TY - CHAP A1 - Meess, Henri A1 - Gerner, Jeremias A1 - Hein, Daniel A1 - Schmidtner, Stefanie A1 - Elger, Gordon T1 - Reinforcement Learning for Traffic Signal Control Optimization BT - A Concept for Real-World Implementation T2 - AAMAS '22: Proceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems KW - Multi-Agent Reinforcement Learning in real-world KW - MARL KW - traffic optimization KW - multimodal traffic KW - DRL Y1 - 2022 UR - https://dl.acm.org/doi/10.5555/3535850.3536081 SN - 978-1-4503-9213-6 SP - 1699 EP - 1701 PB - International Foundation for Autonomous Agents and Multiagent Systems CY - Richland ER - TY - CHAP A1 - Gerner, Jeremias A1 - Rössle, Dominik A1 - Cremers, Daniel A1 - Bogenberger, Klaus A1 - Schön, Torsten A1 - Schmidtner, Stefanie T1 - Enhancing Realistic Floating Car Observers in Microscopic Traffic Simulation T2 - 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC) UR - https://doi.org/10.1109/ITSC57777.2023.10422398 Y1 - 2024 UR - https://doi.org/10.1109/ITSC57777.2023.10422398 SN - 979-8-3503-9946-2 SP - 2396 EP - 2403 PB - IEEE CY - Piscataway ER - TY - CHAP A1 - Meess, Henri A1 - Gerner, Jeremias A1 - Hein, Daniel A1 - Schmidtner, Stefanie A1 - Elger, Gordon ED - Kieu, Minh Le ED - Dam, Koen H. van ED - Thompson, Jason ED - Malleson, Nick ED - Heppenstall, Alison ED - Ge, Jiaqi T1 - Real World Traffic Optimization by Reinforcement Learning: A Concept T2 - International Workshop on Agent-Based Modelling of Urban Systems (ABMUS) Proceedings: 2022 UR - https://doi.org/10.6084/m9.figshare.19733800.v1 Y1 - 2022 UR - https://doi.org/10.6084/m9.figshare.19733800.v1 UR - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:573-29290 SP - 49 EP - 54 PB - figshare CY - [s. l.] ER - TY - INPR A1 - Rößle, Dominik A1 - Gerner, Jeremias A1 - Bogenberger, Klaus A1 - Cremers, Daniel A1 - Schmidtner, Stefanie A1 - Schön, Torsten T1 - Unlocking Past Information: Temporal Embeddings in Cooperative Bird’s Eye View Prediction N2 - Accurate and comprehensive semantic segmentation of Bird's Eye View (BEV) is essential for ensuring safe and proactive navigation in autonomous driving. Although cooperative perception has exceeded the detection capabilities of single-agent systems, prevalent camera-based algorithms in cooperative perception neglect valuable information derived from historical observations. This limitation becomes critical during sensor failures or communication issues as cooperative perception reverts to single-agent perception, leading to degraded performance and incomplete BEV segmentation maps. This paper introduces TempCoBEV, a temporal module designed to incorporate historical cues into current observations, thereby improving the quality and reliability of BEV map segmentations. We propose an importance-guided attention architecture to effectively integrate temporal information that prioritizes relevant properties for BEV map segmentation. TempCoBEV is an independent temporal module that seamlessly integrates into state-of-the-art camera-based cooperative perception models. We demonstrate through extensive experiments on the OPV2V dataset that TempCoBEV performs better than non-temporal models in predicting current and future BEV map segmentations, particularly in scenarios involving communication failures. We show the efficacy of TempCoBEV and its capability to integrate historical cues into the current BEV map, improving predictions under optimal communication conditions by up to 2% and under communication failures by up to 19%. The code will be published on GitHub. UR - https://doi.org/10.48550/arXiv.2401.14325 Y1 - 2024 UR - https://doi.org/10.48550/arXiv.2401.14325 UR - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:573-46123 PB - arXiv CY - Ithaca ER - TY - CHAP A1 - Schlamp, Anna-Lena A1 - Gerner, Jeremias A1 - Bogenberger, Klaus A1 - Schmidtner, Stefanie T1 - User-Centric Green Light Optimized Speed Advisory with Reinforcement Learning T2 - 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC) UR - https://doi.org/10.1109/ITSC57777.2023.10422501 Y1 - 2024 UR - https://doi.org/10.1109/ITSC57777.2023.10422501 SN - 979-8-3503-9946-2 SP - 3463 EP - 3470 PB - IEEE CY - Piscataway ER - TY - JOUR A1 - Meess, Henri A1 - Gerner, Jeremias A1 - Hein, Daniel A1 - Schmidtner, Stefanie A1 - Elger, Gordon A1 - Bogenberger, Klaus T1 - First steps towards real-world traffic signal control optimisation by reinforcement learning JF - Journal of Simulation N2 - Enhancing traffic signal optimisation has the potential to improve urban traffic flow without the need for expensive infrastructure modifications. While reinforcement learning (RL) techniques have demonstrated their effectiveness in simulations, their real-world implementation is still a challenge. Real-world systems need to be developed that guarantee a deployable action definition for real traffic systems while prioritising safety constraints and robust policies. This paper introduces a method to overcome this challenge by introducing a novel action definition that optimises parameter-level control programmes designed by traffic engineers. The complete proposed framework consists of a traffic situation estimation, a feature extractor, and a system that enables training on estimates of real-world traffic situations. Further multimodal optimisation, scalability, and continuous training after deployment could be achieved. The first simulative tests using this action definition show an average improvement of more than 20% in traffic flow compared to the baseline – the corresponding pre-optimised real-world control. UR - https://doi.org/10.1080/17477778.2024.2364715 Y1 - 2024 UR - https://doi.org/10.1080/17477778.2024.2364715 UR - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:573-48585 SN - 1747-7778 PB - Taylor & Francis CY - London ER -