动态人群情绪感染并行仿真算法

系统仿真学报 ›› 2016, Vol. 28 ›› Issue (9): 1964-1969.

动态人群情绪感染并行仿真算法

向南¹, 张明敏², 朱凌云¹

1.重庆理工大学计算机学院,重庆 400054;
2.浙江大学计算机学院,杭州 310058

收稿日期:2016-05-31 修回日期:2016-07-11 出版日期:2016-09-08 发布日期:2020-08-14
作者简介:向南(1984-), 男, 陕西旬阳, 博士, 讲师, 研究方向为情感计算; 张明敏(1968-), 女, 宁波, 副教授, 研究方向为人工智能; 朱凌云(通信作者1969-), 男, 四川巴中, 副教授, 研究方向为人工智能, 物联网。
基金资助:
国家自然科学基金(61502064), 重庆市教委科学技术研究项目(KJ1400907)

Paralleled Dynamic Crowd Emotion Contagion Algorithm

Xiang Nan¹, Zhang Mingmin², Zhu Lingyun¹

1. School of Computer Science and Engineering, Chongqing University of Technology, Chongqing 400054, China;
2. School of Computer Science and Technology, Zhejiang University, Hangzhou 310058, China

Received:2016-05-31 Revised:2016-07-11 Online:2016-09-08 Published:2020-08-14

摘要/Abstract

摘要： 由于动态人群的个体位置是动态变化的,因此情绪传染过程的计算成为一个难题。现有算法在人群仿真中通常需要计算每两物体之间的反应而无法达到实时要求。为了解决这个问题,提出一种基于GPU加速和社会力的感染力并行计算方法。将个人的情感影响范围投射到二维网格上,并用相同大小九宫格表示;并行计算个体与最近邻之间的相互作用力以便确定个体的运动位置;接着计算个体相互之间的情绪感染力向量并渲染仿真结果。整个算法通过GPU并行处理。实验显示,本算法有效提高较大规模动态人群情绪感染仿真的精确性与计算速度。

关键词: 情感计算, 情绪感染, 社会力, GPU加速

Abstract: As the positions of crowd usually change dynamically, then computing the contagion process becomes a challenge. There current algorithms were too time consuming to be adopted as they needed to calculate the reactions between every two objects. In order to solve this problem, a social force based contagion computing algorithm with GPU acceleration was provided. Individuals’ affection fields were projected onto two dimensional mesh grid and represented by the nine-box diary; The social force reactions between individual and nearest neighbors were computed to get the moving position; The contagion results from nearest neighbors were calculated. All of these steps were paralleled processing by GPU. Experiments show that the algorithm can efficiently increase the accuracy and speed of emotion contagion.

Key words: affective computing, emotion contagion, social force, GPU acceleration

中图分类号:

TP391.9

向南, 张明敏, 朱凌云. 动态人群情绪感染并行仿真算法[J]. 系统仿真学报, 2016, 28(9): 1964-1969.

Xiang Nan, Zhang Mingmin, Zhu Lingyun. Paralleled Dynamic Crowd Emotion Contagion Algorithm[J]. Journal of System Simulation, 2016, 28(9): 1964-1969.

参考文献 19

[1]	郭谨一, 刘爽, 陈绍宽, 等. 行人运动仿真研究综述[J]. 系统仿真学报, 2008, 20(9): 2237-2242.(Guo Jinyi, Liu Shuang, Chen Shaokuan, et al.Review of Pedestrian Movement Simulation Studies[J]. Journal of System Simulation (S1004-731X), 2008, 20(9): 2237-2242.)
[2]	B Zhou, X Tang, X Wang.Learning collective crowd behaviors with dynamic pedestrian-agents[J]. International Journal of Computer Vision (S0920-5691), 2015, 111(1): 50-68.
[3]	D C Duives, W Daamen, S P Hoogendoorn.Continuum modelling of pedestrian flows-Part 2: Sensitivity analysis featuring crowd movement phenomena[J]. Physica A Statistical Mechanics & Its Applications (S0378-4371), 2016, 447: 36-48.
[4]	L Tan, M Hu, H Lin.Agent-based simulation of building evacuation: Combining human behavior with predictable spatial accessibility in a fire emergency[J]. Information Sciences (S0165-5515), 2015, 295: 53-66.
[5]	C Wang, J Wang.A Modified Floor Field Model Combined with Risk Field for Pedestrian Simulation[J]. Mathematical Problems in Engineering (S1024-123X), 2016(1): 1-10.
[6]	M L Xu, H Jiang, X G Jin, et al.Crowd Simulation and Its Applications: Recent Advances[J]. Journal of Computer Science and Technology (S1000-9000), 2014, 29(5): 799-811.
[7]	R Zhang, J Qi, M Stradling, et al.Towards a Painless Index for Spatial Objects[J]. ACM Transactions on Data Base Systems (S0362-5915), 2014, 39(3): 19.
[8]	R Zhang, H V Jagadish, B T Dai, et al.Optimized Algorithms for Predictive Range and KNN Queries on Moving Objects[J] Information Systems (S0306-4379), 2010, 35(8): 911-932.
[9]	王志良. 人工情感[M]. 北京: 机械工业出版社, 2009.(Z Wang.Artificial Emotion [M]. Beijing, China: China Machine Press, 2009.)
[10]	W Zeng, P Chen, H Nakamura, et al.Application of social force model to pedestrian behavior analysis at signalized crosswalk[J]. Transportation Research Part C Emerging Technologies (S0968-090X), 2014, 40(1): 143-159.
[11]	X Yang, H Dong, Q Wang, et al.Guided crowd dynamics via modified social force model[J]. Physica A: Statistical Mechanics and its Applications (S0378-4371), 2014, 411: 63-73.
[12]	F Martinez-Gil, M Lozano, F Fernández.MARL-Ped: A multi-agent reinforcement learning based framework to simulate pedestrian groups[J]. Simulation Modelling Practice and Theory (S1569-190X), 2014, 47(47): 259-275.
[13]	N Xiang, Z Pan, M Zhang.Using emotion transmission to simulate spectator behaviors[C]// Proceedings of the 11th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry. USA: ACM, 2012: 51-54.
[14]	T Bosse, M Hoogendoorn, M C A Klein, et al. Modelling collective decision making in groups and crowds: Integrating social contagion and interacting emotions, beliefs and intentions[J]. Autonomous Agents and Multi-Agent Systems (S1387-2532), 2013, 27(1): 52-84.
[15]	L Nyland.GPU Gems 3-Fast N-Body Simulation with CUDA[M]. Boston: Addison-Wesley Professional, 2007: 677-694.
[16]	N K Govindaraju, S Redon, M C Lin, et al.CULLIDE: interactive collision detection between complex models in large environments using graphics hardware[C]// ACM Siggraph/eurographics Workshop on Graphics Hardware, San Diego, California, USA, July, 2003. USA: ACM, 2003: 25-32.
[17]	M Tang, S Curtis, S E Yoon, et al.Interactive Continuous Collision Detection between Deformable Models using Connectivity-Based Culling[J]. IEEE Transactions on Visualization & Computer Graphics (S1077-2626), 2009, 15(4): 25-36.
[18]	P Du, M Tang, R F Tong.A Parallel Collision Detection on Multi-core Platform[J]. Journal of Computer-Aided Design & Computer Graphics (S1003-9775), 2011, 23(5): 833-838.
[19]	向南, 朱凌云, 张明敏. 情感影响下的行人动态聚集过程仿真模型[J]. 系统仿真学报, 2015, 27(10): 2395-2400.(Xiang Nan, Zhu Lingyun, Zhang Mingmin.Pedestrian Dynamic Aggregation Simulation Model with Emotional Affection[J]. Journal of System Simulation (S1004-731X), 2015, 27(10): 2395-2400.)