Intelligent Non-Line-of-Sight Detection and Navigation Error Mitigation Technique for Autonomous Robots

Project Details

Description

The Non-Line-of-Sight (NLOS) detection and mitigation technique proposed in this project is an artificial intelligence (AI) algorithm that is used to improve the accuracy and robustness of the navigation system for automated guided vehicles (AGV). It is a software model built using deep neural networks (DNN) that determines whether the navigation signal is under NLOS mode and applies error corrections accordingly. Due to the powerful modeling capability of DNN, this technique can capture the "hidden" features of the localization signals that cannot be characterized by conventional approaches. And therefore, the accuracy and robustness of navigation can be significantly improved from sub-meters to a few centimeters.

Key findings

Automated real-time detection of bad channel conditions in localization and navigation.
StatusFinished
Effective start/end date01/01/202131/12/2021

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.