论文部分内容阅读
目前对汽车附加转向角的研究主要基于观测的侧向加速度信息对其进行估计,观测精度在极限工况下不能满足汽车稳定性实时控制需要。因此,采用试验分析和建模的方法展开研究,首先基于ADAMS/Car软件建立某乘用车多体动力学整车模型,通过实车稳态回转试验数据验证了模型的精确性。基于该平台量化分析不同工况下附加转向角的影响因素。采用ε-SVR算法建立附加转向角预测模型,输入矢量主要为稳定性控制系统(Stability control system,ESC)配置传感器信息,采用典型蛇形试验和FMVSS 126试验数据分别对模型进行学习和泛化性能测试。测试结果表明,汽车侧向加速度是影响汽车附加转向角最主要因素(数量级达1°),其次是汽车纵向加减速度和路面幅值(数量级达0.3°);建立模型的预测精度和计算实时性均能满足ESC实时控制要求。拓展了极限工况下汽车附加转向角精确观测方法。
At present, the research on the additional steering angle of the vehicle is mainly based on the observed lateral acceleration information, which can not meet the real-time control requirements of the vehicle stability under the limit conditions. Therefore, the research on the method of experimental analysis and modeling is carried out. Firstly, a multi-body dynamic model of a passenger car is established based on ADAMS / Car software. The accuracy of the model is verified by steady-state rotating test data of real car. Based on the platform, the influencing factors of additional steering angle under different working conditions were quantitatively analyzed. The ε-SVR algorithm was used to establish the additional steering angle prediction model. The input vector was mainly used to configure the sensor information for the stability control system (ESC). The typical snake-shaped test and the FMVSS 126 test data were used to study and generalize the performance of the model test. The test results show that the vehicle lateral acceleration is the most important factor that affects the vehicle additional steering angle (1 degree in magnitude), followed by the vehicle longitudinal acceleration and deceleration and the road surface amplitude (up to 0.3 ° in magnitude). The prediction accuracy of the model and the calculation of real-time Sex can meet the ESC real-time control requirements. This paper expands the accurate observation method of vehicle additional steering angle under the limit working condition.