The digital transformation of traditional inspection for pavement and lane markings must be aligned with the requirements of ADAS systems in vehicles and ensure frequent checks.
The maintenance of the road surface takes a big share of the costs of a road during its long years of operation: reparation and paint. Keeping the road surface in good condition is not only for making driving more comfortable: in well-kept roads, vehicles will use less fuel, and drivers will have less run-off-road accidents if the lane markings are visible.
Early detection, when minor interventions can significantly delay the need for costly mayor works. But manual early detection requires trained personnel regularly doing exhaustive and detailed visual inspections with uniform criteria. This is costly, and unattainable for most of the road network, especially secondary roads, which happen to be the most dangerous ones.
For this application, AI technology is ground-breaking: ASIMOB’s Autonomous Road Inspector can automatically inspect every centimetre of road, checking for the relevant parameters: location and identification of different types of damage, evaluation of its size and the length of the lane with defects or their degree of severity… as well as the visibility of the lane markings. A very lightweight installation in a vehicle (normally from the maintenance fleet), that will drive along the road can gather the data, which is analysed to provide useful and detailed information about the condition of the surface for whole networks in a record time.
To provide useful information regarding the damage in the road surface, these analysis of the data can be designed to check against the usual indexes defined by technical bodies that road maintenance and inspection teams use, such as Global Index (IG) of Belgium’s Centre de Recherches Routières (CRR), or the Road Condition Index (IC) of the Spanish Committee for Low-Volume Roads. Automation enables detailed and exhaustive inspections, avoiding heterogeneous evaluations (quite common in manual inspections) and systematically applying the designed criteria.
Lane markings are essential for guiding drivers, especially at night, and also for the good performance of the LKAS (Lane Keeping Assistance), a common feature of the current ADAS Systems of new vehicles.
The standard measurement performed on lane markings is the reflectometry, which measures the quality of the marking’s paint. However, reflectometry measurements do not take into account some critical factors: the problem of the visibility of the lines goes beyond the quality of the paint itself. The colour of the pavement next to the lane marking, the colour of the paint itself, or the presence of water, snow, ice or other elements on the pavement, can change the perception and visibility of the lines for drivers and LKAS. The perception of these lines is not the same during the day in sunny conditions, at night, or in times of rain. The reflectometry measurement provides a fixed number, but the perception of the line changes in different conditions.
Latest News
- Lima Expresa Introduces an AI Accident Detection System to Improve Safety on its Roads
- New Guide to Empower Creation of Road Safety Data Coalitions in Africa
- Newsletter – Issue Nº 5 2023
- Join the IRF at the PIARC World Road Congress 2023
- Successful Crash Investigation Course Hosted in Qatar
- ASIMOB’s Lightweight AI Solution for Road Pavement Inspections: A Game-changer on Road Safety and Efficiency
