基于视觉辅助的汽车转向路感优化控制仿真

Simulation of automotive Steering Road Optimization Control Based on the Visual AIDS Sense

汽车行驶速度过快的情况下,与路面的冲击反力会瞬间增大。路感控制过程需要迅速改变硬件结构参数以保持稳定,过大变化使得控制过程出现异动,导致汽车转向路感系统出现"打手"(过控制)问题,传统的路感系统控制方法针对过控制问题,需要动态变化控制参数,使其与过控制过程形成匹配,但是传统方法很容易引起模型不稳,导致对汽车转向路感系统的控制误差增大。提出视觉辅助的汽车转向路感系统快速控制方法,采用Harris角点控制算法获取监控视频中每幅汽车转向路感图像中的角点位置,对每帧汽车转向图像的角点进行LK光流场运算,依据运动目标的光流连续性获取可能的目标角点信息,通过聚类分析对若干角点位置集进行数字运动汽车转向图像处理中的形态学处理,得到汽车当前所处的实际位置,实现汽车转向路感系统的快速控制。实验结果证明,改进种方法的可行性,并且具有较高的控制精度。

A fast control method based on visual AIDS car steering road feel system is presented, steering Angle point for each vehicle in the road feeling image location of ovideo monitoring is obtained based on the Harris corner control algorithm, LK optical flow field for vehicle steering Angle points of per frame image is calculated, possible target Angle information is obtained based on the light flow continuity of moving targets, the moving car steering image is processed in morphology through clustering analysis to digital angular point location sets, the current actual position is obtained, and the fast control of automobile steering road sense system is realized. Experimental results prove that the proposed method is feasible, and has high control precision.