摘要
为了让移动机器人在动态变化的环境下完成镇定任务,提出了一种基于视觉伺服、可在动态场景中利用模型未知的特征点完成轮式移动机器人的视觉镇定控制方法。首先利用特征点和视觉图像的几何关系计算出不同场景深度的比例;然后,通过单应矩阵分解和坐标系变换,求出各个坐标系之间的相对关系;最后,采用自适应控制器将机器人驱动到期望位姿处。仿真和实验结果表明:在场景发生变动时,移动机器人可以完成镇定控制,即当线速度v→0和角速度w→0时,误差e→0,机器人镇定到期望位姿0.0 m,0.0 m,0.0°处。
A new method is proposed for mobile robot visual servoing using model-free visual targets under dynamic scenes. In the first stage, the ratio of depth information is calculated by using the geometric relationship between feature points and images. For the second stage, the relationship between various coordinate systems can be calculated via homography decompositions and coordinate transformation. Finally, an adaptive controller is adopted to drive the mobile robot to the desired pose. Both the simulation and experimental results show that when the scene changes, the mobile robot can complete the stabilization control. That is, when the linear velocity v→0 and the angular velocity w→0, the error e→0, then the robot stabilization to the desired pose 0.0 m, 0.0 m, 0.0 °.
作者
李宝全
尹成浩
师五喜
LI Bao-quan;YIN Cheng-hao;SHI Wu-xi(School of Electrical Engineering and Automation,Tianjin Polytechnic University,Tianjin 300387,China)
出处
《天津工业大学学报》
CAS
北大核心
2019年第3期47-54,共8页
Journal of Tiangong University
基金
国家自然科学基金资助项目(61603271)
天津市自然科学基金资助项目(15JCYBJC47800,16JCQNJC03800)
关键词
移动机器人
动态场景
视觉镇定
动态特征点
单应矩阵
视觉伺服
自适应控制律
mobile robot
dynamic scene
visual stabilization
dynamic feature points
projection homography matrix
visual servoing
adaptive control law
