2 To establish a creative dialogue between a dancer - actor and a system

2 To establish a creative dialogue between a dancer “actor and a system

The objectives of the "Mises en Sc nes" project are to develop an experimental device connecting an dancer-actor with an intelligent graphic and sound information processing system via the most natural and intuitive methods of interaction possible. From the creative point of view, it is a question of setting up a dialogue between the dancer and the information processing system: the gestures and displacements of the first are a proposal which guides or directs simply the musical and other reactions of the second; in a symmetrical movement, the dancer will react in his turn with the answer suggested by the information processing system.

• From the data-processing point of view, it is all the question of the interface between the man and the machine which is thus posed on new bases and potentially rich scource of direction. It acts indeed, for the machine on the one hand to perceive and interpret what the human user proposes to him in a manner which is only partly formalised, and on the other hand to generate in return a multimode response in accordance with the expressed request. A platform made up of intelligent agents and communicating devices in the network and dedicated to the activity of a dancer can be regarded as a possible actor for assistance with the design, the production and the execution of a spectacle.

• One then is lead to new interfaces, controls of data feedback on the produced effect, feeling.

• Relational interactions between actors, actors and directors, actors and public.

• Semantic interpretations taking account of propositional differentiations.

2.1 The experimental device

The human "dancer “actor" who will carry out the performance will be in interaction with an autonomous information processing system, a kind of numerical partner with whom it will have to establish a creative dialogue. So that that is possible, in fact is needed that the system itself perceives and analyses the performance of the human dancer, imagining a "suitable" response to this performance and finally carries out it. That corresponds to the 3 traditional phases which any autonomous agent follows : perception, decision, action (see Figure 23.1).

Figure 23.1: The autonomous information processing system perceives the evolutions of the dancer via adapted sensors, decides on an adapted answer and carries it out

In other words, it is a question of envisaging:

• How the module of perception will represent the performance of the dancer so as to make it possible for the module of decision to choose the answer to be given?

• How the module of decision will represent the answer which the system will have to make thanks to the action module?

In addition, to be able to establish a dialogue between the dancer and the system, it is necessary to install the bases of a common language between the two. This is a question of making a dialogue between two modules of the system or the system and the dancer. It is the necessary in each case to define:

• A common semiology so that each one does include/understand the other (the "what"?).

• A common technical platform so that they can communicate indeed (the "how"?).

From the semiological point of view, the approach adopted consists in analysing in a similar way the various fields of expression concerned with the spectacle imagined ( dance , music, painting and scene setting), in the form of pairs of qualities (for example "heat “cold"). Each pair of qualities can be represented by a quantitative value, and the scene setting of the spectacle could be based on the writing of partitions indicating, in the course of time, the evolution of each one of these qualities. These partitions, making it possible to analyse the dynamics of the interaction between dancer and system, must authorise the three modules of the autonomous system to communicate one with the other.

2.2 The dancer and the communicating physical device

The experimental device such as it was defined initially utilises one dancer and a multimedia information processing system. From the point of view of scene setting, it is however interesting to introduce facilities present in the scene that the dancer can use. The communicating physical facilities, which have usually as a function to offer services to the users, would have here the role of accessories or elements of scene setting credits. These communicating physical facilities are equipped with:

• an interface with the network;

• data handling capacities of information;

• sensors and/or actuators.

These characteristics enable them to interact between them and with the user. In physical interaction with their environment in the broad sense, these communicating physical facilities can:

• modify their state;

• act on the other purposes;

• interact with the user;

• interact with their environment.

The term "communicating devices" gathers in fact a whole range of very different numerical devices, with the various capacities of interaction. Those known as "communicating" can either perceive the action of a user (direction, distance, acceleration, temperature, contact sensors, etc), or restore information for the user, or to carry out actions (screens, engines step by step), or still allow the control of other facilities for the user (remote controls, PDAs, etc). These facilities can become actors, many taking part directly in the dynamics of the performance of the dancer and the spectacle in general. By the use of tags (active or passive badges), these purposes can allow simple localisation of the dancer in the scene. Finally, with more than one dialogue between a dancer and a system of software agents, the spectacle will be composed of a dialogue between human actor, communicating physical devices and software agents of multimedia composition.

2.3 Operational assistance

This tool constitutes a personal assistant in the form of intelligent prosthesis, instrument of capture and shared interpretation model. Three examples illustrate this operational assistance:

1. A capture instrument : interfaces of sensor type, sensing remotely or placed on the body, video cameras or positioners in space constitute many sources of the tracking of positioning, body language and the physical and emotional state of an artist during a representation.

2. A shared interpretation model : the relational returns of feeling of another actor, the director, the witnesses, constitute guides for the artist in scene to know where, in phase or movement, the effects which function best and the environments which increase or decrease emotional resonance .

3. An interpreter : grammars and grids of interpretation of gesture, rhythm, posture , colour and sound make it possible to explore new forms of representation and scene setting.