考虑共同关注区域静态交谈群组检测的机器人导航及行为评价

doi:10.13973/j.cnki.robot.210276

摘要
图/表
参考文献
相关文章 (1)

全文: PDF (768 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
周磊, 赵坤旭, 宋一诺, 刘景泰. 考虑共同关注区域静态交谈群组检测的机器人导航及行为评价[J]. 机器人, 2022, 44(4): 494-503.DOI: 10.13973/j.cnki.robot.210276. ZHOU Lei, ZHAO Kunxu, SONG Yinuo, LIU Jingtai. Detecting Static Conversation Group with Common Concern Area for Robot Navigation and Behavior Evaluation. ROBOT, 2022, 44(4): 494-503. DOI: 10.13973/j.cnki.robot.210276.

摘要针对静态交谈群组，提出了一种基于共同关注区域的F型社交关系（F-formation）检测算法。该算法将行人的位置、方向作为输入构造群组的共同关注区域，采用一个基于滑动窗口的最大值滤波器检测群组中心，然后根据群组中心进行群组聚类。在实现静态交谈群组检测的基础上，构造群组舒适空间，并基于多层代价地图机制将群组舒适空间模型应用于时间依赖的A^*路径规划算法，实现了考虑群组舒适的移动机器人导航。除此之外，具有社交感知能力的机器人导航行为给予人的舒适感受难以定量评价，本文搭建了一个基于图卷积网络的机器人行为评价模型。实验结果显示该评价网络达到了与人类相似的机器人行为评价能力，可视化分析结果显示了评价网络结果的合理性。通过评价网络对机器人轨迹进行评估，发现考虑静态交谈群组的机器人能够产生更加满足人类舒适感受的运动轨迹。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	周磊
	赵坤旭
	宋一诺
	刘景泰

关键词 ：静态交谈群组检测, F型社交关系, 共同关注区域, 群组舒适空间, 机器人社交导航, 图卷积网络

Abstract：An F-formation detection algorithm based on common concern areas is proposed for static conversation group. It takes the position and direction of pedestrians as input to construct the common concern area of groups. Then a sliding window-based maximum filter is used to detect the group center for clustering. After detecting the static conversation group, a group comfort space is constructed for a time-dependent A^* path planning algorithm based on multi-layer costmap mechanism. Thus, mobile robot navigation considering group comfort is realized. In addition, it is challenging to quantitatively evaluate whether a robot’s navigation behavior is socially acceptable. A graph convolutional network-based model for evaluating robot behavior is constructed. Experimental results show that the evaluation network has a similar capacity to humans in evaluating the robot behavior. Visualization results show the rationality of the evaluation network’s results. According to the evaluation network, robots considering static conversation group can produce more comfortable trajectories.

Key words： static conversational group detection F-formation common concern area group comfort space social aware robot navigation graph convolutional network

收稿日期: 2021-06-22 录用日期: 2021-08-20

TP242

基金资助:国家重点研发计划（2019YFB1310600）；国家自然科学基金（62173189）.

通讯作者: 刘景泰，liujt@nankai.edu.cn E-mail: liujt@nankai.edu.cn

作者简介: 周磊（1995-），男，博士生。研究领域：共融机器人，路径规划，行人轨迹预测，逆强化学习。
赵坤旭（1997-），男，硕士生。研究领域：共融机器人，注视方向估计，机器人导航。
宋一诺（2000-），女，本科生。研究领域：共融机器人，行人跟随。

链接本文:

https://robot.sia.cn/CN/10.13973/j.cnki.robot.210276 或 https://robot.sia.cn/CN/Y2022/V44/I4/494

