[DFG/ANR project MADRAS] Multi-agent modelling of dense crowd dynamics: Predict & Understand

Publications

All the MADRAS project publications and communications are registered on this open access HAL repository.

Peer-reviewed journals

A generalized trajectories-based evaluation approach for pedestrian evacuation models. Safety Science, 2022.
Heterogeneity‑induced lane and band formation in self‑driven particle systems. Scientific Reports, 2022.
Review of pedestrian trajectory prediction methods: comparing deep learning and knowledge-based approaches. IEEE Transactions on Intelligent Transportation Systems, 2022.
Anticipating Collisions, Navigating in Complex Environments, Elbowing, Pushing, and Smartphone-Walking: A Versatile Agent-Based Model for Pedestrian Dynamics. Preprint available here, 2022.

Peer-reviewed contributions to conference proceedings

Time-to-collision models for single-file pedestrian motion. Collective Dynamics (PED2021 conference), 2021.
Initiating lane and band formation in heterogeneous pedestrian dynamics. Collective Dynamics (PED2021 conference), 2021.

Communications in conferences

Differences in pedestrian trajectory predictions for high- and low-density situations. Traffic and Granular Flow 2022, 15 – 17 October 2022, IIT Delhi, India. Abstract.
Modélisation de l’anticipation des collisions et des contacts dans la dynamique des foules piétonnes. 18ième Journées de la Matière Condensée (JMC18), 22 – 26 August 2022, Lyon, France. Abstract.
Stop and go pedestrian waves based on a behavioural model. 18ième Journées de la Matière Condensée (JMC18), 22 – 26 August 2022, Lyon, France. Poster.
Noise-induced breakdown in single-file motion. 18ième Journées de la Matière Condensée (JMC18), 22 – 26 August 2022, Lyon, France. Abstract; Poster.
Gampy: a fast plugin for integration of Python-based deep-learning models to the GAMA platform. 2nd conference GAMA Days 2022, Jun 2022, Online, France. Abstract on HAL; Slides.
A modular framework for multi-behavior and multi-scale simulation of pedestrians. 2nd conference GAMA Days 2022, Jun 2022, Online, France. Abstract on HAL; Slides.
Using synthetic data to improve performance of data-driven algorithms in high density pedestrian situations. 2nd conference GAMA Days 2022, Jun 2022, Online, France. Abstract on HAL; Slides.
Modélisation de l'anticipation des collisions et des contacts dans la dynamique des foules piétonnes (conversation). Regional French Conference on Complex Systems (FRCCS 2022), 20 – 22 June 2022, Paris, France. Abstract.
Stop and go pedestrian wave based on a behavioural model (flash talk). Regional French Conference on Complex Systems (FRCCS 2022), 20 – 22 June 2022, Paris, France.
Review of Pedestrian Trajectory Prediction methods: Comparing physics-based and data-based approaches. 10th International Conference on Pedestrian and Evacuation Dynamics (PED2021). 28 – 30 November 2021, Online, Melbourne, Australia. Abstract; Slides.
Distance vs time-to-collision pedestrian models: verification and validation. 10th International Conference on Pedestrian and Evacuation Dynamics (PED2021). 28 – 30 November 2021, Online, Melbourne, Australia. Abstract; Slides.
Initiating lane and band formation in heterogeneous pedestrian dynamics. 10th International Conference on Pedestrian and Evacuation Dynamics (PED2021). 28 – 30 November 2021, Online, Melbourne, Australia. Abstract; Slides.
Review of Pedestrian Trajectory Prediction methods: Comparing physics-based and data-based approaches. 17ième Journées de la Matière Condensée (JMC17), 24 – 27 August 2021, Online, Rennes, France. Abstract; Slides.
Heterogeneity-induced lane and band formation in self-driven particle systems. 17ième Journées de la Matière Condensée (JMC17), 24 – 27 August 2021, Online, Rennes, France. Abstract; Slides.