Autonomous Driving

Resources^{6 total}

• Project

  Meaningful Control of Autonomous Systems

  At the end, we want a system that interprets sensor input to estimate the uncertainty of the system, while asserting the situation while driving.

• Thesis

  Building motion-prediction models for self-driving vehicles

  Autonomous vehicles should be able to predict the movement of other traffic participants so they can act accordingly.

• Thesis

  Cheating by Segmentation

  This thesis describes a variation on the work from "Learning by Cheating" to use semantic segmentation data instead of bird's-eye view data

• Thesis

  Conditional imitation learning with pyramid perception modules

  In this thesis a driving agent is trained to predict waypoints directly from a forward facing camera, especially when it encounters a traffic light on route.

• Publication

  End-to-End Imitation Learning for Autonomous Vehicle Steering on a Single-Camera Stream

  Vehicles can follow roads based on a camera view, but this has to be done reliably, including in one of the most complex environments of the CARLA simulation.

• Publication

  Learning to Drive Fast on a DuckieTown Highway

  We train a small car to follow the road. In the real-world, roads do not always have the same appearance, so it is trained on the complete view.