what are the current state-of-the-art SLAM algorithms for these scenarios?

Question

Hi, Have found that SLAM is not an algorithm but a framework for localization and mapping. After searching, found that it has been around for 30 years and many approaches have been taken to solve the problem.

I would like know the current state-of-the-art, solid and proven SLAM algorithm for the following scenarios, applicable to an autonomous ground vehicle:

1 Indoor (from a small area such as a room, up to a full floor) 1.1 LiDAR + Odometry/IMU 1.2 LiDAR + Odometry/IMU + Camera

2 Outdoor (from a small parking lot to potentially large areas) 2.1 LiDAR + Odometry/IMU 2.2 LiDAR + Odometry/IMU + Camera

The Google's Cartographer, was opens sourced recently. However I am not sure whether it can be used for 2D, or if it can be used for outdoors.

If possible give a pointer to a supporting paper, or a place showing results. Also, please indicate if the implementation is already available in ROS.

Many thanks.

---

Originally posted by mape1082 on ROS Answers with karma: 3 on 2016-11-07

Post score: 0

---

Original comments

Comment by huanxiner on 2016-11-08:
Try reading some papers yourself.

Answer 1

Cartographer is state-of-the-art and works indoors/outdoors in 2D/3D. We published a paper on our 2D algorithms here: http://static.googleusercontent.com/media/research.google.com/en//pubs/archive/45466.pdf

---

Originally posted by damonkohler with karma: 3838 on 2016-12-01

This answer was ACCEPTED on the original site

Post score: 0