|
|
| Arnoud's | Research |
| Education |
| Activities |
| Projects |
| Publications |
Meaningful Control of Autonomous Systems
Description
The MCAS project is a collaboration of four partners to study the impact of autonomous systems on the society, especially the adaption of systems based on machine learning. The four partners are:- the national research institute for mathematics and computer science (CWI)
- the Netherlands Organisation for applied scientific research (TNO)
- Universiteit van Amsterdam - Faculty of Science
- Universiteit van Amsterdam - Faculty of Law
America’s Power Companies’ advertisement from 1956
depicting a future with autonomous cars.
Inside the project I will work on the use case of Automated and Cooperative Driving.
One can think of topics like semantic segmentation of video data for road geometry estimation, traffic and behavior prediction of other road users, estimating driver's attention, among others.
As part of this research, our team participated in the Waymo 2D Detection Challenge and AI Driving Olympics.
The IAS-17 publication will be presented on the Netherlands Conference on Computer Vision (NCCV 2022) with this poster:
Publications
- Felix Hoekstra, Unsupervised Semantic Segmentation for Autonomous Driving Cars, Bachelor thesis, Universiteit van Amsterdam, July 26, 2024.
- Yue Cao, Wei Shangguan, Arnoud Visser, Junjie Chen, Linguo Chai and Baigen Cai, "Segment-Based Trajectory Prediction and Risk Assessment for RSU-assisted CAVs at Signalized Intersections", in IEEE Transactions on Intelligent Vehicles, June 13, 2024.
- Marina Orozco Gonzalez, Ricardo Chavez Torres, Danilo Toapanta Barahona, Robin Reitsma, Gijs de Jong, Harold Ruiter, Advancing Occupancy and Flow Estimation for Autonomous Driving: Insights from SparseOcc, OpenDriveLab's Autonomous Grand Challenge 2024 - Occupancy and Flow track, Technical Report CV2 project, Universiteit van Amsterdam, June 13, 2024
- Lucas Ponticelli, Taiki Papandreou-Lazos, Erik Meijer, Bart de Zwart, Luuk Versteeg, Tobie Werner, Reproducibility Study Of LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving, OpenDriveLab's Autonomous Grand Challenge 2024 - Mapless Driving track, Technical Report CV2 project, Universiteit van Amsterdam, June 12, 2024
- Midas Amersfoort, Michael Dubbeldam and Arnoud Visser, Improving multi-object re-identification at night with GAN data augmentation, in Intelligent Autonomous Systems 18, Lecture Notes in Networks and Systems (LNNS, volume 795), Springer, pp. 481-493, April 25, 2024.
- Jasper Mulder, Intersection Navigation utilizing Bird’s Eye View representations from 3D Image Features, Bachelor thesis, Universiteit van Amsterdam, March 3, 2023.
- Maël Wildi, Alexandre Alahi and Arnoud Visser, Training traffic light behavior with end-to-end learning, Intelligent Autonomous Systems 17, Lecture Notes on Networks and Systems, vol. 577, pp. 753-764, January 2023
- Midas Amersfoort, Improving FlowCube’s re-identification performance with GAN data augmentation (extended abstract), Master thesis, Vrije Universiteit Amsterdam, August 2022.
- Eric R. Langezaal, Learning to recognize illegible license plates with 3D rendered images, Bachelor thesis, Universiteit van Amsterdam, July 2022.
- Sebastiaan Aflaki, Teaching a Virtual DuckieTown Agent to Stop, Bachelor thesis, Universiteit van Amsterdam, April 2022.
- Thomas Wiggers and Arnoud Visser, Learning to drive fast on a DuckieTown highway, Intelligent Autonomous Systems 16, Lecture Notes in Networks and Systems, vol 412. Springer p. 183-194, April 2022.
- Thomas van Orden and Arnoud Visser, End-to-end Imitation Learning for Autonomous Vehicle Steering on a Single Camera Stream, Intelligent Autonomous Systems 16, Lecture Notes in Networks and Systems, vol 412. Springer p. 212-224, April 2022.
- Maël Wildi, "Conditional imitation learning with pyramid perception modules", Master thesis, École Polytechnique Fédérale de Lausanne, January 2022.
- Thomas van Orden, Cheating by Segmentation, Bachelor thesis, Universiteit van Amsterdam, June 2021.
- Imre Fodi, "Building motion-prediction models for self-driving vehicles", Bachelor thesis, Universiteit van Amsterdam, February 2021.
- Jan-Pieter Paardekoper, Arnoud Visser, Eric Pauwels et al, Meaningful Control of Autonomous Systems, MCAS Project - Final report, February 2021.
- Jan-Pieter Paardekoper, Arnoud Visser, Eric Pauwels et al, Meaningful Control of Autonomous Systems, MCAS Project - Intermediate report, September 2020.