The workflow of FTI-SLAM framework. The SLAM front-end comprises federated learning-enhanced deep neural networks for odometry, embedding, and loop closure detection. After optimising odometry based on loop constraints obtained from the front-end, the SLA (IMAGE)

ELSP

The workflow of FTI-SLAM framework. The SLAM front-end comprises federated learning-enhanced deep neural networks for odometry, embedding, and loop closure detection. After optimising odometry based on loop constraints obtained from the front-end, the SLA

Caption

The workflow of FTI-SLAM framework. The SLAM front-end comprises federated learning-enhanced deep neural networks for odometry, embedding, and loop closure detection. After optimising odometry based on loop constraints obtained from the front-end, the SLAM back-end provides the optimised trajectory.

Credit

Haochen Liu/University of Cambridge, Hantao Zhong/University of Cambridge, Weiyong Si/University of Essex

Usage Restrictions

Credit must be given to the creator.

License

CC BY

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.