Heterogeneous sensing for target tracking: architecture, techniques, applications and challenges

Abstract

Target tracking applications are promising, it has great theoretical and practical significance, but the research faces great challenges. With the development of multi-modal depth sensing technology, a large number of scholars have proposed various target tracking methods based on heterogeneous sensing and got great results. This paper provides an overview of the techniques involved in target tracking in the different layers of the network and a comprehensive analysis of the research progress in heterogeneous sensing techniques in each layer. First, this paper introduced the single sensing scheme and heterogeneous sensing scheme in the physical layer. Second, we presented heterogeneous communication technologies and heterogeneous optimization methods for communication protocols in the network layer. Third, we combined several typical heterogeneous sensory target tracking applications and analyzed the application of cloud computing, edge computing, big data, and blockchain technologies. Finally, we discussed the challenges and future direction of heterogeneous sensory target tracking methods.

A disturbance rejection adaptive filtering approach for human motion tracking *

“…Human motion tracking is a technology to obtain the motion of the human in a three-dimensional space, which has been widely used in the rehabilitation training [1], human-robot collaboration [2], sports training [3] and virtual and reality [4], etc. These human motion tracking technologies are developed based on sensor technologies, such as mechanical [5], optical [6], acoustic [7], electromagnetic [8] and inertial sensors.…”

Measurement Science and Technology’s second century underway

“…Looking back over the 2023 issues, review articles have remained prominent with articles covering smart optical coordinate and surface metrology [3], linear and nonlinear dimensionality reduction from fluid mechanics to machine learning [4], guided ultrasonic wave propagation imaging [5], mechanical ventilation based on machine learning [6], deep learning-based methods in structural reliability analysis [7], 3D optical measurement techniques [8], the application of Josephson voltage standards [9], combustion of metal particles [10], heterogeneous sensing for target tracking [11], quantitative gas property measurements by filtered Rayleigh scattering [12], optical fiber reflectometry detecting static and dynamic Rayleigh spectra [13], stimulated emission depletion microscopy [14], mechanical fault diagnosis based on deep transfer learning [15] and advanced combustion dynamics [16].…”