Experiment 5, Domain: robotic, Title: pathology, Goal: parameters modification in real-time

The aim of this experiment is to show that the parameters of the auto-adaptive system can be modified in real-time.

Hypothesis: the robot is active in an unknown environment to execute several actions according to goals. A goal is represented by a graph of words with topological links.

Goal: the aim is to play with Alain (with the ball or with the computer). But, with a modification, in real-time, of the parameters of the word "Work", there is a serious alteration of the goal. Suddenly, the robot want to work.

The figure 1 shows the morphology of the system before the modification in real-time. The figure 2 shows the focal point. After the modification, the morphology change in parallel of the focal point (figures 3 and 4).

Figure 1. Knowledge morphology before the modification.

Figure 2. The focal point before the modification.

Figure 3. Knowledge morphology after the modification.

Figure 4. The focal point after the modification.

Experiment 4, Domain: code generation, Title: web design, Goal: to generate an HTML page

Hypothesis: a user who don't know the HTML language want to create a simple web page. He uses the provided ontology

The aim of this experiment is to show the flexibility of the Auto-Adaptive System (AAS). An ontology update is sufficient to give new capacities to the AAS.

For this experiment a new classification has been added to the ontology: it is a part of the HTML language. On the figures 1, 2, 3 and 4, we have the new elements in the classification, all new agents for the new aim and a new experience.

Figure 1. Elements of classification.

Figure 2. List of agents.

Figure 3. Experience linked to the HTML generation.

Figure 4. Agent description concerning an HTML page.

During the processing, we have logs on a Terminal (figure 6). The result of the processing is presented on the figures 5 and 7. The AAS generate HTML language to create a web page. We have to work on the ontology to create more complex pages.

Figure 5. Code generated.

Figure 6. Log of the code generation.

Figure 7. The result of the generation.