RailLoc meets and exceeds Network Rail’s updated positioning standards

What does Network Rail’s updated positioning standard state, what has changed and why is RailLoc different to GNSS and Inertial based systems?

  • Each measurement point must have a longitudinal accuracy of 1m at least 99% of the time

  • Each measurement point must have a longitudinal accuracy of 30mm at least 68.2% of the time

  • Correct track identification shall be provided at least 99.5% of the time

This standard aims to ensure that the vast majority of the data is not only within that 1m range (this goes hand in hand with having two thirds of data in the +/-30mm range – shown in a normal distribution) – but that it is also attributed with the correct track. This is an important distinction, this means that a data point that is geospatially accurate to 1m, may not meet the standard – if the track is attributed incorrectly - and this is something that RailLoc solves that traditional GNSS/Inertial system don’t.

RailLoc answers yes to all three of these questions. The previous standard requires +/- 1m with variation only by permission (https://www.railinfrastructuremonitoring.co.uk/reports - “What do our customers want?”), the change to the standard makes the requirements explicit

Ultimately, having a positioning system that produces data to meet these requirements means that so much ambiguity and complexity can be removed from downstream processes.

This new standard gives clear and hard boundaries on what should be achieved, this helps companies like NR choose their solution based on evidence and not a spec sheet (terms like ‘95% confidence’ or ‘in clear open sky’ state what can be achieved, not what shall be achieved).

At Machines With Vision, we ensure that RailLoc data matches the required outcome - not just a possible outcome in optimum conditions.

For more information, please get in touch.

Previous
Previous

Looking ahead to ‘The Rise of IoT and Big Data in Rail 2024’

Next
Next

Omnicom Balfour Beatty and MWV join forces for a case study on PLPR (Plain Line Pattern Recognition)