Student Research Assistant in Bridge Monitoring, Computer Vision & Data Science

The Chair of Structural Mechanics & Monitoring
, led by Prof. Eleni Chatzi, conducts research at the intersection of structural engineering, sensing, data science, and artificial intelligence for infrastructure monitoring.

We are looking for a Student Research Assistant with a background in structural engineering, signal processing, data science, computer vision, or related fields to support our ongoing SwissBWIM project. The project combines bridge structural monitoring and traffic monitoring data to identify vehicle loads and assess the structural condition and performance of highway bridges.

Our project, “SWISS-BWIM:
Multi-functional bridge loads and condition assessment,” develops robust and cost-efficient methods for bridge load identification and condition assessment based on the Bridge Weigh-in-Motion (B-WIM) concept.

B-WIM uses sensors installed directly on bridges to estimate the loads of passing vehicles from the bridge response they induce. The same monitoring data can also provide information on structural behaviour and condition. By combining identified traffic loads with structural response, the project aims to derive performance indicators such as dynamic amplification factors, response envelopes, and load-response relationships, and ultimately support the identification of abnormal behaviour and potential damage.

We have completed monitoring-system installations on several highway bridges in Switzerland and are continuously collecting large-scale, multi-modal datasets, including bridge acceleration, strain and displacement measurements, environmental data, and traffic videos.

The Student Research Assistant will work with an analysis framework which has been developed by our Chair to process and integrate these datasets. Depending on background and interest, there will also be opportunities to develop and test new methods for traffic-load identification, computer-vision-based traffic monitoring, and structural health monitoring under the supervision of experienced researchers

The main tasks include:

• Multi-modal data preprocessing: synchronising measurements from different sensing systems, identifying and cleaning faulty data, and improving data organisation and storage.
• Traffic video analysis: applying deep-learning-based computer vision methods for vehicle detection and tracking; manually labelling data and fine-tuning models where needed; and using camera calibration and geometric transformation to determine vehicle locations and trajectories on the bridge.
• Bridge Weigh-in-Motion analysis: fusing vehicle trajectories with bridge acceleration, strain and displacement measurements to estimate traffic loads.
• Structural monitoring: analysing structural and environmental measurements to extract key performance indicators and investigate changes in bridge behaviour and structural condition.
• Integrated condition assessment: combining identified traffic loads with structural response indicators to support more comprehensive assessment of bridge performance and condition.

Please note that the methods required to complete most of the tasks above have already been developed by our Chair. Depending on the candidate’s background and interests, there will also be opportunities to develop and test new methods.

We are looking for a motivated student who can contribute to several of the areas above. Ideally, you:

• Are currently enrolled at a university in Zurich or the nearby area.
• Have a background in civil/structural engineering, mechanical engineering, data science, computer science, or a related field.
• Have experience or strong interest in two or more of…