A Review of Internet of Things (IoT) based Engineering Applications and Data Fusion Challenges for Multi-rate Multi-sensor Systems

Cooperative target estimation of strap-down missiles considering random measurement delay

“…A novel approach through the re-organization of measurements was proposed to deal with the robust Kalman filtering problem for linear discrete-time systems with norm-bounded parametric uncertainty under both instantaneous and delayed measurements, 29 and a sufficient condition for the existence of a robust Kalman filter is derived. Bernoulli random variables are also used in literature studies 30,31 to explain the random delay phenomenon in multi-rate systems. Two Bernoulli stochastic variables were employed in literature 30 to describe the phenomena of one-step random delays and packet losses in data transmission.…”

“…However, this approach is not appropriate for low-cost, small-sized strap-down vehicles since it requires more computational storage space and the computational cost rises as the delay grows. According to the literature, 31 the system with observation time delay is converted into a multi-rate system without time delay, and then a sequential fusion estimation approach based on linear Kalman filtering and linear minimum variance is presented. Although this approach requires only a little amount of computational work, the robustness of fusion using a sequential approach is weak, and an inaccuracy in one node's estimation will have a significant impact on both the outcomes of following estimations and the fusion itself.…”

“…Inspired by literature studies, [28][29][30][31] a novel distributed cooperative estimate fusion approach is suggested in this study to address the issue of cooperative target estimation for multiple strap-down vehicles while taking into account the unpredictable transmission latency of observation data. The approach implements local estimate for each strap-down detection subsystem and, on the basis of this, implements the global estimation of multiple detection nodes by using the principle of least estimation error covariance trace.…”