How to tune params to avoid the terrible performance on other dataset?

laboshinl / loam_velodyne

Laser Odometry and Mapping (Loam) is a realtime method for state estimation and mapping using a 3D lidar.

http://wiki.ros.org/loam_velodyne

Other

1.7k stars 957 forks source link

How to tune params to avoid the terrible performance on other dataset? #143

Open LiShuaixin opened 5 years ago

LiShuaixin commented 5 years ago

Hello, I tried to test LOAM with Stevens-VLP16-dataset shared by LeGO-LOAM and compare the performance of these two system. I got a terrible result using LOAM when tested it directly without any parameters tunning. However, as the result stated in LeGO-LOAM paper, LOAM should have generated a acceptable result. Could any body else please give me a hint on this?

jasonCastano commented 3 years ago

Hello, I am having the same problem. Could you fix it?