微型电脑应用

巡检机器人在实际应用中受到各种干扰因素的影响，导致障碍物定位难度较大，且定位精度和适应度不高。因此，提出了一种基于双目立体匹配算法的智能巡检机器人障碍物定位方法。通过优化缺陷参量，使用双目立体匹配算法计算障碍物的卷积加权特征，通过视差回归优化实现障碍物的识别和测距。实验结果表明，所提方法取得良好的效果，且边界轮廓的生成速度得到了显著提升，能够有效地识别障碍物边缘，提高了定位精度和速度。

In practical application, the inspection robot is affected by various interference factors, which makes difficult to locate obstacles, and the positioning accuracy and fitness are not high. Therefore, an intelligent inspection robot obstacle location method based on binocular stereo matching algorithm is proposed. By optimizing defect parameters and using binocular stereo matching algorithm to calculate convolutional weighted features of obstacles, obstacle recognition and ranging are achieved through disparity regression optimization. The experimental results show that the proposed method achieves good results, and the generation speed of boundary contours is significantly improved, which can effectively recognize the edges of obstacles and improve positioning accuracy and speed.