Distributed Data Fusion Algorithm Research Articles

Distributed Data Fusion for On-Scene Signal Sensing With a Multi-UAV System

This article presents an algorithmic framework for distributed data fusion using a network of multiple unmanned aerial vehicles (UAVs). Each UAV in the network is capable of sensing signals, performing computational tasks, and exchanging information with its immediate neighbors. Based on the limited local measurement and exchanged information only from neighbors, we aim to propose a fully distributed algorithm to integrate information from various measurements of individual UAVs. As an important building block of the distributed data fusion algorithm, a novel sum consensus scheme is also developed with only assuming row-stochastic weight matrices. Consequently, through the proposed distributed algorithm, each UAV is able to obtain a more reliable fused sensing result with higher accuracy, compared to the local measurement collected by the UAV itself. The effectiveness of the proposed algorithmic framework is validated by both theoretical analysis and numerical simulations.

Multiple RGB-D sensor-based 3-D reconstruction and localization of indoor environment for mini MAV

Micro aerial vehicle (MAV) with lower cost, smaller size and more flexible flight performance plays a greater role in the global position system-denied (GPS-denied) indoor tasks. This paper focuses on the realization of three-dimensional (3-D) panoramic environment reconstruction and target localization without relying on any external navigation aid. By establishing a visual sensor network (VSN) composed of multiple RGB-Depth (RGB-D) sensors, a fast 3-D model of the observed environment is built utilizing an improved speeded up robust feature (SURF) extraction algorithm and iterative closest point-based (ICP-based) reconstruction algorithm. A distributed data fusion algorithm known as kalman-consensus filter (KCF) is then used to estimate a more accurate global position and trajectory of mini MAV. Both software simulations and real indoor flight experiment results demonstrate that the proposed approach achieves a fast and reliable 3-D map of the indoor environment and the accuracy of localization is largely improved.

Distributed state estimation and data fusion in wireless sensor networks using multi-level quantized innovation

Low energy consumption and limited power supply are significant factors for wireless sensor networks (WSNs); thus, distributed state estimation and data fusion with quantized innovation are explored. The universal features of practical WSNs are investigated, and a dynamic transmission strategy is introduced. Furthermore, quantization state estimation based on Bayesian theory is derived. Unlike previous algorithms suitable for processing scalar measurement, the proposed distributed data fusion algorithm is applicable to general vector measurement. Furthermore, the efficiency of the proposed dynamic transmission strategy is analyzed. It is concluded that the proposed algorithm is more efficient than previous methods, and its estimation accuracy comparable to that of the standard Kalman filtering, which is based on analog-amplitude vector measurement.

Distributed Data Fusion Algorithm with Low Energy Consumption and High Accuracy in WSN

Distributed Data Fusion Algorithm with Low Energy Consumption and High Accuracy in WSN

Distributed data fusion algorithms for inertial network systems

New approaches to the development of data fusion algorithms for inertial network systems are described. The aim of this development is to increase the accuracy of estimates of inertial state vectors in all the network nodes, including the navigation states, and also to improve the fault tolerance of inertial network systems. An analysis of distributed inertial sensing models is presented and new distributed data fusion algorithms are developed for inertial network systems. The distributed data fusion algorithm comprises two steps: inertial measurement fusion and state fusion. The inertial measurement fusion allows each node to assimilate all the inertial measurements from an inertial network system, which can improve the performance of inertial sensor failure detection and isolation algorithms by providing more information. The state fusion further increases the accuracy and enhances the integrity of the local inertial states and navigation state estimates. The simulation results show that the two-step fusion procedure overcomes the disadvantages of traditional inertial sensor alignment procedures. The slave inertial nodes can be accurately aligned to the master node.