LOS Vision System

Line of Sight Concept

As the satellite ejects from the dock, a pure blue light on the probe turns on. There is a camera on the docking plate which identifies the satellite's position via this light. In order for this to work the docking camera images are processed by an image recognition algorithm. If the satellite starts to drift out of line of sight then corrective actions can be taken to re-align the satellite.

This same concept enables the satellite to be brought back to the dock with the telemetry system issuing commands to readjust the position and distance of the satellite as it returns.

Function

The below process and software is written and active in the prototype. The images are actual real images from the physical simulation.

1. The docking plate acquires an image from it's LOS camera

2. A stark filter is applied to the image to improve contrast between the blue light and the background noise

3. The new image is fed to an algorithm which determine the position of the satellite in the frame

4. The position is referenced against a boundary detection algorithm to check whether the satellite is leaving the line of site

Timing

The process timings for the algorithm to run the above cases is...

Case 1 = 356ms

Case 2 = 1145ms

Case 3 = 1068ms

Improvements

1. The system currently works off of a static image on a capture by capture basis. This could be improved by using live video and path tracking

2. The algorithm could take into account data from other satellite sensors such as distance

References

1. The image filter routine from step two is from an image manipulation library for Java available at - http://www.jhlabs.com/ip/filters/

2. The source code for the image recognition software is available at the github for this project - https://github.com/mattythorne/ss-cornelius