Unsupervised Learning of Terrain Appearance for Automated Coral Reef Exploration
Unsupervised Learning of Terrain Appearance for Automated Coral Reef Exploration
复制标题
用于自动珊瑚礁探索的地形外观无监督学习
DOI：
--
复制
发表时间：
2009
期刊：
影响因子：
--
通讯作者：
Katrine Turgeon
中科院分区：
文献类型：
--
作者：
P. Giguère;G. Dudek;C. Prahacs;N. Plamondon;Katrine Turgeon
我们描述了一个基于视觉输入驱动的无监督学习的导航和覆盖系统。我们的目标是让机器人保持连续移动以上的地形感兴趣的视觉反馈，以避免leavingthis区域。作为一个特定的应用领域，我们有兴趣在开放水域中这样做，但该方法几乎没有特定领域的假设。具体来说，我们的系统采用了一种无监督学习技术来训练一个k-最近邻分类器，通过图像分割来区分不同地形类型的图像。一个简单的随机探索策略与此分类器一起使用，允许机器人收集数据，同时保持限制在珊瑚礁上方，而不需要保持姿态估计。我们在模拟中测试了这项技术，并在开放水域进行了现场部署。在后者中，机器人在20分钟内成功地在珊瑚礁上方自主航行。
We describe a navigation and coverage system based on unsupervised learning driven by visual input. Our objectiveis to allow a robot to remain continuously moving above a terrain of interest using visual feedback to avoid leavingthis region. As a particular application domain, we are interested in doing this in open water, but the approach makes few domain-specific assumptions. Specifically, our system employed an unsupervised learning technique to train a k-Nearest Neighbor classifier to distinguish between images of different terrain types through image segmentation. A simple random exploration strategy was used with this classifier to allow the robot to collect data while remaining confined above a coral reef, without the need to maintain pose estimates. We tested the technique in simulation, and a live deployment was conducted in open water. During the latter, the robot successfully navigated autonomously above acoral reef during a 20 minutes period.