Geomatics: AI at the service of automotive tests

approach

Historically, the creation of these routes — combining specific percentages of urban, rural and highway driving, with slope constraints — was carried out manually, a time-consuming and complex process. The core of the Geomatics project lies in the development of an algorithm capable of automatically generating paths, responding precisely to the instructions provided by the engineering teams.

GEOGRAPHIC DATA

Contains accurate map information from OpenStreetMap, used to build the route and define the areas crossed.

Elevation data

Integrate slopes and terrain from Aster data, essential for modeling the effort and energy spent on the course.

User criteria

Includes parameters defined by the user: starting point, total distance, city/road/highway proportions, and maximum slope threshold.

innovation

THEAI AT THE SERVICE OF AUTOMOTIVE TESTS

The Geomatics project demonstrates how artificial intelligence transforms a complex manual process into a reliable, efficient and repeatable digital solution — perfectly adapted to the needs of engineers.

Thibaut FRANCE

Head of AI division

methodology

This project required advanced engineering to manipulate large volumes of geographic data and to design a robust algorithm capable of meeting performance and accuracy requirements. The complexity lay not only in calculating the routes, but also in the ability to generate a diversity of valid routes.

Openstreetmap data processing

Acquisition and processing of a large volume of raw data to generate the road graph around the starting point specified by the user.

ENRICHMENT OF THE GRAPH

Addition of key information: road category (city, country, highway), slopes, speed limits — with adaptation to the specificities of each country.

CLEANING AND SIMPLIFICATION

Extensive work to ensure the consistency of the routes (elimination of U-turns, dead ends, cul-de-sac). An essential step in obtaining reliable and varied courses.

PATH GENERATION ALGORITHM

Intelligent selection of points distributed on the map, calculation of the shortest paths between them and construction of a simplified graph, greatly reducing the complexity of the problem.

prospects