MSc Thesis: Robust MPC for Autonomous Vehicles in Uncertain Situations

More than one million people die in traffic accidents every year. One solution to improve traffic safety could be to develop autonomous vehicles (AVs) that avoid dangerous human driving behaviors. However, current AVs still struggle to obtain a robust performance in the real world. Naturally, measurements, and estimations of the own-vehicle (ego-vehicle) and the surrounding vehicles are inherently uncertain, causing many additional challenges with regards to safety. Addressing these challenges with a robust control design is of utmost importance for any real-world application.


An appealing approach to address safety is to plan a trajectory using Model Predictive Control (MPC). These approaches have gained attention due to the ability to enforce rigorous constraints on the ego-vehicle and collision avoidance. Recent work has shown promising results, even in more realistic simulation environments (such as CARLA). However, most recent work lacks rigorous attention to the inherent uncertainties in the environment. Hence, dangerous scenarios can still be encountered in edge-cases, posing a considerable safety hazard for a potential practical application. One promising solution could be to address these issues with Robust MPC.


The purpose of this thesis can be summarized as:


Construct safety-critical scenarios for a Heavy-Vehicle in the CARLA simulator.


Design and implement a Robust Model Predictive Controller that considers the uncertainties related to the ego-vehicle state and dynamics.


Design and implement a Robust Model Predictive Controller that can tackle uncertainties of the surrounding vehicles.


(Optional) A stretch goal for the project is to condense the thesis into a conference paper.


The aim of this thesis is to use robust MPC to solve problems within autonomous driving. The work will include control design, vehicle modelling, sensor analysis and programing. This work will be carried out together with Volvo Group Trucks Technology. The thesis is recommended for one or two students with a strong background in control, optimization, and Python/C++ with a solid mathematical background. Prior experience with vehicle modelling and simulation is meritorious.


The master thesis work is intended to take place January-June 2024.


If you are interested in the proposal, please apply with CV and grades. For questions, please contact Leo Laine +46-76-5535311.


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