基于视觉检测技术的轮毂位置精确测定

Accurate determination of wheel hub position based on visual inspection technology

摘要:

针对工业技术智能化的发展趋势，及电动车轻质合金轮毂机械加工艺，对装卡位置精度要求，本文研究一种有效的轮毂装卡位置精准测定方法，以解决当前检测方法，易受人为因素影响、效率低、稳定性差等问题。提出一种基于机器视觉技术的非接触式轮毂装卡位置精确测定方法，应用CCD工业相机搭建视觉检测系统，运用双特征位置精确测定方法，提取相机摄入图像中轮毂气门孔和中心孔的两处主要特征，通过对双特征数据的分析与计算确定轮毂位置精度误差，并根据计算数据输出装卡位置误差，应用分段位移、逐渐次定位的方式进行反馈补偿。通过实验测量表明，在系统的测量参量值中，角度标准差为0.0226°，测量不确定度为0.0036°，测量精度理想，平均检测时间为0.29s，检测时效性良好，能够满足轮毂自动化加工位置检测的性能要求。

Abstract:

In view of the development trend of intelligent industrial technology, and the requirements of light alloy wheel hub machining process for the accuracy of clamping position, this paper studies an effective accurate determination method of wheel hub clamping position, in order to solve the problems of current detection methods, such as easy to be affected by human factors, low efficiency, poor stability and so on. This paper proposes a non-contact accurate measurement method of wheel hub mounting position based on machine vision technology. A vision detection system is built by using CCD industrial camera. Two main features of wheel hub valve hole and center hole are extracted by using the accurate measurement method of double feature position. The accuracy error of wheel hub position is determined by analyzing and calculating the double feature data, According to the calculated data, the position error of the card is output, and the feedback compensation is carried out by means of subsection displacement and gradual positioning. The experimental results show that the angle standard deviation is 0.0226 ° and the measurement uncertainty is 0.0036 ° in the measurement parameters of the system, the measurement accuracy is ideal, the average detection time is 0.29s, and the detection timeliness is good, which can meet the performance requirements of wheel hub automatic machining position detection.