In the next few decades, Artificial Intelligence (AI) will be the biggest commercial opportunity in the world. As we gain access to an ever-richer tapestry of data and knowledge, the enhancement of deep learning through AI is intrinsically linked to the rise in commercial space and satellite activity.
As the space and satellite industry in Cornwall scales at pace, so too does the region's AI capabilities. From edge AI for manufacturing, to AI algorithms being developed to remove cloud cover and unlock satellite data for business transformation - Cornwall is home to a hugely unique mix of companies tapping into the mutually beneficial relationship between the two technologies, in turn becoming a hub for innovation in AI applications.
Spearheading Cornwall’s acceleration towards becoming the UK's premier location for space manipulated AI and deep learning are the team at Goonhilly Earth Station Ltd, home of the UK’s Space AI institute. The institute provides a melting pot for tech vendors, industry, end users and academia to collaborate and create extraordinarily cutting-edge business opportunities. Underlying the organisation, Goonhilly have also built a state-of-the-art GPU-enabled Data Centre with the latest high-performance GPU solutions and equivalent to Tier 3-4 standard.
In today’s digital world, businesses require on demand access to real-time data, as well as the AI/ Machine Learning analytics, tools and techniques to create new business value from their data. Goonhilly’s data infrastructure readiness has enabled many companies to test and accelerate their offerings in this area.
Meet the maverick companies taking data-led transformations to the next level:
Start-up Aspia Space is leading the way in AI for Earth Observation - having developed a cutting-edge AI algorithm that gives unprecedented cloud-free Earth Observation data intelligence in near time.
Earth Observation data is increasingly relied upon for critical intelligence across multiple industry sectors - from agriculture to environmental monitoring to land management and many more - but Earth observation imaging from satellites is currently hampered by clouds, meaning intelligence is lost during each period of cloud cover.
With funding from AeroSpace Cornwall, Aspia Space has launched a cutting-edge solution that sees through clouds. Its unique patented algorithm combines AI and cloud-penetrating Synthetic Aperture Radar (SAR) Earth observation data to predict detailed multispectral satellite imagery of the ground, from farmland to flood plains to mountain ranges, even in the presence of 100% cloud cover.
This will enable organisations to see both the big picture and finer details with accuracy, identify and understand trends, detect issues and predict risks and make faster, more informed decisions.
4 Earth Intelligence
4 Earth Intelligence (4EI) is also maximising the potential of space data - enabling leading Environmental Intelligence through the unique application of AI and Earth Observation. At the cutting-edge of their field, they use space data for the betterment of humans and the planet.
Their innovative practices stem from a diverse set of knowledge and expertise, delivering solution-focused capabilities including Artificial Intelligence, software development, geospatial mapping and environmental analytics. The company has worked with leading public and private organisations across the globe - enabling them to harness space data to gain further knowledge and in turn, become more environmentally conscious.
Their novel use of space technology and AI allows them to create large scale environmental baselines through the mapping of entire landscapes that can then be turned into an audit of environmental assets. This data can be translated into actionable intelligence for a range of applications - on both a small and large scale.
The focus of 4EI’s Cornwall Team is applying satellite Earth Observation within the Offshore Renewable Energy sector, specifically offshore floating wind. With expanding attention on energy transition, space data can provide a powerful tool to increase efficiencies throughout the supply chain.
Set to revolutionise heavily regulated sectors like medical, aerospace and space beyond recognition, edge AI works by helping machines make intelligent decisions and provide critical feedback whilst keeping data on board a local device.
Embedded software company Bluefruit, have been innovating in edge AI, producing a machine learning solution that achieves high levels of accuracy with small amounts of sample data – one of the main barriers to applied AI. Their first prototype, the Audio Classification Equipment (ACE) project is an embedded software application that can take pre-recorded audio or live audio and categorise the sample into pre-trained categories. Making it possible to ‘listen’ to machines - diagnosing internal faults through sound recognition using real-time AI solutions.
The initial training of any AI algorithm is a critical part of the R&D process, but it’s so computationally intensive that each iteration can take hours, days or even weeks to achieve on standard hardware. A key factor to the success of this project was the ability to use the supercomputing facilities at Goonhilly Earth Station’s Space AI Institute. With the added help of the Goonhilly workstations each training iteration is reduced to a few minutes meaning that Bluefruit were able to complete all of the neural network R&D and training within eight weeks.
Sechoia Ltd is a UK based, self-funded, tech start-up, disruptive technology. They have recently delivered a proof-of-concept software that combines depth imaging, edge computing and machine learning to create real-time data analytics.
Sechoia’s unique combination of 3D mapping with object tracking allows for spatial localisation, creating intuitive datasets which highlight how objects interact with their environment. Using AI in this way opens up a whole new world of smart planning where businesses are able to predict spatial outcomes and future-proof based on data. The construction of a new office for example, can be influenced by algorithms that map and detect the spatial patterns within previous office spaces, unlocking full project potential.
By developing light-weight algorithms, Sechoia have achieved results consistent with techniques that rely on larger, centralised, computational resources, while running on low-power, cost-effective edge devices. This enables real-time vision processing, depth imaging and data analytics to be performed in the field, removing the need to transmit large volumes of data to the cloud for processing.
This tech has huge potential across a range of industries, especially within space and aerospace. AeroSpace Cornwall is funding Sechoia’s next round of technological enhancements and environmental testing.
In a drive to expand their client base, Sechoia are also using their skills to develop efficient, embedded, edge based AI software to supply bespoke software solutions to businesses in the defense sector.
Cornwall-based Assentech is pioneering AI to reduce fugitive emissions - a huge problem that results from farm storage tanks and biogas digesters. In the UK alone there are approximately 900 COMAH* registered sites.
Ewart Cox, Managing Director has been on a mission for 25 years in the tank storage industry, he felt so strongly about reducing fugitive emissions that he set up Assentech in 2009 to focus entirely on helping the tank storage industry to take control and reduce their emissions.
Bypassing the problems of manual testing, Assentech has developed a software architecture that takes pressure measurements every 200 milliseconds referring to complex algorithms to make decisions on whether to change air flow to the valve during a test cycle. Thanks to the use of Machine Learning, the program will observe and analyze the valve behaviour in response to the changing flow, and build a performance report which will be used by another complex algorithm to determine the best method to measure the valve leak rate and use it to test the valve automatically.
Their system, Vent-Less, confirms the functionality of the vent – demonstrating that it’s safe to use and specified to the scope and requirements. This functionality check is important as the operator has confirmation that it is safe to install.
This technology is not only hugely beneficial for the environment - one example demonstrated a reduction in global warming potential from 160 tonnes per annum down to 158kg - but it brings immense benefits to all stakeholders, namely, reduced greenhouse gasses leaking to the environment, fewer costly evaporation losses for the operators, safer environment for staff and neighbours, better ESG monitoring and reduced manufacturing time for the producers.
With innovation aimed at maximising the potential of the technological capabilities within Cornwall, through cross-sectors collaboration and harnessing of existing assets, these businesses are a small snapshot of how Cornwall’s space, AI and deep learning industries are progressing hand in hand.