search

koide3 / glim

GLIM: versatile and extensible range-based 3D localization and mapping framework
MIT License
578 stars 77 forks source link

Reproduce Flatwall experiment #102

Closed se7oluti0n closed 3 weeks ago

se7oluti0n commented 4 weeks ago

Hello, first of all, thank you for generously opensource this repository.

From your paper and the project page, I understand that Glim is able to recovery from lidar degeneration. I try to verify that using the flatwall_01 data but couldn't able to reproduce the recovery. Below is my configuration, could you show me which is wrong in my configuration. Addtional question: Which module are mainly reponsible to recovery from degeration ( odometry_estimation, submapping or global mapping )?

config_sub_mapping_cpu.json config_global_mapping_cpu.json config_odometry_cpu.json config_ros.json config_sensors.json

koide3 commented 3 weeks ago

The odometry estimation module is the main part to deal with point cloud degeneration, and for the flatwall experiment, you need to use OdometryEstimationGPU. You can find the configuration file here. Note that the framework has been largely updated since the first release, and it may show slightly different results for a few sequences.

se7oluti0n commented 3 weeks ago

Thanks for the prompt reply. Close threat here.