解决SLAM地图重用问题

上次更新时间：2016年5月24日

持续关注的帖子

重用地图信息系统的讨论

保存关键帧姿态和点云的实现

已经成功编译poine的所有文件，但编译过程十分缓慢。

注：后续会考虑在Qt下重新构建poine的项目。

关于SLAM系统数据序列化和反序列化

序列化和反序列化是地图重用的关键。

ROS bagfile的启发

The bag file contains interlaced, serialized ROS messages, dumped directly to a single file as they come in over the wire. This is the most performance and disk-friendly recording format possible.
注：由以上可知，或许能够借鉴rosbag的序列化和反序列化实现，因此，要查看rosbag的源码。
注：初步查看，发现rosbag的实现非常复杂，当然也非常强健。

一个成功的实现

序列化内存地图信息，并反序列化磁盘地图数据

其他借鉴

stereo_slam系统的实现。尤其关注其信息发布系统。

Robot 3D Localization using a Kinect and an 3D Map (octomap)
简介：This video shows a mobile robot using a 3D TOF camera (Kinect) to localize itself towards a 3D map of the environment.
The 3D map was created using 6d slam and it is represented as an octree using the Octomap library (which I found very useful).
In the viewer the roof is not shown for visual purposes.

搜索词

slam and path planning

相关资讯

MIT researchers have developed a topological approach to SLAM that allows a robot to map large scale environments by combining smaller maps. Applications for the technology range from Navy AUVs to flying robots that can map caves in Afganistan.