Abstract
Recently, researchers have tried to better understand human behaviors so as to let robots act in more human ways, which means a robot may have its own emotions defined by its designers. To achieve this goal, in this study, we designed and simulated a robot, named Shiau_Lu, which is empowered with six universal human emotions, including happiness, anger, fear, sadness, disgust and surprise. When we input a sentence to Shiau_Lu through voice, it recognizes the sentence by invoking the Google speech recognition method running on an Android system, and outputs a sentence to reveal its current emotional states. Each input sentence affects the strength of the six emotional variables used to represent the six emotions, one corresponding to one. After that, the emotional variables will change into new states. The consequent fuzzy inference process infers and determines the most significant emotion as the primary emotion, with which an appropriate output sentence as a response of the input is chosen from its Output-sentence database. With the new states of the six emotional variables, when the robot encounters another sentence, the above process repeats and another output sentence is then selected and replied. Artificial intelligence and psychological theories of human behaviors have been applied to the robot to simulate how emotions are influenced by the outside world through languages. In fact, the robot may help autistic children to interact more with the world around them and relate themselves well to the outside world.
Similar content being viewed by others
References
Andreu J, Angelov P (2013) Towards generic human activity recognition for ubiquitous applications. J Ambient Intell Humaniz Comput 4(2):155–156
Bellman RE, Kalaba RE, Zadeh LA (1964) Abstraction and pattern classification. RAND Corporation
Buckley JJ (1992) Theory of the fuzzy controller: an introduction. Fuzzy Sets Syst 51(3):249–258
Carrino F, Sokhn M, Le Calvé A, Mugellini E, Khaled OA (2013) Personal information management based on semantic technologies. J Ambient Intell Humaniz Comput 4(3):401–407
Chen Y, Chen Y (2006) Affective computing model based on rough fuzzy sets. IEEE Int Conf Cogn Inf 2:835–838
Chen CW, Kouh JS, Tsai JF (2013) Maneuvering modeling and simulation of AUV dynamic systems with Euler–Rodriguez quaternion method. China Ocean Eng 27(3):403–416
Coeckelbergh M (2012) Are emotional robots deceptive? IEEE Trans Affect Comput 3(4):388–393
Ekman P (2003) Emotions revealed: recognizing faces and feelings to improve communication and emotional life, Holt Paperbacks, 2nd edn (March 20, 2007). Henry Holt and Company, New York
Fong B, Westerink J (2012) Affective computing in consumer electronics. IEEE Trans Affect Comput 3(2):129–131
Fullér R, Zimmermann H-J (1993) Fuzzy reasoning for solving fuzzy mathematical programming problems. Fuzzy Sets Syst 60(2):121–133
He T, Chen H (2010) The mining and analysis of affective law. In: International conference on computational and information sciences, pp 1130–1133
Lanatà A, Valenza G, Scilingo EP (2013) Eye gaze patterns in emotional pictures. J Ambient Intell Humaniz Comput 4(6):705–715
Lee CM, Narayanan S, Pieraccini R (2001) Recognition of negative emotions from the speech signal. In: IEEE workshop on automatic speech recognition and understanding, pp 240–243
Lee CM, Narayanan S, Pieraccini R (2001) Recognition of negative emotions from the speech signal. In: IEEE workshop on automatic speech recognition and understanding, pp 240–243
Lin S, Zhigang L (2012) Generation of basic emotions for virtual human in the virtual environment. In: IEEE symposium on electrical and electronics, engineering, pp 585–588
Mamdani EH, Assilian S (1975) An experiment in linguistic synthesis with a fuzzy logic controller. Int J Man Mach Stud 7(1):1–13
Negoita C (1981) The current interest in fuzzy optimization. Fuzzy Sets Syst 6(3):261–269
Pedrycz W (2010) Human centricity in computing with fuzzy sets: an interpretability quest for higher order granular constructs. J Ambient Intell Humaniz Comput 1(1):65–74
Scherer KR (2005) What are emotions? And how can they be measured? Social Sci Inf 44(4):695–729
Sugeno M (1985) An introductory survey of fuzzy control. Inf Sci 36(1–2):59–83
Wu D (2012) Fuzzy sets and systems in building closed-loop affective computing systems for human–computer interaction: advances and new research directions. In: IEEE international conference on fuzzy systems, pp 1–8
Yang M-S (1993) On a class of fuzzy classification maximum likelihood procedures. Fuzzy Sets Syst 57(3):365–375
Yang M-S (1993) A survey of fuzzy clustering. Math Comput Model 18(11):1–16
Young L, Camprodon JA, Hauser M, Pascual-Leone A, Saxe R (2010) Disruption of the right temporoparietal junction with transcranial magnetic stimulation reduces the role of beliefs in moral judgments. Proc Natl Acad Sci 107:6753–6758
Zadeh LA (1965) Information and control. Fuzzy sets 8(3):338–353
Zhao Y, Wang X, Goubran M, Whalen T, Petriu EM (2013) Human emotion and cognition recognition from body language of the head using soft computing techniques. J Ambient Intell Humaniz Comput 4(1):121–140
Zimmermann H-J (1976) Description and optimization of fuzzy system. Int J Gen Syst 2(4):209–215
Acknowledgments
The work was partially supported by TungHai University under the project GREENs and the National Science Council, Taiwan under Grants NSC 102-2221-E-029-003-MY3, and NSC 100-2221-E-029-018.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by A. Castiglione.
Appendices
Appendix A
The sentences collected in the questionnaire are as follows. “Please assess emotional values from the viewpoints of a girl if the following sentences are heard by you. Also, please indicate what the appropriate offset value in your mind for this numerical design is.”
Will you marry me?
\(\mathbf {\ldots }\)
The average value of the 20 offsets collected is \(\pm 2.2(=( {0*0})+( {1*3})+( {2*11})+( {3*5})+( {4*1}))/20\)
we choose the value range between \(-2\) and 2 as the tuning values of newly produced emotions.
Appendix B
Table 6 lists a part of the sentences collected in the Input-sentence database and their corresponding emotional updating variables: happiness (E\(_{\mathrm{h}})\), anger (E\(_{\mathrm{a}})\), fear (E\(_{\mathrm{f}})\), sadness (E\(_{\mathrm{sad}})\), disgust (E\(_{\mathrm{d}})\), and surprise (E\(_{\mathrm{sur}})\).
Appendix C
Table 7 list a part of the content of the output-sentence database. In fact, for the same input sentence and the same level of the same emotion, Shiau-Lu may select different output sentences. For example, when Shiau_Lu is now in its happiness mood, for the input sentence “You look so beautiful!” there are two choices: “You always make me feel happy!” and “Thanks.” The choice is performed based on a random function.
Appendix D
Table 8 lists “Value range” and “Judged as” of happiness, which is a positive emotion, and fear, which is a negative emotion, corresponding to the items along the X-axis shown in Fig. 7.
Appendix E
Several input sentences and their responses are put together as a script in which S is the input sentence, O is the output sentence and all the inputs are given one by one without any delay. So there is no attenuation. The initial emotions are all neutral, i.e., (happiness, anger, fear, sadness, disgust, surprise)=(0, 0, 0, 0, 0, 0), and S= “Do you want to eat worm?” The purpose is to increase the degree of Shiau_Lu’s disgust. Figure 15a summarizes the process of this input sentence. The six emotional updating variables are E\(_{\mathrm{h}}\):-7, E\(_{\mathrm{a}}\):4, E\(_{\mathrm{f}}\):7, E\(_{\mathrm{sad}}\):3, E\(_{\mathrm{d}}\):9, E\(_{\mathrm{sur}}\):-3 (see Appendix B). The six emotion variables (short as E.V.) are then (-7, 4, 7, 3, 9, -3). After a random number is generated for each emotion variable, the tuning scores obtained are (0, 2, 1, -1, 2, 1). Then the tuned E.V.=(-7, 6, 8, 2, 11, -2). After a random number is generated for each of fear and disgust, the fear is neutral and disgust is positive since according to Fig. 7 and Appendix D, 8 and 11 may be neutral or positive. At last, the primary emotion is disgust and positive-disgust is chosen. So O as shown in Fig. 15b is “You make me sick!”
Continuously, as shown in Fig. 16a, S is “You are a beauty.” to increase the degree of its happiness and reduces the degree of its disgust. Now the mood returns to neutral. O as shown in Fig. 16b is “Thanks.” In Fig. 17a, S is “Will you marry me?” which may make it feel happy. O as shown in Fig. 17b is “I know I am cute, but I am a robot.” In Fig. 18, S is “Let’s go skydiving.” which may increase the degree of its fear. At this moment, the mood approaches fear and happiness. In Figs. 19, 20, 21, the sentence “You look so beautiful.” is continuously input three times, consequently the mood changing from fear to happiness and then strong happiness.
Rights and permissions
About this article
Cite this article
Leu, FY., Liu, Jc., Hsu, YT. et al. The simulation of an emotional robot implemented with fuzzy logic. Soft Comput 18, 1729–1743 (2014). https://doi.org/10.1007/s00500-013-1217-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-013-1217-1