基于激光雷达增强的无人机视觉SLAM地图构建方法的研究

Research on UAV Visual SLAM Construction Method Enhanced by Lidar

同时定位与建图(SLAM)技术已广泛应用于各类自主移动平台中,根据使用的传感器类型主要分为使用视觉传感器的SLAM方案和使用激光雷达的SLAM方案。视觉SLAM的建图效果较为依赖环境中的光照条件,所以视觉SLAM在室外环境及动态环境的建图效果较好。为了使无人机能够随时自主感知周围环境的变化,利用各个传感器之间冗余和互补的特性来获取更加丰富的环境信息,可以更准确的检测出飞行环境中的障碍物,设计一种单线激光雷达与相机组合检测的SLAM方法。首先对相机数据进行预处理,然后标定出相机与激光雷达传感器的外参矩阵,使用Cartographer算法,利用相机在环境中进行定位,并用激光雷达对障碍物进行检测。通过搭建Gazebo仿真环境验证所提出的检测方法的可行性与有效性,仿真实验结果表明该方法能够提升环境障碍物检测的准确性。

Simultaneous Localization and Mapping(SLAM)technology has been widely used in various autonomous mobile platforms.According to the type of sensor used,it is mainly divided into SLAM schemes using visual sensors and SLAM schemes using lidar.The mapping effect of visual SLAM is more dependent on the illumination conditions in the environment,so the mapping effect of visual SLAM is better in outdoor environment and dynamic environment.In order to make the UAV autonomously perceive the changes of the surrounding environment at any time,and use the redundancy and complementary characteristics between various sensors to obtain more abundant environmental information,so as to more accurately detect obstacles in the flight environment,a SLAM method for combination detection of single-line lidar and camera is designed.Firstly,the camera data is preprocessed,and then the external parameter matrix of the camera and lidar sensor is calibrated.Using the Cartographer algorithm,the camera is used to locate in the environment,and the obstacles are detected by lidar.The feasibility and effectiveness of the proposed detection method are verified by setting up a Gazebo simulation environment.The simulation results show that the method can improve the accuracy of environmental obstacle detection.