Can we learn where people go?


  • Marion Gödel Munich University of Applied Sciences, Munich, Germany and Technical University of Munich, Munich, Germany
  • Gerta Köster Munich University of Applied Sciences, Munich, Germany
  • Daniel Lehmberg Munich University of Applied Sciences, Munich, Germany and Technical University of Munich, Munich, Germany
  • Manfred Gruber Munich University of Applied Sciences, Munich, Germany
  • Angelika Kneidl accu:rate GmbH Institute for crowd simulation, Munich, Germany
  • Florian Sesser accu:rate GmbH Institute for crowd simulation, Munich, Germany



(4-8) pedestrian dynamics, predictive simulation, machine learning, random forest


In most agent-based simulators, pedestrians navigate from origins to destinations. Consequently, destinations are essential input parameters to the simulation. While many other relevant parameters as positions, speeds and densities can be obtained from sensors, like cameras, destinations cannot be observed directly. Our research question is: Can we obtain this information from video data using machine learning methods? We use density heatmaps, which indicate the pedestrian density within a given camera cutout, as input to predict the destination distributions. For our proof of concept, we train a Random Forest predictor on an exemplary data set generated with the Vadere microscopic simulator. The scenario is a crossroad where pedestrians can head left, straight or right. In addition, we gain first insights on suitable placement of the camera. The results motivate an in-depth analysis of the methodology.


M. Davidich and G. Köster, “Predicting Pedestrian Flow: A Methodology and a Proof of Concept

Based on Real-Life Data,” PLoS ONE, vol. 8, no. 12, pp. 1-11, 2003.

P. M. Kielar, A. Borrmann, “Modeling pedestrians' interest in locations: A concept to improve

simulations of pedestrian destination choice,” Simulation Modelling Practice and Theory, vol. 61, pp.

-62, 2016.

V. A. Sindagi and V.M. Patel, “A Survey of Recent Advances in CNN-based Single Image Crowd

Counting and Density Estimation,” Pattern Recognition Letters, vol. 107, pp. 3-16, 2018.

Y. Ma, E. W. M. Lee and R. K. K. Yuen, "An Artificial Intelligence-Based Approach for Simulating

Pedestrian Movement," in IEEE Transactions on Intelligent Transportation Systems, vol. 17, no. 11,

pp. 3159-3170, 2016.

J. Krumm and e. Horvitz, “Predestination: Inferring destinations from partial trajectories,” in

Proceedings of the 8th International Conference on Ubiquitous Computing UbiComp '06, pp.

-260, 2006.

J. K. Laurila et al., “The mobile data challenge: Big data for mobile computing research,” in

Proceedings of the Workshop on the Nokia Mobile Data Challenge, in Conjunction with the 10th

International Conference on Pervasive Computing, p. 1-8,2012.

A. Bera, S. Kim, D. Manocha, “Online parameter learning for data-driven crowd simulation and

content generation,” Computers and Graphics (Pergamon), vol. 55, pp. 68-79, 2016.

N. Bisagno, N. Conci, B. Zhang, “Data-Driven crowd simulation”, in 14th IEEE International

Conference on Advanced Video and Signal Based Surveillance, Lecce, Italy, 2017, pp.

F. Dietrich, “Data-Driven Surrogate Models for Dynamical Systems”, Ph. D. dissertation, Dept.

Scientific Computing, Technical University of Munich, Germany.

S. Kim et al., “Interactive and adaptive data-driven crowd simulation,” in Proceedings of IEEE

Virtual Reality, Greenville, SC, 2016, pp. 29-38.

J. Porzycki et al., “Dynamic data-driven simulation of pedestrian movement with automatic

validation,” in Traffic and Granular Flow '13, Jülich, Germany, 2013, pp. 129-136.

B. Liu et al., “A social force evacuation model driven by video data,” Simulation Modelling Practice

and Theory, vol. 84, pp. 190-203, 2018.

M. J. Seitz and G. Köster, “Natural discretization of pedestrian movement in continuous space,”

Physical Review E, vol. 86, no. 4, 2012.

I. von Sivers and G. Köster, “Dynamic Stride Length Adaptation According to Utility and Personal

Space”, Transportation Research Part B: Methodological, vol. 75, pp. 104-117, 2015.

Vadere Crowd Simulation (2016) [Online]. Available:

L. Breiman, “Random Forests,” Machine Learning, vol. 45, no. 1, pp. 5-32, 2001.

A. Cutler, D.R. Cutler, J.R. Stevens, “Random Forests” in Ensemble Machine Learning, C. Zhang,

Y. Ma, Ed. Boston, 2012, pp. 157-175.

A. C. Müller and S. Guido, „Introduction to Machine Learning with Python – A Guide for Data

Scientists”, O’Reilly Media, Inc., 2017, p.89.

F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning

Research, vol. 12, pp. 2825-2830, 2011.

D. Helbing, A. Johansson, H. Z. Al-Abideen, “Dynamics of crowd disasters: An empirical study,”

Physical Review E, vol. 75, no. 4, 2007.




How to Cite

Gödel, M., Köster, G., Lehmberg, D., Gruber, M., Kneidl, A., & Sesser, F. (2020). Can we learn where people go?. Collective Dynamics, 5, 134–141.



Proceedings of Pedestrian and Evacuation Dynamics 2018