Study on Data-driven ORCA Preference Velocity

doi:10.16182/j.issn1004731x.joss.19-FZ0330

Abstract

Abstract: Crowd evacuation is an important aspect of emergency management. Because of economic and safety reasons, emergency evacuation drill cannot be carried out repeatedly, however modeling and simulation has the obvious advantages of low cost, high repeatability, etc. Optimal Reciprocal Collision Avoidance (ORCA) is a velocity-based Collision Avoidance method, which has fine granularity and high computational efficiency, and is widely used in robot collision avoidance and other domains. In this paper, ORCA is introduced into crowd evacuation simulation, and the optimal parameters of ORCA under different kinds of speed strategies are identified by data-driven method. The optimal strategy is selected from the strategy set and its optimal parameter configuration is determined.

Key words: crowd evacuation simulation, ORCA model, data-driven, differential evolution algorithm

CLC Number:

TP391.9

Wang Jie, Chen Bin, Yuan Peng, Ma Liang, Qiu Xiaogang. Study on Data-driven ORCA Preference Velocity[J]. Journal of System Simulation, 2019, 31(12): 2731-2739.

References

[1] Fiorini P, Shiller Z.Motion planning in dynamic environments using velocity obstacles[J]. The International Journal of Robotics Research (S1741-3176), 1998, 17(7): 760-772.
[2] Berg J V D, Lin M, Manocha D. Reciprocal Velocity Obstacles for real-time multi-agent navigation[C]// Proceedings of the 2008 IEEE International Conference on Robotics and Automation. NJ: Piscataway, 2008.
[3] Hansen N, Ostermeier A.Adapting arbitrary normal mutation distributions in evolution strategies: The covariance matrix adaptation[C]// Proceedings of IEEE international conference on evolutionary computation. IEEE. NJ: Piscataway, 1996: 312-317.
[4] Wolinski D, J Guy S, Olivier A H, et al. Parameter estimation and comparative evaluation of crowd simulations[J]. Computer Graphics Forum (S1467-8659), 2014, 33(2): 303-312.
[5] Wolinski D, Guy S J, Olivier A H, et al.Optimization-based Pedestrian Model Calibration for Evaluation[J]. Transportation research procedia (S2352-1465), 2014: 228-236.
[6] Helbing D, Farkas I, Vicsek T.Simulating dynamical features of escape panic[J]. Nature (S0028-0836), 2000, 407(6803): 487-490.
[7] Zhong J, Cai W.Differential evolution with sensitivity analysis and the Powell's method for crowd model calibration[J]. Journal of Computational Science (S1877-7503), 2015, 9: 26-32.
[8] Helbing D, Johansson A, Al-Abideen H Z. Dynamics of crowd disasters: An empirical study[J]. Physical review E (S1539-3755), 2007, 75(4): 046109.
[9] Seer S, Brandle N, Ratti C, et al.Kinects and human kinetics: a new approach for studying pedestrian behavior[J]. Transportation Research Part C-emerging Technologies (S0968-090X), 2014: 212-228.
[10] Zhang J, Klingsch W, Schadschneider A, et al.Transitions in pedestrian fundamental diagrams of straight corridors and T-junctions[J]. Journal of Statistical Mechanics: Theory and Experiment (S1742-5468), 2011(6): P06004.
[11] 王静虹. 非常规突发情况下大规模人群疏散的不确定性研究[D]. 合肥: 中国科学技术大学, 2013.
Wang Jinghong.Research on the uncertainties of crowd large-scale evacuation under unconventional emergencies[D]. Hefei: University of Science and Technology of China, 2013.
[12] Helbing D, Johansson A.Pedestrian, crowd and evacuation dynamics[M]. NY: Springer New York, 2009.
[13] Storn R, Price K.Differential evolution-a simple and efficient heuristic for global optimization over continuous spaces[J]. Journal of Global Optimization (S1573-2916), 1997, 11(4): 341-359.
[14] Moussaïd M, Helbing D, Theraulaz G.How simple rules determine pedestrian behavior and crowd disasters[J]. Proceedings of the National Academy of Sciences (S1091-6490), 2011, 108(17): 6884-6888.
[15] Johansson A, Helbing D, Shukla P K.Specification of the social force pedestrian model by evolutionary adjustment to video tracking data[J]. Advances in Complex Systems (S0219-5259), 2007: 271-288.
[16] Ma L, Chen B, Wang X, et al.The analysis on the desired speed in social force model using a data driven approach[J]. Physica A: Statistical Mechanics and its Applications (S0378-4371), 2019, 525: 894-911.
[17] Kim S, Guy S J, Liu W, et al.Brvo: Predicting pedestrian trajectories using velocity-space reasoning[J]. The International Journal of Robotics Research (S1741-3176), 2015, 34(2): 201-217.
[18] He L, Pan J, Wang W, et al.Proxemic group behaviors using reciprocal multi-agent navigation[C]// 2016 IEEE International Conference on Robotics and Automation (ICRA). NJ: IEEE, 2016: 292-297.
[19] Zhao M, Zhong J, Cai W.A Role-Dependent Data-Driven Approach for High-Density Crowd Behavior Modeling[J]. ACM Transactions on Modeling and Computer Simulation (TOMACS)(S1558-1195), 2018, 28(4): 28.