Optimal cooperative sensing using multiple sensor platforms

Date

2016

Authors

Farmani, Negar

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

In this work, we present a distributed multi-target tracking system for cooperative unmanned systems equipped with vision sensors and communication modules. The sensors have some limitations and constraints which make the target tracking challenging. We, first, address the sensor manager and the path planner problems for a single unmanned system to track multiple mobile targets in an optimal manner. To that end, a Dynamic Weighted Graph technique determines the high target density areas and a Model Predictive Control is used to optimize the gimbaled camera pose. A real-time path planner with two-time horizon steps optimizes the unmanned system path by incorporating the uncertainty of target state estimations and the relative distance among the unmanned system and the targets. Then, we incorporate a distributed task allocation algorithm to track multiple mobile targets cooperatively when the number of unmanned systems is less than the number of targets. The algorithm benefits from sharing data amongst unmanned systems. To make the system scalable, the Density Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is used. The DBSCAN algorithm is integrated with the optimal sensor manager and path planner in order to distributively track a large number of targets with a limited number of resources. A set of Extended Kalman Filter (EKF) is used by each unmanned system to estimate the location of mobile targets. Later, we enhance the optimal path planner technique for cooperative unmanned systems while the communication range is limited. The new path planner technique is affected by the trade-off between communicating with neighbors and performing the assigned tasks. Finally, we address the problem of unknown delayed measurements in the observations and present a technique to improve the target tracking system by integrating the Hidden Markov Model and Out Of Order Sigma Point Kalman Filter (O3SPKF). The effective performance of the.

proposed systems are shown through simulation results.

Description

This item is available only to currently enrolled UTSA students, faculty or staff. To download, navigate to Log In in the top right-hand corner of this screen, then select Log in with my UTSA ID.

Keywords

Cooperative System, Geo-localization, Path Planning, Sensor Management, Target Tracking, Unmanned Systems

Citation

Department

Electrical and Computer Engineering