Abstract
One limitation of current data-driven automatic crowd modeling methods is that the models generated have low interpretability, which limits the practical applications of the models. In this article, we propose a new data-driven crowd modeling approach that can generate universal behavior rules with better interpretability. Higher interpretability helps people better understand and analyze the rules. Furthermore, the proposed approach considers both static and dynamic features during modeling to generate a realistic crowd, based on the assumption that humans tend to consider different features with respect to their states. In the proposed method, the automatic behavior rule generation problem is formulated as a symbolic regression problem. Then, the problem is solved by multi-objective genetic programming. On one hand, to improve the interpretability of the behavior rules found, a new mechanism is proposed to guide the algorithm to find concise and dimensionally consistent solutions. On the other hand, decisions made by considering static and dynamic features respectively are combined to improve the generated crowd realism. To validate the effectiveness of the proposed method, three real-world datasets are utilized for training and testing. The simulation results demonstrate that the proposed method is able to find universal behavior rules that are competitive to previous work in accuracy while having better interpretability.
- Alexandre Alahi, Kratarth Goel, Vignesh Ramanathan, Alexandre Robicquet, Li Fei-Fei, and Silvio Savarese. 2016. Social LSTM: Human trajectory prediction in crowded spaces. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’16). 961--971.Google ScholarCross Ref
- Raghda Alqurashi and Tom Altman. 2018. Multi-level multi-stage agent-based decision support system for simulation of crowd dynamics. In Proceedings of the 2018 23rd International Conference on Engineering of Complex Computer Systems (ICECCS’18). 82--92.Google ScholarCross Ref
- Federico Bartoli, Giuseppe Lisanti, Lamberto Ballan, and Alberto Del Bimbo. 2018. Context-aware trajectory prediction. In Proceedings of the 2018 24th International Conference on Pattern Recognition (ICPR’18). IEEE, Los Alamitos, CA, 1941--1946.Google ScholarCross Ref
- Kalyanmoy Deb, Amrit Pratap, Sameer Agarwal, and T. Meyarivan. 2002. A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation 6, 2 (2002), 182--197.Google ScholarDigital Library
- Xiaojie Du, Yuan Chen, Ahmed Bouferguene, and Mohamed Al-Hussein. 2018. Multi-agent based simulation of elderly egress process and fall accident in senior apartment buildings. In Proceedings of the 2018 Winter Simulation Conference (WSC’18). 929--940.Google ScholarDigital Library
- Vashti Galpin, Natalia Zoń, Pia Wilsdorf, and Stephen Gilmore. 2018. Mesoscopic modelling of pedestrian movement using Carma and its tools. ACM Transactions on Modeling and Computer Simulation 28, 2 (March 2018, Article 11, 26 pages. DOI:https://doi.org/10.1145/3155338Google ScholarDigital Library
- Agrim Gupta, Justin Johnson, Li Fei-Fei, Silvio Savarese, and Alexandre Alahi. 2018. Social GAN: Socially acceptable trajectories with generative adversarial networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’18). 2255--2264.Google ScholarCross Ref
- Dirk Helbing, Anders Johansson, and Habib Zein Al-Abideen. 2007. Dynamics of crowd disasters: An empirical study. Physical Review E 75, 4 (2007), 046109.Google ScholarCross Ref
- Rachel Joanne Hunt, Mark Richard Johnston, and Mengjie Zhang. 2016. Evolving Dispatching Rules with Greater Understandability for Dynamic Job Shop Scheduling. Massachusetts Institute of Technology, Cambridge, MA.Google Scholar
- Manh Huynh and Gita Alaghband. 2019. Trajectory prediction by coupling scene-LSTM with human movement LSTM. In Proceedings of the International Symposium on Visual Computing. 244--259.Google ScholarCross Ref
- Anders Johansson, Dirk Helbing, and Pradyumn K. Shukla. 2007. Specification of the social force pedestrian model by evolutionary adjustment to video tracking data. Advances in Complex Systems 10 (2007), 271--288.Google ScholarCross Ref
- Maarten Keijzer and Vladan Babovic. 1999. Dimensionally aware genetic programming. In Proceedings of the 1st Annual Conference on Genetic and Evolutionary Computation—Volume 2 (GECCO’99). 1069--1076.Google ScholarDigital Library
