工程科学学报，第39卷.第10期：1591-1597,2017年10月 Chinese Journal of Engineering,Vol.39,No.10:1591-1597,October 2017 D0I:10.13374/j.issn2095-9389.2017.10.019;htp://journals.ustb.edu.cn 面向老人迟滞性特征的可信情感计算 韩晶)，解仑)区，贺苗)，王志良”，任福继2) 1)北京科技大学计算机与通信工程学院，北京1000832)合肥工业大学情感计算与先进智能机器安徽省重点实验室，合肥230009 区通信作者，E-mail:xielun@usth.edu.cm 摘要根据老人认知心理特征中的迟滞性特征，结合基于Gss认知重评的情感计算模型，考虑迟滞性特征对个体情感状 态转移的影响，在认知重评参数的基础上，建立迟滞性因子，对当前情感状态下的认知重评能力进行修正，从而对情感计算模 型进行有效的修正，使人机交互更加自然和谐.为了实现情感计算模型的可信，结合现场可编程门阵列硬件平台和高级加密 标准密码算法对情感计算模型中的老人情感信息进行加密处理，实现情感计算过程中的数据可信.采用可编程片上系统技 术在现场可编程门阵列芯片EP4CE115F29C7中搭建功能实现所需的所有硬件组件，结合硬件组件编写逻辑程序并实现高级 加密标准密码算法，在情感计算过程中实现数据的实时传输和安全处理，实现可信的情感计算.最后实验结果表明受到迟滞 性特征影响下的情感计算模型与老人的真实情感具有高度一致性，且可信计算有助于提升老人的正向情感状态 关键词情感计算：迟滞性：可信计算：认知重评：高级加密标准密码算法 分类号TP751.1 Trusted affective computing based on hysteresis characteristics of the elderly HAN Jing,XIE Lun,HE Miao),WANG Zhi-liang),REN Fu-ji) 1)School of Computer and Communication Engineering,University of Science and Technology Beijing,Beijing 100083,China 2)Affective Computing and Advanced Intelligent Machines Key Laboratory,Hefei University of Technology,Hefei 230009,China Corresponding author,E-mail:xielun@ustb.edu.cn ABSTRACT By using the hysteresis characteristics in the cognitive psychological characteristics of the elderly,combined with an af- fective computing model based on Gross cognitive reappraisal,this paper takes consideration of the influence of the hysteresis charac- teristics on individual emotional state transition.It then establishes a hysteresis factor,based on cognitive reappraisal parameters,to correct the cognitive reappraisal ability of current emotional states,then simulates and analyzes the revised affective computing model using MATLAB.This improves the hysteresis factor effectively and revises the affective computing model resulting in more natural and harmonious human-robot interaction.To achieve the trusted affective computing model,this paper combines an FPGA hardware plat- form with an AES cipher algorithm for elderly affective information encryption processing,to realize the data trust in the affective com- puting.In this study,all the hardware components were built in the FPGA chip EP4CE115F29C7 using SOPC technology,the logic program was wrote,and the AES cipher algorithm was implemented with combination of the hardware components,to realize real-time data transmission and security during the process of affective computing,and its trust was improved.Finally,the experimental results show that the affective computing model under hysteresis characteristics is highly consistent with true elderly emotion,and trusted com- puting encourages a positive emotional state in the elderly. KEY WORDS affective computing;hysteresis;trusted computing;cognitive reappraisal;AES cipher algorithm 近年来，智能服务机器人的应用领域正逐步从工的情感交互成为一个最关键的步骤).智能服务机器 业和家庭转向个人服务，其中智能服务机器人与用户人通过有效的人机情感交互，能够极大地提高用户的 收稿日期：2016-12-15 基金项目：国家自然科学基金面上资助项目(61672093)：国家自然科学基金重点资助项目(61432004)：国家重点研发计划重点专项课题资助 项目(2016YFB1001404)工程科学学报,第 39 卷,第 10 期:1591鄄鄄1597,2017 年 10 月 Chinese Journal of Engineering, Vol. 39, No. 10: 1591鄄鄄1597, October 2017 DOI: 10. 