How can I find some valuable problems about SLAM which can be my master research projects?

Question

hi, my dear ROS friends! I am a first-year student for my Master Degree and SLAM is totally a whole new thing for me. There are so many things I need to learn. I am learning how to operate ROS to do SLAM. And I have wrote one simple EKF-SLAM code which were tested on a vehicle robot these days but the landmarks recognition part is from my partner. Im very confused right now cause I have to determine my research projects in two weeks however I haven't read many SLAM papers. I am trying to read some but I can't totally understand what the papers said without practice by my hand such as using a LIDAR to do SLAM. So what should I do?

---

Originally posted by blairlpp on ROS Answers with karma: 31 on 2016-06-01

Post score: 2

Answer 1

about ROS :just follow the Tutorials at wiki.ros.org

about SLAM : if you are doing laser 2D slam, you need to learn how to using gmapping and hectorslam. just search it on ros. they were two best laser slam algorithm.

in addition,gmapping needs odometer(motor encoder data and IMU),while hectorslam not. it's better for you to improve your own code on the base of gmapping or hectorslam.

recently SLAM research are mostly in visual slam field and 3D laser slam.

for 3D laser slam, you can find it on youtube. it's similar to 2D laser slam. for Vslam, there is a lot new things to learn. mostly computer vision related. popular direction includes: RGBD slam (with kinect depth data), stereo slam, Monocular slam.

---

Originally posted by panda with karma: 63 on 2016-06-02

This answer was ACCEPTED on the original site

Post score: 1

Answer 2

This papers describes everything in detail. link text

Anyone looking for open problems in SLAM these days should refer to this paper.

---

Originally posted by ZainMehdi with karma: 225 on 2017-09-11

This answer was NOT ACCEPTED on the original site

Post score: 1