基于车载摄像机移动机器人的模糊神经网络跟踪

摘要

针对车载单目摄像机的轮式移动机器人的目标跟踪问题，提出一种模糊跟踪控制方案。方案分为两部分，分别为摄像机控制部分和轮式移动机器人控制部分。摄像机控制部分通过相关特征点之间的连线的夹角设计摄像机云台控制器，使摄像机能够实时跟踪目标，保持目标在图像平面的期望位置上。轮式移动机器人控制部分通过设计T-S模糊神经网络控制器使其能够在视觉伺服反馈下实时调整速度和姿态，实现目标的有效跟踪。仿真结果表明了本文方法的有效性。%Aiming at tracking targets by wheeled mobile robots (WMR )with cameras,a fuzzy neural network control scheme was presented.The scheme includes two parts:one is to control the camera and the other is to control the wheeled mobile robots.In the camera control part,a Pan-Tilt controller was designed by using some angles between related lines connecting feature points so that the feature points on the camera image plane can track the target feature points timely and keep them in the desired positions on the image plane.For the tracking control of the wheeled mobile robots,a tracking controller was proposed by using the T-S fuzzy neural network.The configuration and speed of wheeled mobile robots can be adjusted from by virtue of visual servoing feedback to implement an effective target tracking.The simulation results illustrate the effectiveness of the scheme propvided.