PRoTECT: PRedicting The driving behavior under varying Environmental ConsTraints

Project Description

To enable modern cars to drive autonomously or highly automated is one of the most challenging undertakings of our time. To achieve this goal the cars need to comprise their environment and the present traffic situation as exact as possible, while meeting real-time constraints. But that is still not enough – additionally a sufficient prediction of the situation is needed to be able to plan a secure driving corridor. Current Advanced Driver Assistance Systems (ADAS) are already executing and involving such predictions. But most systems, which are currently in the market, are based on simple heuristics and use only the information directly provided by the sensor systems. At this point the PRoTECT project applies with the purpose to improve these prediction accuracies by incorporating also such information a human driver would conclude from the context of the situation and its knowledge thereof. That way, a human driver would e.g. expect a neighboring vehicle to keep a larger distance to its surrounding vehicles or generally speaking to drive more carefully in the case of a slippery road. In addition to that one has to take into account geo-located specialties as crowded roads in India or rectangular roads as in Manhattan. With the thereby improved knowledge about future situations, considerably improvements to the behavior of new ADAS generations in the sense of a humanized driving will become possible.






Project Team

Florian Wirthmüller
Ulm University, Institute of Databases and Information Systems
Daimler AG (Böblingen), Crowd Data & Analytics for Automated Driving
Dr. Jochen Hipp
Daimler AG (Böblingen), Crowd Data & Analytics for Automated Driving
Dr. Joachim Herbst
Daimler AG (Böblingen), Crowd Data & Analytics for Automated Driving
Prof. Dr. Manfred Reichert
Ulm University, Institute of Databases and Information Systems

Project Partners

Ulm University, Institute of Databases and Information Systems

Daimler AG (Böblingen), Crowd Data & Analytics for Automated Driving

Duration

The project was started in early 2018.

Publications

| 2021 | 2020 | 2019 | 2015 |

2021

Wirthmueller, Florian and Schlechtriemen, Julian and Hipp, Jochen and Reichert, Manfred (2021) Teaching Vehicles to Anticipate: A Systematic Study on Probabilistic Behavior Prediction Using Large Data Sets. IEEE Transactions on Intelligent Transportation Systems, IEEE. (Accepted for Publication) file
Wirthmueller, Florian and Klimke, Marvin and Schlechtriemen, Julian and Hipp, Jochen and Reichert, Manfred (2021) Predicting the Time Until a Vehicle Changes the Lane Using LSTM-Based Recurrent Neural Networks. IEEE Robotics and Automation Letters, 6(2): 2357-2364, 10.1109/LRA.2021.3058930. file

2020

Wirthmueller, Florian and Schlechtriemen, Julian and Hipp, Jochen and Reichert, Manfred (2020) Towards Incorporating Contextual Knowledge into the Prediction of Driving Behavior. In: 23rd IEEE International Conference on Intelligent Transportation Systems (ITSC 2020), Rhodes, Greece, 20-23 September 2020, IEEE, pp. 1-7. file
Wirthmueller, Florian and Klimke, Marvin and Schlechtriemen, Julian and Hipp, Jochen and Reichert, Manfred (2020) A Fleet Learning Architecture for Enhanced Behavior Predictions during Challenging External Conditions. In: IEEE Symposium Series on Computational Intelligence (SSCI 2020), Canberra, Australia, 1 - 4 December 2020, IEEE, pp. 2739-2745. file

2019

Wirthmueller, Florian and Hipp, Jochen and Sattler, Kai-Uwe and Reichert, Manfred (2019) CPD: Crowd-based Pothole Detection. In: 5th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2019), Heraklion, Crete, Greece, May 3-5, 2019, SciTePress, pp. 33-42. file

2015

Schlechtriemen, Julian and Wirthmueller, Florian and Wedel, Andreas and Breuel, Gabi and Kuhnert, Klaus-Dieter (2015) When will it change the lane? A probabilistic regression approach for rarely occurring events. In: IEEE Intelligent Vehicles Symposium (IV), Seoul, South Korea, June 28 - July 1, 2015, (): 1373-1379, IEEE Computer Society Press.