**3.2.5 Visualization of unifiable states**

20 Will-be-set-by-IN-TECH

Fig. 16. Extension of the state-transition model using the extension-by-unification method.

In the extension-by-unification method, we extend the state-transition model by the following

2. Unify a state with the same ID between the existing and the new state-transition models.

The existing state-transition model *A* can be represented by state *S*, state transition *γ*, and

,*O*, *γ*� , *λ*� ,*s*�

*<sup>t</sup>*,*<sup>i</sup>* (*s*�

*<sup>t</sup>*,*<sup>i</sup>* (*s*�

*A*�� =< *I*, *S*��,*O*, *γ*��, *λ*��,*s*��

*<sup>t</sup>* ∈ *S*�

*<sup>t</sup>* ∈ *S*�

*A* :=< *I*, *S*,*O*, *γ*, *λ*,*s*<sup>0</sup> > (22)

*S*�� = *S* ∪ *S*� (24)

*st*,*<sup>i</sup>* = *γst*,*<sup>i</sup>* (*st* ∈ *S*, *i* ∈ *I*) (25)

*st*,*<sup>i</sup>* = *λst*,*<sup>i</sup>* (*st* ∈ *S*, *i* ∈ *I*) (27)

<sup>0</sup> = *s*0, the extended state-transition model *A*�� will be as

<sup>0</sup> > (23)

, *i* ∈ *I*) (26)

, *i* ∈ *I*) (28)

<sup>0</sup> > (29)

. First, we calculate state

**3.2.4 Extension-by-unification method**

This process is illustrated in Figure 16. Here, we formulate the above process.

1. Develop a state-transition model to realize a new function.

Similarly, the new state-transition model *A*� is represented as follows:

We accumulate the state-transition model *A*�� by unifying *A* and *A*�

Next, the transition between the state *S*�� is calculated as follows:

*γ*��

*γ*�� *s*� *t* ,*<sup>i</sup>* = *γ*� *s*�

*λ*��

*λ*�� *s*� *<sup>t</sup>*,*<sup>i</sup>* = *<sup>λ</sup>*� *s*�

<sup>0</sup> to be *s*��

*A*� :=< *I*, *S*�

procedure:

initial state *s*0:

as follows:

Here, *S* ∩ *S*� �= ∅.

By defining initial state *s*��

follows:

By using the above algorithms, the possibility of unification between scripts can be identified as "Unifiable," "Unifiable (Occurrence of isolated state)," or "Conflict". Similarly, scripts can be classified as "Executable" or "Unexecutable." By comparing a script and an adaptor definition for the existing scripts, we can obtain a list of scripts annotated with 6 (3 × 2) classes. Our script management server displays the above list at the bottom of each wiki page. By displaying the list, the developer can easily find a script that can be included in his/her current application.

Figure 17 shows overview of the editing system and Figure 18 shows example of using the web based interface.

Fig. 17. Overview of the editing system.

**4. Summary**

**5. References**

In this chapter, we have introduced our health exercise robot TAIZO and the development background. TAIZO has the function to accept commands by voice and keypad, and demonstrates by using its body in collaboration with human demonstrator. Through the experiment, we have measured not only the error rate of both voice and key inputs, but also, the demonstrative effect of using each method. Through real demonstrations, we have confirmed that the robot is effective for giving the incentive to engage in health exercises. We have also discussed about some practical problems to make difficult the development of communication function for humanoid robot. By introducing the component based architecture and extension-by-unification method to develop the dialog script, scripts created in the past can easily be reused in the new application. In the conventional extension-by-connection method, the developer had to develop each function in turn, because

Speech Communication with Humanoids: How People React and How We Can Build the System 187

it did not support the "merging" of scripts that had been developed simultaneously.

more in the feature, by developing practical techniques as described in this chapter.

Communication, [in print] (2009)

Collection, Harvard Business School Publishing

Telecommunications 2006, pp.143–150, 2006.

Dialogue Systems, IJCAI-2001, pp.64–69, 2001.

SEAT,SAT: AIST-OpenRTP Project, http://openrtp.jp/seatsat

Robotics, 5(1), pp.67-86, 2008.

vol.3, pp. 442–447 (2000)

Conference on Robotics and Automation, pp.1416-1421

Notes in Computer Science, Springer, 2008, pp.109–124.

We believe, in future, how advanced the computer graphics are, humanoid robots still keep its strongness to affect the human to move. We hope such powerful medium would be used

Y.Matsusaka, H.Fujii, I.Hara: Health Exercise Demonstration Robot TAIZO and Effects

Y. Moon. (2001). Sony AIBO: The World's First Entertainment Robot," in The Harvard Case

K. Wada, T. Shibata, T. Saito and K. Tanie. (2002). Robot Assisted Activity for Elderly People

Y. Matsusaka, "History and Current Researches on Building Human Interface for Humanoid

K. Ganeshan, "Introducing True 3-D Intelligent Ed-Media: Robotic Dance Teachers," in

Y. Katagiri, T. Takahashi, Y. Takeuchi, "Social Persuasion in Human-Agent Interaction,", in

K. Hayashi, T. Kanda, T. Miyashita, H. Ishiguro and N. Hagita, "ROBOT MANZAI -Robot

H. Hirukawa, F. Kanehiro, S. Kajita, "OpenHRP: Open Architecture Humanoid Robotics Platform," in Tracts in Advanced Robotics, Springer, pp.99–112, 2003.

N.Iwahashi: Language Acquisition Through a Human-Robot Interface, In Proc. ICSLP-2000,

D.Roy: Grounded Spoken Language Acquisition: Experiments in Word Learning, IEEE

Transactions on Multimedia. vol.5, no.2, pp. 197–209 (2003)

of Using Voice Command in Robot-Human Collaborative Demonstration, Proc. IEEE/RSJ International International Symposium on Robot and Human Interactive

and Nurses at a Day Service Center," in Proceedings of the IEEE International

Robots," in Modeling Communication with Robots and Virtual Humans, Lecture

Proceedings of World Conference on Educational Multimedia, Hypermedia and

Proceedings of Second IJCAI Workshop on Knowledge and Reasoning in Practical

Conversation as A Passive-Social Medium-," International Journal of Humanoid

a) Overview of the development interface. Visualization of the state-transition model (left), XML based editing panel (right top), real-time annotation of existing scripts (right bottom). Editing task script. When the developer types the keyword "Hello", the existing script from the script database is annotated as "conflict" and suggest to reuse. At this step, the system only accepts 3 ("Hello", "What can you do?", "Come here") phrases.

b) When the developer check the "greet" script, which already contains several vocabulary for greeting, it is unified to the task script. As a result, the developer only had to increment the application specific vocabulary to realize the whole script with many vocabularies. At this step, the system accepts 7 ("Hello", "Good morning", "Good afternoon", "Thank you", "Nice to meet you", "What can you do?", "Come here") phrases.

Fig. 18. Example of using the web based interface.