[1] 韩建达,方勇纯,赵新,等.机器人的智能发育[J].人工智能, 2018(3):28-35.Han J D, Fang Y C, Zhao X, et al.Intelligent development of robots[J].AI-View, 2018(3):28-35.
[2] 何玉庆,赵忆文,韩建达,等.与人共融——机器人技术发展的新趋势[J].机器人产业, 2015(5):74-80.He Y Q, Zhao Y W, Han J D, et al.Tri-Co robot-The new trend in the development of robotics[J].Robot Industry, 2015(5):74-80.
[3] Truong X T, Ngo T D.Dynamic social zone based mobile robot navigation for human comfortable safety in social environments[J].International Journal of Social Robotics, 2016, 8:663-684.
[4] Fox D, Burgard W, Thrun S.The dynamic window approach to collision avoidance[J].IEEE Robotics&Automation Magazine, 1997, 4(1):23-33.
[5] Snape J, van den Berg J, Guy S J, et al.The hybrid reciprocal velocity obstacle[J].IEEE Transactions on Robotics, 2011, 27(4):696-706.
[6] Roesmann C, Feiten W, Woesch T, et al.Trajectory modification considering dynamic constraints of autonomous robots[C]//7th German Conference on Robotics.Munich, Germany:VDE, 2012:1-6.
[7] Rosmann C, Feiten W, Wosch T, et al.Efficient trajectory optimization using a sparse model[C]//European Conference on Mobile Robots.Piscataway, USA:IEEE, 2013:138-143.
[8] Foster M E, Alami R, Gestranius O, et al.The MuMMER project:Engaging human-robot interaction in real world public spaces[M]//Lecture Notes in Computer Science, Vol.9979.Berlin, Germany:Springer, 2016:753-763.
[9] Lead T, Lead W P.Safe robot navigation in dense crowds[R/OL].(2020-05-26)[2021-04-25].https://cordis.europa.eu/project/id/779942/reporting.
[10] Papadakis P, Rives P, Spalanzani A.Adaptive spacing in humanrobot interactions[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscataway, USA:IEEE, 2014:2627-2632.
[11] 张森,刘景泰.基于多维度服务情景的人的舒适需求建模[J].机器人, 2019, 41(4):493-506.Zhang S, Liu J T.Modeling of human's comfort needs based on multi-dimensional service situations[J].Robot, 2019, 41(4):493-506.
[12] 周磊,张森,赵英利,等.基于非对称高斯函数的个人/群组动态舒适空间建模[J].机器人, 2021, 43(3):257-268.Zhou L, Zhang S, Zhao Y L, et al.Modeling of personal/group dynamic comfort space based on asymmetric Gaussian function[J].Robot, 2021, 43(3):257-268.
[13] 张森,周磊,刘梦,等.一种面向狭小、拥挤情景的服务机器人运动规划方法[J].机器人, 2021, 43(3):269-278.Zhang S, Zhou L, Liu M, et al.A motion planning scheme of service robots for cramped and crowded situations[J].Robot, 2021, 43(3):269-278.
[14] Kendon A.Spatial organization in social encounters:The Fformation system[J].Man Environment Systems, 1976, 6:291-296.
[15] Setti F, Russell C, Bassetti C, et al.F-formation detection:Individuating free-standing conversational groups in images[J].PLoS ONE, 2015, 10(9).DOI:10.1371/journal.pone.0123783.
[16] Setti F, Lanz O, Ferrario R, et al.Multi-scale F-formation discovery for group detection[C]//IEEE International Conference on Image Processing.Piscataway, USA:IEEE, 2013:3547-3551.
[17] Setti F, Hung H, Cristani M.Group detection in still images by F-formation modeling:A comparative study[C]//14th International Workshop on Image Analysis for Multimedia Interactive Services.Piscataway, USA:IEEE, 2013.DOI:10.1109/WIAMIS.2013.6616147.
[18] Akbari A, Farsi H, Mohamadzadeh S.Deep neural network with extracted features for social group detection[J].Journal of Electrical and Computer Engineering Innovations, 2021, 9(1):47-56.
[19] Yang F K, Peters C.AppGAN:Generative adversarial networks for generating robot approach behaviors into small groups of people[C]//28th IEEE International Conference on Robot and Human Interactive Communication.Piscataway, USA:IEEE, 2019.DOI:10.1109/RO-MAN46459.2019.8956425.
[20] Yang F K, Yin W J, Bjorkman M, et al.Impact of trajectory generation methods on viewer perception of robot approaching group behaviors[C]//29th IEEE International Conference on Robot and Human Interactive Communication.Piscataway, USA:IEEE, 2020:509-516.
[21] Yang F K, Peters C.App-LSTM:Data-driven generation of socially acceptable trajectories for approaching small groups of agents[C]//7th International Conference on Human-Agent Interaction.New York, USA:ACM, 2019:144-152.
[22] Sathyamoorthy A J, Patel U, Paul M, et al.CoMet:Modeling group cohesion for socially compliant robot navigation in crowded scenes[J].IEEE Robotics and Automation Letters, 2022, 7(2):1008-1015.
[23] Katyal K, Gao Y X, Markowitz J, et al.Learning a group-aware policy for robot navigation[DB/OL].(2020-12-22)[2021-04-26].https://arxiv.org/abs/2012.12291.
[24] Vega A, Manso L J, Macharet D G, et al.Socially aware robot navigation system in human-populated and interactive environments based on an adaptive spatial density function and space affordances[J].Pattern Recognition Letters, 2019, 118:72-84.
[25] Bera A, Randhavane T, Prinja R, et al.SocioSense:Robot navigation amongst pedestrians with social and psychological constraints[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscataway, USA:IEEE, 2017:7018-7025.
[26] Lu D V, Hershberger D, Smart W D.Layered costmaps for context-sensitive navigation[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscataway, USA:IEEE, 2014:709-715.
[27] Quigley M, Conley K, Gerkey B, et al.ROS:An open source robot operating system[C]//ICRA Workshop on Open Source Software.Piscataway, USA:IEEE, 2009:3-32.
[28] Kollmitz M, Hsiao K, Gaa J, et al.Time dependent planning on a layered social cost map for human-aware robot navigation[C]//European Conference on Mobile Robots.Piscataway, USA:IEEE, 2015.DOI:10.1109/ECMR.2015.7324184.
[29] Papenmeier F, Uhrig M, Kirsch A.Human understanding of robot motion:The role of velocity and orientation[J].International Journal of Social Robotics, 2019, 11:75-88.
[30] Manso L J, Nunez P, Calderita L V, et al.SocNav1:A dataset to benchmark and learn social navigation conventions[J].Data, 2020, 5(1).DOI:10.3390/data5010007.
[31] Baghel R, Kapoor A, Bachiller P, et al.A toolkit to generate social navigation datasets[M]//Advances in Intelligent Systems and Computing, Vol.1285.Berlin, Germany:Springer, 2020:180-193.
[32] Kip T N, Welling M.Semi-supervised classification with graph convolutional networks[DB/OL].(2018-09-09)[2020-09-30].https://arxiv.org/abs/1609.02907.