TITLE:
J7-50096 - Improving B-WIM performance with big data and Artificial Intelligence BrAId
DESCRIPTION:
The increasing number of heavy commercial vehicles, such as trucks and buses, along with new transportation solutions like road trains and electronically controlled convoys, impacts the safety and management of road infrastructure. Overloaded vehicles in particular cause issues as they increase safety risks and accelerate road damage. Vehicle weight control is crucial to prevent illegal road use.
Traditional weighing methods, such as static weighing, are accurate but expensive and inefficient for heavy traffic. Therefore, systems for weighing vehicles while driving (WIM) are used. However, installing most WIM systems is problematic and can damage the road surface. An alternative is the B-WIM bridge system, which does not require cutting the road or road closures, but currently, its accuracy and reliability do not meet legal standards for penalizing violators.
B-WIM systems are based on measuring bridge deformations, which causes errors, especially on longer bridges or with multiple axles on the bridge. Reliable measurement of axle loads remains an open question.
The proposed project aims to address these shortcomings using artificial intelligence (AI). The project's objectives include developing methods to evaluate the reliability of B-WIM measurements with AI, improving measurement accuracy by combining data from deformation measurements, B-WIM results, and traffic camera photographs, and creating a public B-WIM data repository for further research. This approach could significantly improve the accuracy and reliability of B-WIM measurements and open a new field of research in this area.
Traditional weighing methods, such as static weighing, are accurate but expensive and inefficient for heavy traffic. Therefore, systems for weighing vehicles while driving (WIM) are used. However, installing most WIM systems is problematic and can damage the road surface. An alternative is the B-WIM bridge system, which does not require cutting the road or road closures, but currently, its accuracy and reliability do not meet legal standards for penalizing violators.
B-WIM systems are based on measuring bridge deformations, which causes errors, especially on longer bridges or with multiple axles on the bridge. Reliable measurement of axle loads remains an open question.
The proposed project aims to address these shortcomings using artificial intelligence (AI). The project's objectives include developing methods to evaluate the reliability of B-WIM measurements with AI, improving measurement accuracy by combining data from deformation measurements, B-WIM results, and traffic camera photographs, and creating a public B-WIM data repository for further research. This approach could significantly improve the accuracy and reliability of B-WIM measurements and open a new field of research in this area.
DURATION:
2023/10/01 - 2026/09/30
FUNDING:
ARIS
MORE:
