Adaptive Cooperative Localization Using Relative Position Estimation for Networked Systems With Minimum Number of Communication Links
Open Access
- 5 March 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 32368-32382
Abstract
In this paper, a novel solution for cooperative localization problem involving a network with multiple mobile agents accessing to only one beacon agent, is presented. The solution can be applied to a network with minimum number of communication links (subjecting the network topology to be an undirected spanning tree). The objective is to provide each agent in the network with information on its absolute position without using any on-board global positioning sensors such as GPS receivers. The proposed solution uses an adaptive relative position estimation algorithm for each pair of the neighboring mobile agents. The proposed estimation algorithm requires each agent to measure the relative inter-agent distance. Local velocity vector is also measured and transmitted to the neighboring agents. The estimation mechanism incorporating a signum function outperforms the recently established relative position estimation algorithms, in terms of positioning and tracking errors. An adaptive cooperative localization (ACL) algorithm is formed by augmenting the relative position estimation in a cooperative observer scheme suitably applicable for accomplishing localization task involving a network of mobile agents. The salient feature of the proposed ACL algorithm is that the communication graph among the agents needs only to have one undirected path between two agents in the network. Such convenience promotes easy practical implementation and lite computation for each agent. The proof of the proposed algorithm is provided using Lyapunov stability theorem. Three simulation case studies are presented to evaluate the performance of the solution in different scenarios, including the stationary and moving beacon agent as well as the non-cooperatively controlled and cooperatively controlled network of mobile agents. Comparative studies reveal that ACL algorithm is superior than the recently investigated linear-convex algorithm. The number of communication links required for the localization task to be carried out by the proposed algorithm is minimum, thus promoting a more preferable energy-efficient solution.Keywords
Funding Information
- TWAS-USM Postgraduate Fellowship
- The World Academy of Sciences
- USM Research University (1001/PELECT/8014029)
- RCMO Bridging Research (304/PELECT/6316106)
- School of Electrical and Electronic Engineering KPI
- Universiti Sains Malaysia
This publication has 53 references indexed in Scilit:
- Distributed Leader-Assistive Localization Method for a Heterogeneous Multirobotic SystemIEEE Transactions on Automation Science and Engineering, 2015
- Distributed Self Localization for Relative Position Sensing Networks in 2D SpaceIEEE Transactions on Signal Processing, 2015
- A survey of multi-agent formation controlAutomatica, 2015
- A Barycentric Coordinate Based Distributed Localization Algorithm for Sensor NetworksIEEE Transactions on Signal Processing, 2014
- Accuracy of iPhone Locations: A Comparison of Assisted GPS, WiFi and Cellular PositioningTransactions in GIS, 2009
- A Collaborative Multi-robot Localization Method without Robot Identification2009
- Monte Carlo localization of wireless sensor networks with a single mobile beaconWireless Networks, 2008
- Wireless sensor network localization techniquesComputer Networks, 2007
- Higher-order sliding modes, differentiation and output-feedback controlInternational Journal of Control, 2003
- Distributed multirobot localizationIEEE Transactions on Robotics and Automation, 2002