Master’s Thesis – Data-driven traffic modelling for CV simulation

Background


At the Customer Feature Data Analytics group, we lead transformation towards customer centric data-driven V&V. Our purpose is to enable fact-based decisions based on data and to make sure Volvo Trucks lives up to and exceeds customer expectations on features through developing and using data analytic methods.
We support all types of features, among them, features impacting the vehicle fuel/energy economy, CO2 emission as well as the vehicle performance and drivability. For this reason, we work very closely with Driving Performance & Aero group which defines, verifies, and validates the customer attributes with the help of complete vehicle simulation and vehicle testing in order to deliver the right products and services to all Volvo Group brand customers.
Due to global transition, the transport sector is currently undergoing large adaptation and change in technology. This also requires current methods and development processes to be updated to capture the most important factors affecting the customer features.
New propulsion systems, e.g., Battery electric vehicles, will represent a big shift in the transportation industry and logistics operators, truck owners and drivers will need to adapt to eventual differences in daily operation, such as recharging which might differ from the traditional refueling routine.
Traffic conditions in combination with driver interactions will be a key factor in the vehicle range and therefore also in the recharging strategy to be adopted for optimal operation.
The purpose of this master thesis work is to support traffic modelling in the simulation environment by analyzing truck data. The goal is to develop a method that can be used in the future to define the most representative traffic scenarios for a given road.


Proposed work outline
The thesis could include, but not be limited to:
Literature study of existing traffic modelling techniques used for complete vehicle simulation purposes
Benchmark of existing traffic modeling techniques with the in-house traffic model
Data pre-processing and time-series analysis
Statistics and modelling of traffic scenarios based on the analysis of a set of customer trucks.

Suitable background
We expect that you are finishing your studies in Computer Science, Applied Physics or similar. Background in Mechanical Engineering, Automotive Engineering or equivalent is a plus. We believe you have a technical interest in the automotive industry. You should be familiar with Python. Knowledge and experience in time-series analysis and PySpark coding is a plus.


Proposed start date is Feb-Mar 2023.
Physical location: Volvo GTT (Group Truck Technology), Lundby, Göteborg

Thesis Level: Master

Language: English


Starting date: March 2023


Number of students: 1/2


Tutor
Anna Marti Bigorra, Specialist Data Scientist, anna.marti.bigorra@volvo.com


Kindly note that due to GDPR, we will not accept applications via mail. Please use our career site.