| 摘要: |
| 目的 提出一种兼具主观测量和生理数据验证的机器人语音交互认知负荷的评价方案。方法 设计实验使受试者对不同的机器人语音交互系统进行体验，完成相应的任务并接受主观认知负荷测量，同时加以生理电信号测量。主观测量部分利用层次分析法（AHP）对传统的NASA-TLX量表进行改进，使得测量更加客观与细致；并利用生理信号测量的结果验证主观评价的可靠程度与实际意义。结果 得出不同语音交互系统认知负荷的主观评价数据与生理测量数据，两者相互印证，从而得到可靠的认知负荷评价结果，并进一步证明了评价系统本身的有效性。结论 基于AHP的基本原理，改进了NASA-TLX量表，并建立起了一套包含四个准则十四个指标的语音交互认知负荷评价体系，利用心率与皮肤电信号测量进行了辅助验证，提出了一种研究机器人语音交互用户体验的技术手段。 |
| 关键词: 语音交互 认知负荷 AHP 生理电 机器人 |
| DOI：10.19554/j.cnki.1001-3563.2020.18.023 |
| 分类号:TB472 |
| 基金项目: |
|
| Robot Voice Interaction Cognitive Load Evaluation Based on AHP and Physiological Signal |
|
CHEN Hao-long, HE Ren-ke
|
|
Hunan University, Changsha 410082, China
|
| Abstract: |
| It aims to introduce a method for evaluating the cognitive load of robot voice interaction based on both subjective measurement and physiological data verification. Experiments were designed for subjects to experience different robot voice interaction systems, finish different tasks and accept subjective cognitive load measurement and physiological electrical signal measurement. The analytic hierarchy process (AHP) was used in subjective measurement to improve the traditional NASA-TLX, which made the measurement more objective and detailed. And physiological signal measurement was used to verify the effectiveness and reliability of the subjective evaluation data. During experiments of different voice interaction systems, the subjective evaluation data and physiological measurement data of cognitive load were obtained. Reliable cognitive load evaluation results were obtained through their mutual verification, which further proved the effectiveness of the evaluation system. Based on the principles of AHP, the study improves NASA-TLX, establishes a voice interactive cognitive load evaluation system with 4 criteria and 14 indicators, completes auxiliary validation by heart rate data and skin electrical signal measurement, and proposes a technological means to research user experiment of robot voice interaction. |
| Key words: voice interaction cognitive load AHP physiological signal robot |
渝公网安备 50010702501716号