30 credits – Machine-Learning Based Load-Change Detection for Heavy Vehicles i Sodertalje

Scania is now undergoing a transformation from being a supplier of trucks, buses and engines to a supplier of complete and sustainable transport solutions.

Carrying out a thesis project at Scania is an excellent way of making contacts for your future working life. Many of our current employees started their career with a thesis project within the company.

Background:
Information about total weight and weight distribution (between the wheels) is used by many electronic systems in modern heavy vehicles. For example, the braking system uses this to distribute the correct braking effort at each wheel, making the mass estimation a safety critical functionality. Several sensors and algorithms are used to estimate the mass properties in different ways, but are not always quick enough when the truck is being loaded or unloaded, which can lead to unpredictable braking performance and hazardous incidents.

Assignment:
The project assignment is to develop a system that quickly detects or predicts changes in the vehicle’s weight, by using currently available sensors and information in Scania trucks. This system will have most value for vehicles that are exposed to large and frequent load changes, which often are vehicles operated in a cyclic manner. By analyzing location data (e.g. GPS) from these vehicles with driving pattern recognition and machine learning algorithms, the system should be able to predict when and where the vehicle load will change. If fused with sensor information (from e.g. accelerometers) and historical vehicle mass data (from other mass estimation functions), a good guess of the new vehicle weight and weight distribution is also possible.

The project will require a theoretical survey of applicable methods,

test design and planning, as well as extensive data processing and analysis.

Education:
Preferred experience and competence of applicants:

• Mechanical engineering, applied physics, electrical engineering, vehicle engineering, or similar.
• Machine learning and pattern recognition
• Signal processing
• Sensor fusion
• Scripting/programming (e.g. Matlab)

Number of students: 2
Start date: January-February 2019
Estimated time needed: 20 weeks