基于激光测距AdaBoost分类的室内环境识别

移动机器人所处环境的地点语义信息能够提高机器人自主定位、路径规划和人机互动的能力.为了让机器人识别环境中不同地点类型,提出一种对机器人所处环境地点类型进行语义分类的方法.该方法对激光传感器的测距数据进行特征提取,通过提取的样本集利用强化学习AdaBoost方法构建分类器,对于环境中多类型地点分类识别,将获得的二分类器有顺序地排列建立分类列表形成多分类器,将获得的多分类器运用到房间、走廊和门口的分类识别中.实验结果表明:移动机器人通过该方法都能对环境下不同地点类型进行有效的分类识别.

Indoor Environment Recognition Based on AdaBoost Classification of Laser Ranging

The semantic information about the type of place improves the capabilities of a mobile robot in various domains including localization,path-planning,or human-robot interaction.In order to allow the robot to identify different types of place in the environment,a method identifying different types of place in semantic information is proposed.The features are extracted based on laser range data,and AdaBoost,a supervised learning algorithm,is used to train a set of classifiers for place recognition.For the environmental multi types of place classification,multi-classifier uses the best sequence of binary classifiers to get a decision list.The obtained multi-classifier is applied to classification of rooms,corridors and doorways.Experimental results demonstrate the effectiveness of the approach for classifying different places in various environments for mobile robot.