- Nicholas Keller and Xiaolin Hu. 2019. Towards data-driven simulation modeling for mobile agent-based systems. ACM Transactions on Modeling and Computer Simulation 29, 1 (Feb. 2019), Article 1, 26 pages. DOI:https://doi.org/10.1145/3289229Google ScholarDigital Library
- Wee Lit Koh and Suiping Zhou. 2011. Modeling and simulation of pedestrian behaviors in crowded places. ACM Transactions on Modeling and Computer Simulation 21, 3 (Feb. 2011), Article 20, 23 pages. DOI:https://doi.org/10.1145/1921598.1921604Google ScholarDigital Library
- Kang Hoon Lee, Myung Geol Choi, Qyoun Hong, and Jehee Lee. 2007. Group behavior from video: A data-driven approach to crowd simulation. In Proceedings of the 2007 ACM SIGGRAPH/Eurographics Symposium on Computer Animation. 109--118.Google ScholarDigital Library
- Alon Lerner, Yiorgos Chrysanthou, and Dani Lischinski. 2007. Crowds by example. Computer Graphics Forum 26, 3 (2007), 655--664.Google ScholarCross Ref
- Guanghui Lu, Leiting Chen, and Weiping Luo. 2016. Real-time crowd simulation integrating potential fields and agent method. ACM Transactions on Modeling and Computer Simulation 26, 4 (March 2016), Article 28, 16 pages. DOI:https://doi.org/10.1145/2885496Google ScholarDigital Library
- Francisco Martinez-Gil, Miguel Lozano, and Fernando Fernández. 2014. MARL-Ped: A multi-agent reinforcement learning based framework to simulate pedestrian groups. Simulation Modelling Practice and Theory 47 (2014), 259--275.Google ScholarCross Ref
- Ramin Mehran, Alexis Oyama, and Mubarak Shah. 2009. Abnormal crowd behavior detection using social force model. In Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’09). 935--942.Google ScholarCross Ref
- Yi Mei, Su Nguyen, and Mengjie Zhang. 2017. Constrained dimensionally aware genetic programming for evolving interpretable dispatching rules in dynamic job shop scheduling. In Proceedings of the Asia-Pacific Conference on Simulated Evolution and Learning. 435--447.Google ScholarCross Ref
- David J. Montana. 1995. Strongly typed genetic programming. Electronic Commerce 3, 2 (1995), 199--230.Google Scholar
- Soraia Raupp Musse and Daniel Thalmann. 2001. Hierarchical model for real time simulation of virtual human crowds. IEEE Transactions on Visualization and Computer Graphics 7, 2 (2001), 152--164.Google ScholarDigital Library
- Sachin Patil, Jur P. van den Berg, Sean Curtis, Ming C. Lin, and Dinesh Manocha. 2011. Directing crowd simulations using navigation fields. IEEE Transactions on Visualization and Computer Graphics 17, 2 (2011), 244--254.Google ScholarDigital Library
- Stefano Pellegrini, Andreas Ess, and Luc J. Van Gool. 2010. Improving data association by joint modeling of pedestrian trajectories and groupings. In Computer Vision—ECCV 2010. Lecture Notes in Computer Science, Vol. 6311. Springer, 452--465.Google Scholar
- Stefano Pellegrini, Andreas Ess, Konrad Schindler, and Luc Van Gool. 2009. You’ll never walk alone: Modeling social behavior for multi-target tracking. In Proceedings of the 2009 IEEE 12th International Conference on Computer Vision. IEEE, Los Alamitos, CA, 261--268.Google ScholarCross Ref
- Mark Pfeiffer. 2018. Learning to Navigate: Data-Driven Motion Planning for Autonomous Ground Robots. Ph.D. Dissertation. ETH Zurich.Google Scholar
- Paul Scovanner and Marshall F. Tappen. 2009. Learning pedestrian dynamics from the real world. In Proceedings of the 2009 IEEE 12th International Conference on Computer Vision. 381--388.Google Scholar
- Wei Shao and Demetri Terzopoulos. 2007. Autonomous pedestrians. Graphical Models 69, 5 (2007), 246--274. DOI:https://doi.org/10.1016/j.gmod.2007.09.001Google ScholarDigital Library
- Mankyu Sung, Michael Gleicher, and Stephen Chenney. 2004. Scalable behaviors for crowd simulation. Computer Graphics Forum 23 (2004), 519--528.Google ScholarCross Ref
- Sing Kuang Tan, Nan Hu, and Wentong Cai. 2018. Data-driven agent-based simulation for pedestrian capacity analysis. In Proceedings of the International Conference on Computational Science. 103--116.Google ScholarCross Ref
- Leonel Toledo, Ivan Rivalcoba, and Isaac Rudomin. 2018. Fuzzy and data-driven urban crowds. In Proceedings of the International Conference on Computational Science. 280--290.Google ScholarCross Ref
- Jason Tsai, Natalie Fridman, Emma Bowring, Matthew Brown, Shira Epstein, Gal A. Kaminka, Stacy Marsella, et al. 2011. ESCAPES: Evacuation simulation with children, authorities, parents, emotions, and social comparison. In Proceedings of the 10th International Conference on Autonomous Agents and Multi Agents Systems (AAMAS’11). 457--464.Google Scholar
