Data Fusion Techniques for Object Identification in Airport Environment UTC Aerospace Systems

Format:

Conference/Event

Author/Creator:

Thupakula, Thupakula, author.

Conference Name:

Aerospace Technology Conference & Exposition (2017-09-26 : Fort Worth, Texas, United States)

Language:

English

Physical Description:

1 online resource

Place of Publication:

Warrendale, PA SAE International 2017

Summary:

AbstractAirport environments consist of several moving objects both in the air and on the ground. In air moving objects include aircraft, UAVs and birds et cetera On ground moving objects include aircraft, ground vehicles and ground personnel et cetera Detecting, classifying, identifying and tracking these objects are necessary for avoiding collisions in all environmental situations. Multiple sensors need to be employed for capturing the object shape and position from multiple directions. Data from these sensors are combined and processed for object identification.In current scenario, there is no comprehensive traffic monitoring system that uses multisensor data for monitoring in all the airport areas. In this paper, for explanation purposes, a hypothetical airport traffic monitoring system is presumed that uses multiple sensors for avoiding collisions. The referenced system employs multiple types of sensors for object data collection in different situations, wherein the collected multi object data is combined to classify and identify the objects, and identified objects are accurately tracked for collision prediction.This paper discusses a data fusion model of multisensor data for object identification in an airport environment to allow the traffic monitoring system to determine the shape, type and position of an object. As a future scope of this paper, the object shape, type and position data from the object identification stage is provided as input to the next stage in the airport traffic monitoring system to track the object movements for collision prediction. Multiple type sensors are arranged in different configurations such as complementary, competitive and cooperative arrangements. Data from these sensors is combined for object detection and identification. Optimal fusion model and object model mapping algorithms are discussed for the object identification purpose. A case study of the competitive sensor data fusion is also discussed in this paper

Notes:

Vendor supplied data

Publisher Number:

2017-01-2109

Access Restriction:

Restricted for use by site license