移动机器人基于多传感器信息融合的室外场景理解

本文研究了移动机器人多传感器信息融合技术，提出一种融合激光测距与视觉信息的实时室外场景理解方法．基于三维激光测距数据构建了高程图描述场景地形特征，同时利用条件随机场模型从视觉信息中获取地貌特征，并以高程图中的栅格作为载体，应用投影变换和信息统计方法将激光信息与视觉信息进行有效融合．在此基础上，对融合后的环境模型分别在地形和地貌两个层面进行可通过性评估，从而实现自主移动机器人实时室外场景理解．实验结果和数据分析验证了所提方法的有效性和实用性．

Outdoor scene comprehension of mobile robot based on multi-sensor information fusion

We propose for the mobile robot an algorithm for real-time outdoor scene comprehension based on multisensor information fusion. An elevation map is built from 3D laser points for describing the terrain features, while the ground surface characters are acquired from visual information by using the conditional random field model. By projecting the grids of the elevation map onto the image, the fusion of terrain features and ground surface characters are achieved by using the statistical method. Then, the grids which fuse multi-sensor information are evaluated for the traversability of the outdoor environments, thus achieving the outdoor scene comprehension of the mobile robot. Experimental results and data analysis show the validity and practicability of the proposed approaches.