Imagine a self-driving car that can 'see' as clearly in a heavy blizzard or thick fog as it does on a sunny day. This project builds a brain for cars that combines different types of sensors—like heat-sensing cameras and radar—to make sure the vehicle doesn't get confused by rain or snow. It's like giving a car a set of high-tech goggles and a smarter instinct for when to slow down in bad weather.
Autonomous vehicles currently struggle in rain, fog, and snow, leading to safety risks and limited adoption. Most systems are trained for clear weather, making them unreliable in the unpredictable conditions required for global commercial rollout.
A weather-aware perception and decision-making system. This includes high-fidelity digital twins, sensor noise models, and a data readiness level (DRL) toolset to quantify image and point cloud quality.
If you are a vehicle OEM dealing with safety failures in snow or rain — this project developed a weather-aware perception and decision-making system that reaches TRL 7. This allows your autonomous platforms to operate reliably in harsh environments.
If you are a fleet operator dealing with route shutdowns during extreme weather — this project developed a robust sensor fusion system including LiDAR and thermal cameras. This ensures your trucks can navigate safely through fog and snow without human intervention.
If you are a transit authority dealing with accidents involving pedestrians in low visibility — this project developed improved object and person classification for vulnerable road users. This increases the safety of autonomous shuttles in complex urban weather.
Based on available project data, specific pricing or implementation costs are not provided.
Yes, the project aims to integrate solutions into OEM platforms to reach TRL 7, indicating a high level of industrial readiness.
Based on available project data, specific licensing terms are not mentioned, though the project involves 19 partners including 9 industry members.
The project uses the ISO 34503 taxonomy to define operational design domains, ensuring alignment with international standards.
The system is designed for integration into OEM platforms using a cost-effective multisensory setup and X-in-the-loop testing.
The consortium is heavily weighted toward commercial application, with 9 industry partners (47% industry ratio) and 4 SMEs. This balance, combined with 5 universities and 4 research institutes across 7 countries, suggests a strong pipeline from theoretical research to industrial deployment. The inclusion of specialized testing infrastructure, such as Arctic tracks and wind tunnels, provides a significant competitive advantage for validation.
Contact Hogskolan i Halmstad in Sweden
Contact us to explore licensing opportunities for weather-aware AI perception systems.