- Jur Van Den Berg, Stephen J. Guy, Ming Lin, and Dinesh Manocha. 2011. Reciprocal n-body collision avoidance. In Robotics Research. Springer, 3--19.Google Scholar
- Xiang Wei, Wei Lu, Lili Zhu, and Weiwei Xing. 2018. Learning motion rules from real data: Neural network for crowd simulation. Neurocomputing 310 (2018), 125--134.Google ScholarCross Ref
- Tao Xiang and Shaogang Gong. 2008. Video behavior profiling for anomaly detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 30, 5 (2008), 893--908.Google ScholarDigital Library
- Kota Yamaguchi, Alexander C. Berg, Luis E. Ortiz, and Tamara L. Berg. 2011. Who are you with and where are you going? In Proceedings of the 24th IEEE Conference on Computer Vision and Pattern Recognition (CVPR’11). 1345--1352.Google Scholar
- Jun Zhang, Wolfram Klingsch, Andreas Schadschneider, and Armin Seyfried. 2012. Ordering in bidirectional pedestrian flows and its influence on the fundamental diagram. Journal of Statistical Mechanics: Theory and Experiment 2012, 02 (2012), P02002.Google ScholarCross Ref
- Mingbi Zhao, Stephen John Turner, and Wentong Cai. 2013. A data-driven crowd simulation model based on clustering and classification. In Proceedings of the 2013 IEEE/ACM 17th International Symposium on Distributed Simulation and Real Time Applications. IEEE, Los Alamitos, CA, 125--134.Google ScholarDigital Library
- Mingbi Zhao, Jinghui Zhong, and Wentong Cai. 2018. A role-dependent data-driven approach for high-density crowd behavior modeling. ACM Transactions on Modeling and Computer Simulation 28, 4 (2018), 28.Google ScholarDigital Library
- Jinghui Zhong and Wentong Cai. 2016. A hyper-heuristic framework for agent-based crowd modeling and simulation. In Proceedings of the 2016 International Conference on Autonomous Agents and Multiagent Systems. 1331--1332.Google Scholar
- Jinghui Zhong, Wentong Cai, Michael Lees, and Linbo Luo. 2017. Automatic model construction for the behavior of human crowds. Applied Soft Computing 56 (2017), 368--378. DOI:https://doi.org/10.1016/j.asoc.2017.03.020Google ScholarDigital Library
- Jinghui Zhong, Wentong Cai, Linbo Luo, and Mingbi Zhao. 2016. Learning behavior patterns from video for agent-based crowd modeling and simulation. Autonomous Agents and Multi-Agent Systems 30, 5 (2016), 990--1019.Google ScholarDigital Library
- Jinghui Zhong, Yew-Soon Ong, and Wentong Cai. 2016. Self-learning gene expression programming. IEEE Transactions on Evolutionary Computation 20, 1 (2016), 65--80.Google ScholarDigital Library
- Bolei Zhou, Xiaogang Wang, and Xiaoou Tang. 2012. Understanding collective crowd behaviors: Learning a mixture model of dynamic pedestrian-agents. In Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’12). 2871--2878.Google ScholarCross Ref
- Suiping Zhou, Dan Chen, Wentong Cai, Linbo Luo, Malcolm Yoke Hean Low, Feng Tian, Victor Su-Han Tay, Darren Wee Sze Ong, and Benjamin D. Hamilton. 2010. Crowd modeling and simulation technologies. ACM Transactions on Modeling and Computer Simulation 20, 4 (2010), 20.Google ScholarDigital Library
Index Terms
Dimensionally Aware Multi-Objective Genetic Programming for Automatic Crowd Behavior Modeling
Recommendations
A Role-Dependent Data-Driven Approach for High-Density Crowd Behavior Modeling
Special Issue on PADS 2016In this article, we propose a role-dependent (RD) data-driven modeling approach to simulate pedestrians’ motion in high-density scenes. It is commonly observed that pedestrians behave quite differently when walking in a dense crowd. Some people explore ...
Crowd modeling and simulation technologies
As a collective and highly dynamic social group, the human crowd is a fascinating phenomenon that has been frequently studied by experts from various areas. Recently, computer-based modeling and simulation technologies have emerged to support ...
Automatic rule identification for agent-based crowd models through gene expression programming
AAMAS '14: Proceedings of the 2014 international conference on Autonomous agents and multi-agent systemsAgent-based modelling of human crowds has now become an important and active research field, with a wide range of applications such as military training, evacuation analysis and digital game. One of the significant and challenging tasks in agent-based ...
Comments