Grant Privett, Defence Science Technology Laboratory (Dstl); Calum Meredith, Defence Science Technology Laboratory (Dstl); Andy Winterbotham, Defence Science Technology Laboratory (Dstl); Jamie Symons, Defence Science Technology Laboratory (Dstl); Lauchie Scott, Defense R&D Canada; Allie Fawcett, Bornea Dynamics; Simon Lyddiatt, Bornea Dynamics; Ricky Gill, University of Manitoba; Brendan Doherty, University of Manitoba; Riley Sweeney, University of Manitoba; Philip Ferguson, University of Manitoba; Paul Harrison, Magellan Aerospace; Matthew Driedger, Magellan Aerospace; Vojtech Balaban, Magellan Aerospace
Keywords: Space to space, SDA, SWIR
Abstract:
The Redwing Space Domain Awareness (SDA) satellite is currently under development by Defence Research and Development Canada (DRDC) in partnership with Magellan Aerospace. Redwing includes a number of technology demonstration payloads, including the 6U CubeSat Little Innovator in Space Situational Awareness (LISSA) and its primary payload, a Short Wave Infrared (SWIR) camera package known as CompactSWIR. The CompactSWIR payload is being developed by the UK’s Defence Science and Technology Laboratory (Dstl) in partnership with Bornea Dynamics. Redwing is expected to be launched in 2027 and will occupy a Sun Synchronous Low Earth Orbit (SSO, LEO) with LISSA. LISSA will have a mission duration of 1 year and will detect LEO space objects in the southern hemisphere below 60 degrees South latitude.
It has become apparent that systems with in-orbit optical SDA cameras can experience stray light from the strong reflections of sunlight from the polar ice caps; in particular, the Antarctic ice sheet. This stray light can cause significant issues with image quality and cause the loss of opportunities to collect data when orbiting near the poles – when the scene is most likely to contain satellites. Fortunately, while within the visible range ice may reflect more than 90% of incident light, the reflectivity drops rapidly to just a few percent for wavelengths greater than 1.4 micron. Consequently, ice – when viewed in the SWIR band – is as reflective as coal, so the stray light issue should be greatly mitigated and lessen the need for optical baffling on a very small satellite platform. An Indium Gallium Arsenide (InGaAs) based SWIR sensor, with a quantum efficiency above 80% at ~0.95 – 1.65 micron (covering J band), has therefore been selected for LISSA. It is certain that some stray light will remain during passes over the Arctic during the summer, so the image processing software solution deployed must be able to work effectively and robustly in challenging sky background conditions.
A key limiting factor for most CubeSat missions is the need for small-size and low-weight components that can be operated within a modest power budget. This includes a limited data rate for communication links with the ground. Further, the lengthy development and qualification lifecycles typically associated with space missions adds to the complexity and cost of what are, typically, low budget endeavours. However, the burgeoning market in commercial off the shelf (COTS) components and subsystems for satellites provides hope for low-cost development and short project duration in the future. The intention throughout has therefore been to create a concept demonstrator system using affordable COTS components that will generate good quality metric observation data, which can then be correlated with a satellite catalogue and ultimately used to improve orbit knowledge of a detected satellite or highlight recent changes to its orbit. This requires data reduction on orbit to reduce the larger-sized imagery data to metric observations in order to meet data link limitations.
In adopting affordable hardware – including the on-board dual-core Xiphos Q7 processor board – it was accepted that great care would be required to ensure the data was processed in a manner that made efficient use of both the available dual 766 MHz ARM CPUs and the limited 512 MB memory available. Software has previously been developed by Dstl and reported at AMOS (Privett et al., 2017), that is capable of providing preprocessing of data from astronomical sensor observations, interfacing with freely available plate-solving software and detecting satellites. In preparation for use of this software pipeline with the CompactSWIR sensor, further development has taken place to take account of J-band requirements for plate-solving and photometric calibration; to ensure that satellite detections can be robustly tracked across time series frames; and to generally ensure that the pipeline is suitable for use on-board the satellite with the hardware limitations described. This newly developed software data reduction pipeline has been called Syrinx. It is worth noting that, as observed from orbit, the paths traversed by satellites are not always straight lines or gentle curves but can become parabolic in shape, so some approaches used to detect satellites present in imagery obtained by ground-based sensors were not applicable.
Initially the software was prototyped in Python 3 before being translated to C and cross-compiled for subsequent use and to allow testing, verification and elimination of memory leaks. The use of C was a requirement of the hardware due to the limitations described above. The software has been tested on both the engineering model and flight model CompactSWIR payloads and shown to function as expected in a ground test environment.
Test activities were undertaken using the CompactSWIR engineering model on a driven mount from under a 20 magnitudes/arcsecond2 sky in the UK to test the payload’s performance but also the software pipeline against real imagery. The test showed the software employing comparison stars down to J-Band magnitude 8 and demonstrated the software successfully extracting trails for both the targets and other satellites in the scene.
It is hoped that later iterations of LISSA-like systems will combine larger field of view sensors and collecting apertures with a greater processing capability so the catalogue correlation and orbit improvement processes can be undertaken either on orbit, or greatly reducing the quantity of data required for transmission to the ground.
Date of Conference: September 16-19, 2025
Track: Space-Based Assets