光学与光电技术, 2019, 17 (5): 30, 网络出版: 2019-11-14  

一种用于机器人避障的深度相机姿态自标定方法

Depth Camera Pose Self-Calibration Method for Robot Obstacle Avoidance
作者单位
贵州电网有限责任公司电力调度控制中心, 贵州 贵阳 550002
摘要
针对巡检机器人深度视觉系统需要频繁人工标定问题, 提出了一种用于机器人避障与防跌落的深度相机姿态自标定方法。该方法结合相机坐标系与世界坐标系相对关系, 利用平整地面拟合出参数化平面, 在此基础上建立世界坐标系, 进而求解出表示世界坐标系与相机坐标系关系的参数。该方法能够解决传统算法对深度相机标定需要人工标定, 且需要在场景中放置标志信息等操作复杂的问题, 提高了标定效率。实验数据表明提出的方法能准确标定出深度相机姿态参数, 满足了障碍检测与机器人防跌落需求。
Abstract
Aiming at the frequent manual calibration of the patrol robot depth vision system, a depth camera pose self-calibration method for robot obstacle avoidance and fall prevention is proposed. The method comprehensively considers the relationship between the camera coordinate system and the world coordinate system, and uses the flat ground to fit the parameterized plane. On this basis, the world coordinate system is established, and the relationship between the world coordinate system and the camera coordinate system is solved. It effectively solves the problem that the traditional algorithm requires manual calibration for depth camera calibration, and requires complicated operation such as placing marker information in the scene, by improving the calibration efficiency. The experimental data show that the proposed method can accurately calibrate the pose parameters of the depth camera, and meets the obstacle detection and robot anti-drop requirements.

陈锦龙, 杜江, 安成, 肖倩宏, 宋弦. 一种用于机器人避障的深度相机姿态自标定方法[J]. 光学与光电技术, 2019, 17(5): 30. CHEN Jin-long, DU Jiang, AN Cheng, XIAO Qian-hong, SONG Xian. Depth Camera Pose Self-Calibration Method for Robot Obstacle Avoidance[J]. OPTICS & OPTOELECTRONIC TECHNOLOGY, 2019, 17(5): 30.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!