Title: GPS availability prediction based on air-ground collaboration

Authors: Zun Liu; Wenlian Huang; Yanyan Chen; Jie Chen

Addresses: College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, 518060, China ' College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, 518060, China ' School of Software, Henan University of Science and Technology, Luoyang, 471023, China; College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, 518060, China ' School of Software, Henan University of Science and Technology, Luoyang, 471023, China; College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, 518060, China

Abstract: Robots such as unmanned aerial vehicles (UAVs) have been widely applied in emergency rescue scenes. However, there is a lack of distribution of ground GPS signals in the complex environment, which indirectly affects the take-off and landing of UAVs. To solve this problem, we have proposed an air-ground collaborative mapping system based on the Gaussian Process (GP) and convolutional neural network (CNN). Firstly, CNN is used to predict whether the GPS signals are available. And the GP is used to interpolate and predict the areas not visited by the UAVs, then the GPS signal distribution map is obtained. Compared with the traditional mapping methods, the system does not require size parameters and can build maps more efficiently and quickly.

Keywords: mapping; robots; rescue scene; Gaussian process; CNN; convolutional neural network.

DOI: 10.1504/IJSSC.2023.133240

International Journal of Space-Based and Situated Computing, 2023 Vol.9 No.3, pp.147 - 157

Accepted: 29 Mar 2023
Published online: 03 Sep 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article