Experimental development of novel computer vision algorithms that will lead to a fully automated and autonomous system for inspection, managing and reporting defects temporally on Great Britain's railways.

Automating today's manual processes associated with access right management and spot bidding and wrapping these digital processes in an intuitive, integrated, modern, bespoke and scalable user system. The benefits of this innovation are not only cost efficiency but also enabling a much better experience to freight customers to drive modal shift.

During phase 2, Hack Partners Limited will build on the IoT sensor and geospatial data model developed in phase 1, to develop a prototype product (which houses the phase 1 research and development) and test and validate this prototype in the real-world. End users will have a tool to help: * Better understand how delays propagate throughout the network and the most probable causes of delays * Discover where capacity can be increased using practical data not theoretical

Public description We will use IoT sensor information and geospatial data to create a technology that will detect train movement at a more granular level. As a result, we will be able to understand in more detail where trains are on the rail network. This can technology can be used to: Better understand how delays propagate throughout the network Discover where capacity can be increased using practical data not theoretical Detect how leaf fall can impact performance Understand how speed restrictions impact performanc

Hubble is an artificial intelligence assisted lineside inspection and maintenance planning solution helping the rail industry add more resilience to infrastructure and operations. Starting with lineside inspections for vegetation growth which continues to be an urgent priority for the railway leading to: - Delays and cancellations to passenger and freight services resulting in less available capacity - High maintenance costs which ultimately are passed back to operators, the tax payer, and passengers - An overall poor customer experience for the passenger and freight operators - The increased utilisation of front-line engineers which must travel by van to perform trackside inspections Current methods for addressing vegetation management are either too slow and labour intensive increasing costs and risk; or automated but very expensive (ORBIS' LiDAR) and do not focus how route level users makes use of the data resulting in a situation where asset manangers are "data rich, but insight poor". Both approaches as a result are not used very often leading to poor quality information on when and where vegetation management should occur. This results in the infrastructure and operational resilience of the railway to be susceptible to adverse environmental and future weather conditions and the availability of the railway not being at its maximum due to inaccurate maintenance plans. Hubble uses artificial intelligence which follows the latest Vegetation Management Standards produced by Network Rail to addresses multiple problems associated with vegetation management, namely the lack of: - Decision support tools which have clear geographical mapping of vegetation throughout the network - Integrated workflow management system allowing for better maintenance planning - An accurate, reliable, and integrated objective method for identification of non-compliant vegetation. By identifying which vegetation poses a risk to the resilience of rail infrastructure and operations Hubble ensures that the GB Rail remains a point of pride and economic growth for the nation, one hedge at a time.