Information Technology Reference
In-Depth Information
Fig. 14.
The final adaptation of the system to the desires of the inhabitants
The reason for this behavior of the system lies in the use of the antibodies that are
specialized to rooms, weekdays, and hours. When an adaptation is required at some
hour, the antibodies responsible for the normal behavior in this hour are not
completely eliminated; rather some of them survive for some time and can be easily
reactivated if necessary. This makes the system able to quickly re-adapt when the
deviating behavior is no longer required, and this is important for a flexible use of the
system in the context of the intelligent home.
6 Conclusion
We have described an Artificial Immune System that has been developed for the
control of an intelligent home. Such a system should be able to learn the normal
behavior of the inhabitants which is assumed to be constant for most of the time. This
assumption is certainly correct for most people. The system must be able to
differentiate between days, times, and rooms in the house. In addition, the use in the
home requires the ability to quickly adapt to spontaneously sent signals from the users
and to re-adapt to the normal behavior later.
We have presented in this paper an implementation of an AIS that satisfies the
needs of the intelligent home and we have demonstrated how it operates by a certain
test scenario that in particular deals with the problem of adaptation to commands
deviating from the normal behavior. We have tested the system in a number of other
scenarios not included in this paper. In these scenarios the regular use, the regular use
with a break of two weeks of no use, the regular use with a change of use after a
number of weeks, irregular use with frequent changes, and the control of several
