虚拟试验场RGR路面模型重构研究

Reconstruction of RGR pavement model in virtual proving ground

路面模型是虚拟试验场车辆舒适性分析,道路载荷谱分析等必不可少的输入。为了将国内西部某汽车试验场的真实道路转换为路面模型,首先采用PPS激光扫描仪,IMU惯导模块,GNSS基站等设备集成的移动路面扫描系统进行扫描,获得了物理道路的3维扫描点云。然后采用样条曲线拟合道路中心线,并沿中心线构建2维平面栅格点。接着以自然相邻点插值法计算平面栅格点的高程值。最后将3维栅格点云转换为可在多体动力学仿真软件ADAMS中使用的RGR(regular grid road)路面模型。以重构的路面模型为基础,结合FTire轮胎模型,整车多体动力学模型搭建起完整的虚拟试验场仿真系统。为了检验仿真精度,采用带有轮心六分力仪,加速度传感器的路谱采集车进行道路谱载荷采集试验。在大特征破坏路面坑洼路和平稳随机型路面鹅卵石路下,轮心垂向力和轮心垂向加速度等仿真结果与实测数据均吻合良好,表明重构的路面模型及其虚拟试验场具有较好的精度。

Road surface model is an essential input for vehicle comfort analysis and road load spectrum analysis in virtual proving ground.Here,to convert real road of a certain automobile proving ground in western China into a road surface model,firstly,the mobile road surface scanning system integrated with PPS laser scanner,IMU inertial navigation module,GNSS base station and other equipment was used to do road surface scanning,and 3-D scanning point cloud of the physical road was obtained.Then,spline curve was used to fit road centerline,and 2-D planar grid points were constructed along the centerline.Furthermore,the natural adjacent point interpolation methodwas used to calculate high rang values of planar grid points.Finally,the 3-D scanning point cloud was converted into the regular grid road(RGR)model able to be used in the multi-body dynamic simulation software ADAMS.Based on the reconstructed RGR pavement model,combining with the tire model FTire and the whole vehicle’s multi-body dynamic model,a complete virtual proving ground simulation system was built.In order to check simulation accuracy,a road spectrum collecting vehicle with a wheel center six-component force-meter and acceleration sensorswas used to conduct road spectrum load collection tests.It was shown thatthe simulation results of wheel center vertical force and wheel center vertical acceleration agree well with the measured data under conditions of pothole road with large feature damage pavement and cobblestone road with stationary random pavement,so the reconstructed RGR pavement model and its virtual test field have better accuracy.