loading page

A Monocular vision positioning and tracking system based on deep neural network
  • +1
  • Huijun Li,
  • Yu Zhang,
  • Bin Ye,
  • Hailong Zhao
Huijun Li
China University of Mining and Technology

Corresponding Author:[email protected]

Author Profile
Yu Zhang
China University of Mining and Technology
Author Profile
Bin Ye
China University of Mining and Technology
Author Profile
Hailong Zhao
China University of Mining and Technology
Author Profile

Abstract

In order to locate the mobile robots in three-dimensional indoor environment, mostly global navigation satellite system-denied space, a monocular visual space positioning algorithm based on deep neural network is proposed. First, we employ the lightweight YOLOv5 algorithm for target detection, and the LibTorch deep learning framework is used for model deployment to improve the inference speed. Moreover, a multi-layer perceptron (MLP) neural network with four inputs and two outputs is constructed, which regress the coordinates of the robot in the field coordinate system to complete the target localization, and this method is compared with the mathematical model solving algorithm to reflect the accuracy and superiority of positioning algorithm based on deep neural network. The proposed positioning and tracking system has been successfully applied to ICRA robot competition, and results show that the positioning error estimated by our method is within 10cm whilst having good real-time performance.
02 Nov 2022Submitted to The Journal of Engineering
04 Nov 2022Submission Checks Completed
04 Nov 2022Assigned to Editor
23 Nov 2022Reviewer(s) Assigned
29 Dec 2022Review(s) Completed, Editorial Evaluation Pending
03 Jan 2023Editorial Decision: Revise Major
20 Jan 20231st Revision Received
27 Jan 2023Submission Checks Completed
27 Jan 2023Assigned to Editor
29 Jan 2023Reviewer(s) Assigned
12 Feb 2023Review(s) Completed, Editorial Evaluation Pending
17 Feb 2023Editorial Decision: Accept