• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    公众号

vijay4313/SLAM-using-GTSAM: Simultaneous Mapping and Localization framework for ...

原作者: [db:作者] 来自: 网络 收藏 邀请

开源软件名称(OpenSource Name):

vijay4313/SLAM-using-GTSAM

开源软件地址(OpenSource Url):

https://github.com/vijay4313/SLAM-using-GTSAM

开源编程语言(OpenSource Language):

MATLAB 100.0%

开源软件介绍(OpenSource Introduction):

SLAM-using-GTSAM

Simultaneous Mapping and Localization framework for drone

Introduction: This project is aimed at implementing a vision-based Simultaneous Localization and Mapping (SLAM) to genreate a 3D map of the environment using a GTSAM and iSAM packages from GTSAM matlab toolbox. The premise of the project is to find the survivors in a power plant that has been destroyed by a recent earthquake. Due to the hazardous materials, drones (quadcopters) will be used for this mission. There are two parts to the mission. First, building the map of the current condition of the plant using a light weight drone with nothing but a single camera and an IMU. Then use a second drone is capable of carrying other needed supplies for the rescue mission. The second drone will use the map from the first drone and go to the rescue target location with the supplies.

Algorithm: The algorithm for estimating 3D world co-ordinates of camera poses and AprilTags using GTSAM involves 3 steps:

  1. Initialisation For GTSAM packages to optimize the world co- ordinates properly, the world co-ordinated are to be initialized properly. Knowing that bottom-left corner of tag 10 will be origin in world co- ordinates, the poses of the camera frames con- sisting of tag 10 can be computed with RPnP approach.

  2. Building the Factor Map The factor graph is then built using GTSAM pack- ages for MATLAB, with the initial estimates, mea- surements in pixel co-ordinates and noise models.

  3. Optimising the factor graph The factor graph, thus created, is optimized using Levenberg-Marquadt non-linear optimizer.

The iSAM part of the project too follows a similar approach. However, the factor graph are created and optimized in batches (usually 1 frame at a time), as opposed to all the frames at once.




鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
热门推荐
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap