C2I 2022: Data & Connectivity shortlist

This year's Data & Connectivity shortlist showcased innovation around smarter energy meters, autonomous vehicles and railway crossing safety.

Project OPTIMUS utilises AI to improve railway crossing safety
Project OPTIMUS utilises AI to improve railway crossing safety

Project: Honeywell delivers industry-leading smart energy solutions faster using AI

Partners: Monolith AI with Honeywell

Honeywell is researching and developing new products like Smart Meters to measure energy consumption for residential and commercial properties, and also solutions for methane emissions monitoring.

They collaborated with engineers at Monolith to explore how advanced machine learning methods could be used in engineering workflows to understand and predict complex product behaviours faster and in extreme operation conditions.

They selected Monolith's software because its no-code environment was tailored to engineers and was particularly easy to learn. Within days of using the technology on test data they found new insights that delivered real business value.

Through the addition of pressure and temperature sensors to the residential gas flow meter, the Honeywell Smart Meter has an added benefit of being able to autonomously shut down if gas-grid over-pressure is detected, enhancing safety for millions of customers.
Prior to investigating AI, Honeywell followed a traditional and familiar development process that involved physically and virtually testing many different product variables.

This process would typically take 18 months to ensure the calibration error was below the legally required one per cent. With AI they could achieve this 25 per cent faster.

Using a machine learning solution has an additional benefit: it will continue to learn and keep improving after being deployed.

Project: Europe's first zero-occupancy autonomous vehicle journey on-road completed by Oxbotica & Applied EV

Partners: Oxbotica and Applied EV

Oxford-based Oxbotica develops software for vehicle autonomy and has successfully demonstrated its technology in vehicles that were built for people to drive.

But what about truly autonomous vehicles (AVs) that require no in-vehicle driver intervention and can be acquired by customers who require such vehicles on a 24/7 basis for a range of tasks? This prompted Oxbotica to seek a truly zero-occupancy AV and were soon introduced to Applied EV via a mutual mining connection.

Bayswater North, Australia-based Applied EV has developed the Blanc Robot, a programmable, 'application-ready' platform operated via the company’s Digital Backbone software and remote control. What they wanted, however, was full autonomy and a collaboration was born.

In May 2022 the Blanc Robot, driven by Oxbotica Driver, Oxbotica’s full stack autonomy system, debuted on a public road in Oxford. This was the first safe and sustainable deployment of a zero-occupancy, fully autonomous, new-type electric vehicle on publicly accessible roads in Europe, and the first of its kind to have a registration plate.

Applied EV CEO Julian Broadbent described the collaboration as ‘a 1+1 = 3 scenario - the perfect chance to show how two companies together stand a better chance of solving a problem than either on its own.’

Project: OPTIMUS

Partners: Leicester University, Synoptix and Innovate UK KTN

Safety at railway crossings looks set to be greatly enhanced following a collaboration between an interdisciplinary team at Leicester University, systems engineering specialists Synoptix, and an Innovate UK KTN.

There are 5887 level crossings across the UK's rail network and in 2019 alone, there were 181 incidents and two fatalities. To counter this, Network Rail conducts data gathering of level crossing use in a process called 'census'.

Census data is the key input into risk models used to assess technological interventions that can improve safety at level crossings, but due to the number and distribution of level crossings, it is an incomplete picture as census activities are carried out periodically. The data generated is then scaled to a national level, using assumptions. However, this method of collecting and processing census data is prone to inaccuracies.

To counter this, the collaborating partners developed OPTIMUS, an AI-based automated census device built around the concept of "all on the edge" that allows for continuous census for long time periods with no gaps, allowing Network Rail to obtain an accurate picture of traffic at individual level crossings.

In 2022 the team began comparing the accuracy and speed of OPTIMUS to a human completing the same task.