Synthetic aperture radar (SAR) is a key technology for Earth observation, that allows the unobstructed Earth observation and imaging including change detection and disaster management with high spatial resolution and independence of distance and weather conditions. Advancement in processing techniques and analyses will soon make it possible to have realtime monitoring. Due to the increasing ubiquity of wireless communications and the ever-increasing utilization of the radio spectrum, radio frequency interference is expected to become a major issue impacting spaceborne SAR technologies.

This research project aims to investigate interference topic in modern spaceborne SAR systems. It will construct a modelling framework to capture terrestrial interference using both analytic tools from stochastic geometry and simulation tools. It will then develop novel machine learning methods to detect interfering regions in reflected SAR signal based on both training samples, supervised and unsupervised learning methods. The research will thus develop interference mitigation techniques that will enhance SAR observation under the increasingly crowded radio spectrum.

The results of this project are expected to enhance the reliability of spaceborne SAR earth observation with direct applications in defence & security as well as in agriculture farming which is totally aligned with SmartSat CRC second research area “Earth observations from space” objectives and applications.

This research aims to provide a solution for energy-efficient AI-based on-board processing of hyperspectral imagery supporting automated early detection of fire smoke. We propose using modified and resampled MODIS imagery data that emulates the swath as well as spectral, spatial, and radiometric resolution of HyperScout-2 channel 1 hyperspectral imagery. In doing so, we intend to provide a solution that meets on-board processing limitations and up/downlink data transfer restrictions of the Kanyini – HyperScout-2/ with Intel’s Myriad X VPU chip.

Based on a semi-automatically created fire smoke training dataset, our proposed AI processing approach is performed at two levels: a) on-board feature and band selection, and b) ground AI neural network tasks – in order to optimize on-board processing and downlink data transfer. Expected outputs include on-board and ground AI algorithms for fire smoke detection, applicable for various hyperspectral imagery datasets.

The project will develop the concept of an intelligent payload termed Satellite Proximity Surveillance System or SatProx that can automatically monitor the near region (< 30km) of a geostationary (GEO) satellite for potential approaches by an adversarial satellite. Such unexpected rendezvous manoeuvres, which occur surreptitiously at low relative speeds (< 0.7 m/s), could indicate hostile on orbit activities such as shadowing, interference and hijacking. By providing early warning, SatProx buys valuable time for the host satellite to engage in mitigation strategies.

The project will produce the design and specification of SatProx, including the sensor suite (e.g., optical, IR, thermal, LiDAR) and edge data processing subsystem that can support real-time inference on the data stream to automatically detect and raise alarm on potential incoming spacecraft. The project will also develop graphical simulation software that can render the GEO environment for the optical channel (visible spectrum) under varying conditions environmental conditions, including simulated encroachment by other adversarial satellites. Based on the simulation software, machine learning algorithms for real-time adversarial satellite detection and physical characterisation (shape, pose and trajectory of adversarial satellite) will be developed. The algorithms will then be demonstrated on the simulation environment with edge processing hardware (embedded GPU or FPGA) in the loop.

Very Low Earth Orbit (VLEO) is a region of space above the Kármán line and up to The International Space Station at approximately 450km altitude. This region is a challenge to operate in due to atmospheric effects limiting the total lifespan of a typical satellite.

This project will investigate the development of a purpose-built attitude and orbit control system (AOCS) for operation in VLEO to create new mission opportunities, improving data products and communication services. The operational advantage of closer proximity to ground and high revisit rates will bring enhanced commercial products and capabilities to consumers, expanding the ease and access to space.

The new AOCS will bring innovation on orbital control in regions of significant atmospheric density, coupling new propulsion methods and attitude manoeuvring concepts for sustainment of new small satellite operations.

This project aims to conduct a case study on the process of development of the South Australian Government’s first satellite mission, Kanyini.

The proposed research will capture the issues, activities, challenges and opportunities for improvement the process of conception, design, manufacture, testing, launch and operation of small satellite missions and identify learnings that will guide future missions and process improvements. The proposal given here covers the space mission’s development up until completion of ground qualification testing.

A second follow-on project will follow the mission through its launch and on-orbit operations.

Current ISR satellites typically play a passive data collection and dissemination role, with processing typically done further downstream at ground stations. This “offline” processing approach introduces significant delays in converting data to actionable insights, prevents low-latency coordination between end-users (e.g., field operatives) and space-based assets, and precludes more intelligent sensing capabilities; for example, adaptive tasking of the sensor suite based on real-time data enhancement and analytics to improve intelligence gathering.

The project builds upon existing work at BAE Systems and the partner academic institutions (UoA and UNSW) on machine learning for ISR applications. The project will develop novel algorithms and workflows to enable machine learning on nanosatellites for space based ISR from multi-modal sensors. Note that the capacity of current edge computing hardware (e.g., Nvidia Jetson series) is still significantly smaller than standard hardware, thus necessitating algorithms for model pruning and data pre-processing to perform on-board machine learning. Operating in space also presents unique obstacles to updating the pre-trained on-board models, in terms of procuring data and supervisory labels for retraining, and bandwidth constraints in updating models. The project will tackle the above challenges to produce a novel space analytics engine that is reconfigurable after launch, which significantly increases the value proposition of on-board processing.