of ABS/ASR/ACC system based on information fusion technology [J]. Computer Simulation, 2004, 21(6): 155-158. [3] Qi Zhiquan, Liu Zhaodu, Shi Kaibin, et al. Research on ABS reference speed determination method based on automobile ABS/ASR/ACC integrated system [J]. Automotive Engineering, 2003, 25(6): 617-620. [4] MA Yue-feng, LIU Zhao-du, QI Zhi-quan, et al. Adaptive Noise Cancellation Method Used for Wheel Speed ​​Signal of Integrate ABS/ASR System [J]. Journal of Beijing Institute of Technology, 2006, 15(2): 144-147. [5] Zhao Junqing. Research on the integrated optimization of structure/control of active suspension and electric power steering in automobiles [D]. Hefei: Hefei University of Technology, 2005. [6] Pan Jijun. Experimental analysis of embedded system based on ARM [J]. Microcomputer Information, 2006, 2-2: 129-132. [7] Chen Kuiyuan, Ma Xiaoping, Ji Xuewu. Research on control technology of electric power steering system [J]. Journal of Jiangsu University: Natural Science Edition, 2004, 25(1): 21-24.