운동 성능 요구 조건을 고려한 차량용 전자 제어 장치의 통합 평가 시스템 개발

Abstract

Because of the increasing role that the Electronic Control Unit (ECU) plays in an automobile, various methods has been attempted to verify the reliability of this product. Hardware in the Loop Simulation (HILS) is a method widely used to verify the safety of an ECU-installed vehicle. The behavior of the vehicle can be analyzed by using the vehicle dynamic model in HILS. Although HILS can evaluate the control algorithm in the ECU and predict vehicle behavior, it has a limitation in evaluating the reliability of the ECU requirements. Requirement Based Testing (RBT) is a method used to verify the quality of an embedded system, the representative one being the electronic control unit, which is adopted in vehicles. RBT tests whether or not the embedded system satisfies pre-defined requirements. The RBT software creates the test oracle, which is the virtual ECU, based on the system requirements. The same inputs are provided to both the ECU and test oracle. The quality of the embedded system can be verified by comparing the outputs of the ECU with those of the test oracle. The system requirements of the embedded system generally refer to output signals or actuator responses. If the actuator response affects the behavior of a vehicle, the output of the embedded system may be the behavior of the vehicle. Also, the dynamic performance of the vehicle can be regarded as the system requirement. If the embedded system testing is conducted considering the behavior of the vehicle as a system requirement, the reliability of the ECU and the behavior of the vehicle can be evaluated concurrently. The testing of the embedded system considering the vehicle dynamic performance means the testing is targeted not only to the ECU unit but to the whole vehicle system. HILS, which predicts the behavior of a vehicle, can be used in embedded system testing. This method has the advantage of being able to test vehicle dynamic performance as well as an embedded system under the same environment. This method can be used to evaluate the turning performance of an articulated vehicle adopting All Wheel Steering (AWS) ECU. This vehicle has three axles: The second axle is steered by the front driver angle, and the third axle is steered by the articulation angle. The system requirements of the vehicle are considered swing out, turning radius and off tracking. A simplified mathematical model is developed to predict the behavior of an articulated vehicle. The test uses the mathematical model in the RBT software. In this paper, the entire articulated vehicle, which adopts two AWS ECUs and the mathematical model, is tested.