When robots are able to see and respond to their surroundings, a whole new world of possibilities opens up. To bring these possibilities to life, the robotics industry is increasingly adopting camera-based vision systems, especially when a robotic system needs to interact with a dynamic environment or moving target. However, this kind of vision system is known to have low data transmission rates, packet loss during communication and noisy measurements as major disadvantages. These problems can perturb the control performance and the quality of the robot-environment interaction. To improve the quality of visual information, in this paper, we propose to model the dynamics of the motion of a target object and use this model to implement an Extended Kalman Filter based on Intermittent Observations of the vision system. The effectiveness of the proposed approach was tested through experiments with a robotic arm, a camera device in an eye-to-hand configuration, and an oscillating suspended block as a target to follow.

翻译：当机器人能够感知并响应周围环境时，全新的应用可能性便随之开启。为实现这些可能性，机器人行业正越来越多地采用基于摄像头的视觉系统，特别是当机器人系统需要与动态环境或移动目标交互时。然而，此类视觉系统存在数据传输速率低、通信过程中数据包丢失以及测量噪声大等主要缺陷。这些问题会干扰控制性能及机器人-环境交互质量。为提升视觉信息质量，本文提出对目标物体的运动动力学进行建模，并基于视觉系统的间歇观测实现扩展卡尔曼滤波器。通过使用机械臂、眼对手构型的摄像头装置以及作为跟随目标的振荡悬挂块进行实验，验证了所提方法的有效性。