The tedious and repetitive process of measuring and assessing track condition and the lack of generational exchange among the aging skilled expert personnel is generating ever-increasing costs for maintaining rail infrastructure.
The lack of emergence of new revolutionary solutions and decisive response of infrastructure managers to this problem will cause further aggravation of the problem and high financial losses in the near time frame.
The device, attached to existing rolling stock, allows measuring and assessing the condition of railroad infrastructure during operational rolling stock runs.
Created for infrastructure managers and maintenance entities, it has been designed to be easily installed on operational wagons, locomotives, or diagnostic vehicles. Automatically identifies more than 15 different types of defects on rolling surface, fastenings, sleepers and track bed, such as sleeper cracks, loss of ballast, which are impossible to detect using alternative non-manual assessment methods.
The system is able to acquire images and measurements from independent cameras and scanners for the complete inspection of the track infrastructure with speeds up to 120 kph. An advanced AI algorithm is processing the data recognizing the possible defects with a machine learning model based approach.
Performing track inspections using machine vision instead of manual inspections, significantly simplifying the planning and execution of rail infrastructure maintenance processes with less personel engagement.
Measuring up to 400 km of railway tracks per day (depending on the rolling stock used) – 40 times more than with manual inspection in the same time frame.
Introducing an unprecedented business model that strongly relieves the customer of the personnel and monetary expenses associated with acquiring, maintaining, and operating the system.
Unification of expert knowledge while saving and storing the volatile expertise and replacing it with a digital memory, effecting in remarkably repeatable and reliable defect detection and evaluation with up to 96% accuracy.
Dedicated streamlined application that allows interaction with the system for AI prediction refinement and generating digital reports on the condition of rail infrastructure from preset measurement sections.
Easily adaptable to different types of rolling stock and specific needs of the customer, allowing easy expansion of the system with new functionalities (extension of the analyzed area, creating digital twins, BIM, etc.).
Provide reports from measurements of inspected infrastructure on customer demand in no more than one business day.
