Skip to content

Training a deep learning system for localization without an explicitly defined map.

Notifications You must be signed in to change notification settings

bnurbekov/LSTM_based_localization

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Status

Work in progress (not cleaned up)

Model

Multi-layered LSTM was used for the change of pose regression.

Regularization

Dropout.

Preliminary results

Training loss visualization

The results of the training are shown on the graph below:

Image of training loss

Validation set loss

The validation set loss was equal to 0.001 (1 mm)

About

Training a deep learning system for localization without an explicitly defined map.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages