Examenspresentation av Richard Johnsson och Oskar Westerlund

Opponents: Oskar Hurtig and Marcus Lindohf
Supervisors: Erik Börve Erik at AB Volvo

Abstract:
In this thesis the area of robust trajectory planning through model predictive control in combination with trajectory predictions is explored. The combination is used to obtain an autopilot algorithm controlling a truck that is operating in a highway scenario with surrounding traffic. A novel transformer-based observer has been developed to address the problem of generating accurate trajectory predictions for the robust model predictive controller, capable of expressing uncertainty through the Monte Carlo dropout methodology. The trajectory predictions together with the uncertainty estimations from the observer are later combined with the model predictive controller to increase the robustness of the planned trajectory. The results indicate a move towards more preferable performance in safety-critical situations.

Welcome!

Oskar, Marcus and Nikolce