How can road damage be detected before it becomes a problem? Through the InfraSENTIC research project, the BERNARD Gruppe is working with its project partners to develop a new approach to intelligent and continuous infrastructure monitoring.
The BERNARD Gruppe is part of the InfraSENTIC research project, funded by the FFG. The aim of the project is to monitor the condition of roads continuously, in a data-driven and efficient manner in future. Rather than relying on time-consuming individual measurements or periodic inspections, near-production vehicles will act as mobile sensor nodes.
Cameras and vehicle sensors collect anonymised condition data during normal driving, which is processed directly within the vehicle. This creates a robust basis for detecting damage at an early stage and planning targeted maintenance measures. The BERNARD Gruppe is contributing its expertise, in particular, to the development of the system architecture and the subsequent application of the overall system. This includes the selection of suitable pilot regions, the analysis of relevant damage and hazard scenarios, and the development of a probabilistic digital road twin. This[MB1] ‘twin’ combines crowd-sourced data, sensor data, and geo- and weather information, enabling the condition and risk evaluation of infrastructure sections. Furthermore, the BERNARD Gruppe is supporting the assessment of the system in real-world vehicle operations and is developing a scalable product module for infrastructure operators, public authorities and fleets. The intelligent integration of edge AI, vehicle data and environmental information creates a new approach to infrastructure management. Critical road sections can be identified at an early stage, maintenance measures efficiently prioritised and the service life of the infrastructure sustainably extended. In this way, the research project supports the transition from reactive to condition- and risk-based maintenance of transport infrastructure.
Felix Laimer, BERNARD Gruppe
Picture: © Vertical Vehicle
Funded by:
