Mercurial > hg > Members > atton > intelligence_robotics

comparison slide.md @ 5:9dbbe7f42909

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Add pictures
author Yasutaka Higa <e115763@ie.u-ryukyu.ac.jp>
date Thu, 18 Jun 2015 23:08:28 +0900
parents 51b87e0db067
children 3325edf9139f
comparison
equal deleted inserted replaced
4:51b87e0db067 5:9dbbe7f42909
41 * The robot should be trained with sociology data related to one country, and evolve its behaviour by engaging with people of another country in a small number of interactions. 41 * The robot should be trained with sociology data related to one country, and evolve its behaviour by engaging with people of another country in a small number of interactions.
42 * For the implementation of the gestures and the interaction experiment, we used the humanoid robot ARMAR-IIIb. 42 * For the implementation of the gestures and the interaction experiment, we used the humanoid robot ARMAR-IIIb.
43 * As the experiment is carried out in Germany, the interactions are with German participants, while preliminary training is done with Japanese data, which is culturally extremely different. 43 * As the experiment is carried out in Germany, the interactions are with German participants, while preliminary training is done with Japanese data, which is culturally extremely different.
44 44
45 # Introduction: ARMAR-IIIb 45 # Introduction: ARMAR-IIIb
46 <img src="pictures/ARMAR-IIIb.png" style='width 400px, height 400px'> </img> 46 <img src="pictures/ARMAR-IIIb.png" style='width: 350px; height: 350px; margin-left: 200px;'>
47 47
48 # Introduction: Target scenario 48 # Introduction: Target scenario
49 * The idea behind this study is a typical scenario in which a foreigner visiting a country for the first time greets local people in an inappropriate way as long as he is unaware of the rules that define the greeting choice. 49 * The idea behind this study is a typical scenario in which a foreigner visiting a country for the first time greets local people in an inappropriate way as long as he is unaware of the rules that define the greeting choice.
50 * (e.g., a Westerner in Japan) 50 * (e.g., a Westerner in Japan)
51 * For example, he might want to shake hands or hug, and will receive a bow instead. 51 * For example, he might want to shake hands or hug, and will receive a bow instead.
77 * 'intimacy' : is influenced by physical distance, eye contact, gender, location, and culture. (Social Distance) 77 * 'intimacy' : is influenced by physical distance, eye contact, gender, location, and culture. (Social Distance)
78 * 'Time' : time of the day is important for the choice of words. 78 * 'Time' : time of the day is important for the choice of words.
79 * 'Politeness', 'Power Relationship', 'culture' and more. 79 * 'Politeness', 'Power Relationship', 'culture' and more.
80 80
81 # Greeting Selection: Factors on Classification 81 # Greeting Selection: Factors on Classification
82 TODO: Insert FIGURE 1
83 * the factors to be cut are greyed out. 82 * the factors to be cut are greyed out.
83
84 <img src="pictures/factors.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'>
84 85
85 # Model of Greetings: Assumptions (1 - 5) 86 # Model of Greetings: Assumptions (1 - 5)
86 * The simplification was guided by the following ten assumptions. 87 * The simplification was guided by the following ten assumptions.
87 * Only two individuals (a robot and a human participant): we do not take in consideration a higher number of individuals. 88 * Only two individuals (a robot and a human participant): we do not take in consideration a higher number of individuals.
88 * Eye contact is taken for granted. 89 * Eye contact is taken for granted.
103 * They are categorical data, as they can assume only two or three values. 104 * They are categorical data, as they can assume only two or three values.
104 * Output 105 * Output
105 * The outputs can also assume only a limited set of categorical values. 106 * The outputs can also assume only a limited set of categorical values.
106 107
107 # Model of Greetings: Features, mapping discriminants, classes, and possible status 108 # Model of Greetings: Features, mapping discriminants, classes, and possible status
108 TODO: FIGURE 2 109 <img src="pictures/classes.png" style='width: 60%; margin-left: 150px;'>
109 110
110 # Model of Greetings: Overview of the greeting model 111 # Model of Greetings: Overview of the greeting model
111 * Greeting model takes context data as input and produces the appropriate robot posture and speech for that input. 112 * Greeting model takes context data as input and produces the appropriate robot posture and speech for that input.
112 * The two outputs evaluated by the participants of the experiment through written questionnaires. 113 * The two outputs evaluated by the participants of the experiment through written questionnaires.
113 * These training data that we get from the experience are given as feedback to the two mappings. 114 * These training data that we get from the experience are given as feedback to the two mappings.
115
116 # Model of Greetings: Overview of the greeting model
117 <img src="pictures/model_overview.png" style='width: 75%; margin-left: 120px;'>
114 118
115 # Greeting selection system training data 119 # Greeting selection system training data
116 * Mappings can be trained to an initial state with data taken from the literature of sociology studies. 120 * Mappings can be trained to an initial state with data taken from the literature of sociology studies.
117 * Training data should be classified through some machine learning method or formula. 121 * Training data should be classified through some machine learning method or formula.
118 * We decided to use conditional probabilities: in particular the Naive Bayes formula to map data. 122 * We decided to use conditional probabilities: in particular the Naive Bayes formula to map data.