摘要
本文利用相机对目标物体特征点的测量信息解决机器人相对位置和姿态的估计问题。首先,针对机器人运动过程中相机获得的目标物体连续图像,利用扩展Kalman滤波器估计特征点的三维信息。然后,在SE(3)空间上设计了一个非线性位姿观测器。利用得到的特征点三维信息构造了Lyapunov函数,并解耦成分别和姿态误差以及位置误差相关的两部分。利用Lyapunov稳定性原理设计了观测器的姿态误差和位置误差补偿律,证明了观测器在平衡点处是渐近稳定的。最后,通过数值仿真实验验证了所提算法的有效性。
This paper studies the estimation problem of the relative position and attitude for robots based on measured feature points of the target object from the camera.Firstly,an extended Kalman filtering is used to estimate the 3-D information of the feature points in the continuous image of the target object obtained by the camera.Secondly,a nonlinear pose observer is designed on the SE(3),and Lyapunov function is constructed by the estimated 3-D information of the feature points,which can be decomposed into two parts related to attitude error and position error.Moreover,the attitude error and position error compensation laws of the observer are designed by using the Lyapunov stability theory,and it is proved that the observer is asymptotically stable at the equilibrium point.Finally,the effectiveness of the proposed algorithm is verified by simulation experiment.
作者
滕游
刘安东
俞立
Teng You;Liu Andong;Yu Li(College of Information Engineering,Zhejiang University of Technology,Hangzhou 310023)
出处
《高技术通讯》
CAS
2021年第6期639-645,共7页
Chinese High Technology Letters
基金
国家自然科学基金-浙江两化融合联合基金(U1709213)项目资助。