13374 / j. issn2095鄄鄄9389. 2017. 10. 019; http: / / journals. ustb. edu. cn 面向老人迟滞性特征的可信情感计算 韩 晶1) , 解 仑1) 苣 , 贺 苗1) , 王志良1) , 任福继2) 1) 北京科技大学计算机与通信工程学院, 北京 100083 2) 合肥工业大学情感计算与先进智能机器安徽省重点实验室, 合肥 230009 苣通信作者, E鄄mail: xielun@ ustb. edu. cn 摘 要 根据老人认知心理特征中的迟滞性特征,结合基于 Gross 认知重评的情感计算模型,考虑迟滞性特征对个体情感状 态转移的影响,在认知重评参数的基础上,建立迟滞性因子,对当前情感状态下的认知重评能力进行修正,从而对情感计算模 型进行有效的修正,使人机交互更加自然和谐. 为了实现情感计算模型的可信,结合现场可编程门阵列硬件平台和高级加密 标准密码算法对情感计算模型中的老人情感信息进行加密处理,实现情感计算过程中的数据可信. 采用可编程片上系统技 术在现场可编程门阵列芯片 EP4CE115F29C7 中搭建功能实现所需的所有硬件组件,结合硬件组件编写逻辑程序并实现高级 加密标准密码算法,在情感计算过程中实现数据的实时传输和安全处理,实现可信的情感计算. 最后实验结果表明受到迟滞 性特征影响下的情感计算模型与老人的真实情感具有高度一致性,且可信计算有助于提升老人的正向情感状态. 关键词 情感计算; 迟滞性; 可信计算; 认知重评; 高级加密标准密码算法 分类号 TP751郾 1 Trusted affective computing based on hysteresis characteristics of the elderly HAN Jing 1) , XIE Lun 1) 苣 , HE Miao 1) , WANG Zhi鄄liang 1) , REN Fu鄄ji 2) 1) School of Computer and Communication Engineering, University of Science and Technology Beijing, Beijing 100083, China 2) Affective Computing and Advanced Intelligent Machines Key Laboratory,Hefei University of Technology, Hefei 230009, China 苣 Corresponding author, E鄄mail: xielun@ ustb. edu. cn ABSTRACT By using the hysteresis characteristics in the cognitive psychological characteristics of the elderly, combined with an af鄄 fective computing model based on Gross cognitive reappraisal, this paper takes consideration of the influence of the hysteresis charac鄄 teristics on individual emotional state transition. It then establishes a hysteresis factor, based on cognitive reappraisal parameters, to correct the cognitive reappraisal ability of current emotional states, then simulates and analyzes the revised affective computing model using MATLAB. This improves the hysteresis factor effectively and revises the affective computing model resulting in more natural and harmonious human鄄robot interaction. To achieve the trusted affective computing model, this paper combines an FPGA hardware plat鄄 form with an AES cipher algorithm for elderly affective information encryption processing, to realize the data trust in the affective com鄄 puting. In this study, all the hardware components were built in the FPGA chip EP4CE115F29C7 using SOPC technology, the logic program was wrote, and the AES cipher algorithm was implemented with combination of the hardware components, to realize real鄄time data transmission and security during the process of affective computing, and its trust was improved. Finally, the experimental results show that the affective computing model under hysteresis characteristics is highly consistent with true elderly emotion, and trusted com鄄 puting encourages a positive emotional state in the elderly. KEY WORDS affective computing; hysteresis; trusted computing; cognitive reappraisal; AES cipher algorithm 收稿日期: 2016鄄鄄12鄄鄄15 基金项目: 国家自然科学基金面上资助项目(61672093);国家自然科学基金重点资助项目(61432004);国家重点研发计划重点专项课题资助 项目(2016YFB1001404) 近年来,智能服务机器人的应用领域正逐步从工 业和家庭转向个人服务,其中智能服务机器人与用户 的情感交互成为一个最关键的步骤[1] . 智能服务机器 人通过有效的人机情感交互,能够极大地提高用户的