Collecting time and space position from tracking mounts

Download the PDF version of the article here

Time and space position information (TSPI) data from tracking mounts has traditionally been derived by using a suite of different software programs in different evironments.

Typically, this would include:

■ Star calibration – The calibration for systematic errors is performed using the stars as reference. This requires real time control of the mount from the calibration software.

■ The calibration for systematic errors from fixed reference targets has been performed in one program as an alternative to star calibration or as a complement to star calibration.

■ The 2D tracking of the images has been performed in a second program.

■ The corrections for systematic errors of the 2D tracks have been performed in a third program.

■ The 3D triangulation by using 2D tracks from multiple tracking mounts has been performed in a fourth program.

■ The result presentation has been done in a fifth program.

There are, of course, a number of disadvantages to using a large number of different software packages to solve one application. The most obvious disadvantages are from a maintenance point of view and for the operators who must learn a number of different GUIs. One other major disadvantage is that when an error is discovered in the TSPI result, the operator must go back and redo the complete chain of processing.

Until now there has been no commercial-off-the-shelf (COTS) product available to handle the complete application from one single GUI. But now there is TrackEye. The process is divided in two different steps:

■ Calibration of the mounts for systematic errors from fixed reference targets. The mounts are moved to generate a moving image of the reference target, which is tracked automatically. From this data the calibration parameters are generated as well as the
correction parameters. Both the calibration and correction parameters are stored in an archive that is part of software.

■ The analysis of the real test. By using a predefined template the operator’s task is reduced to 2D tracking in one image sequence from each tracking mount.

With this approach the operators only need to learn one user interface, but an even larger advantage is the ability to detect and correct any type of error in the analysis. With the complete process, including 2D tracking, in one GUI you can immediately view the result of a 2D tracking in the final TSPI data. The image sequences from the different mounts are tracked in parallel and, after each image has been tracked, the graph containing the TSPI data is updated. All the processing between the 2D tracking and the final data presentation is included in the predefined template.

Having the complete application in one GUI also allows all data associated with one test to be stored together in one database.

COTS motion analysis software that produces TSPI data also contains a number of basic functions that can be used in other test range applications. TSPI from tracking mounts is one of the most demanding applications at a test range from a motion analysis perspective. Almost all other applications that are using images as sensors are a subset of TSPI from tracking mounts and can, therefore, also be solved by the same COTS software. The great advantage for the users is that they only need to learn one user interface and to maintain one software package.