TY  - CONF
AU  - Alia, Ahmed
AU  - Chraibi, Mohcine
AU  - Seyfried, Armin
TI  - A Dynamic Distance Social LSTM for Predicting Pedestrian Trajectories in Crowded Environments
M1  - FZJ-2025-03048
PY  - 2025
AB  - This work introduces dynamic distance Social Long Short-Term Memory, a deep learning approach for pedestrian trajectory prediction in crowded environments. The approach integrates a new dynamic distance-based loss function into Social Long Short-Term Memory, enhancing collision avoidance without compromising displacement accuracy. The method is trained and evaluated on a heterogeneous density dataset and four homogeneous density datasets, covering various crowd-density levels. Experimental results show that the proposed approach outperforms baseline methods in reducing collision rates without decreasing displacement accuracy and, in most cases, even improving it.
T2  - Modelling, Data Analytics and AI in Engineering
CY  - 7 Jul 2025 - 11 Jul 2025, Porto (Portugal)
Y2  - 7 Jul 2025 - 11 Jul 2025
M2  - Porto, Portugal
LB  - PUB:(DE-HGF)6
DO  - DOI:10.34734/FZJ-2025-03048
UR  - https://juser.fz-juelich.de/record/1044146
ER  -