Luoyang Institute of Technology, School of Electrical Engineering and Automation
针对未标定手眼机器人视觉伺服控制问题，提出了基于图像轨迹规划与时序控制的视觉伺服方法。首先给出了图像平面和任务空间的轨迹规划方法，其次设计了卡尔曼滤波器估算目标与相机相对位姿，结合相机测量实现了雅可比矩阵的自适应估算。在此基础上设计了基于图像视觉反馈的视觉伺服控制器，实现了机器人的目标视觉跟踪。文末通过MATLAB仿真和实验研究对该算法进行了验证，并与经典PBVS、IBVS方法进行了比较试验，仿真及实验结果验证了该算法的有效性。
A visual servo method based on image trajectory planning and sequential control was proposed for uncalibrated eye-in-hand robot system. First the trajectory planning method in image space and task space was presented. Second the Kalman filter was adopted to estimate the relative pose of the target to camera, and combined with camera measurement, the adaption of Jacobian matrix was realized. A visual servo controller with image feedback was designed to realize the visual tracking of target. At the end of this paper, the algorithm was verified by MATLAB simulations and experimental researches, and compared with classical PBVS and IBVS method. The results verify the effectiveness of the algorithm.
