论文部分内容阅读
在实际的应用中，由于室外移动机器人面临着复杂的环境，道路中充满阴影、水迹等环境噪声．这些噪声严重地损害了以往的各种视觉导航算法的鲁棒性．提出了一种新的面向室外移动机器人导航的阴影消除算法．详述了使用模糊神经网在低分辨率图像上对阴影进行识别，利用遗传算法进行网络结构优化，最后消除原始图像中阴影的方法和过程．同时给出了利用该算法在ＴＨＭＲＩＩ室外移动机器人上进行的实验结果．
In practical applications, due to the complex environment that outdoor mobile robots face, roads are full of environmental noise such as shadows and watermarks. These noises seriously impair the robustness of various visual navigation algorithms in the past. A new shadow removal algorithm for outdoor mobile robot navigation is proposed. The method and process of using fuzzy neural network to recognize shadows on low resolution image, using genetic algorithm to optimize the network structure and finally eliminate the shadow in the original image are described in detail. At the same time, the experimental results on THMRII outdoor mobile robot are given.