基于机器视觉的救援机器人自动避障技术研究

Research on Automatic Obstacle Avoidance Technology of Rescue Robot Based on Machine Vision

针对矿山救援机器人易受粉尘、光线昏暗等因素影响,导致其自动避障能力下降问题,研究一种基于机器视觉的救援机器人自动避障策略,可以实现在复杂环境下的有效避障。设计一种基于改进非局部均值滤波和多尺度B样条小波变换的机器视觉算法,以获得更高质量图像并精确获得障碍物边缘,确保救援机器人自主识别并避开障碍物。仿真结果表明:该算法相比传统算法在图像降噪和边缘检测上均体现出优越性。

To cope with the decline of automatic obstacle avoidance ability of mine inspection robots due to their susceptibility to dust,dim light and other factors,studies an automatic obstacle avoidance strategy based on machine vision,which can effectively avoid obstacles in complex environments.A machine vision algorithm based on improved non-local mean filtering and multi-scale B-spline wavelet transform is proposed,by which higher quality images and the edges of obstacles both can be abtained to ensure that the rescue robot can recognize and avoid obstacles independently.Simulation results show that the proposed algorithm is superior to traditional ones in terms of image noise reduction and edge detection.