Information Technology Reference
In-Depth Information
Figure 5.5.
Example of the diagnosis of hypovigilance of the driver monitoring system device
and comparison with EEG measures (taken from [BOV 05])
Evidently, this type of supervision technology is very complementary to the
assistance devices described in sections 5.3.1 and 5.3.2.
First of all, the levels of vigilance and distraction of the driver directly influence
driving behaviors (in terms of reaction time when faced with an obstacle, for
example, and lateral trajectory management). The receptivity of the human to
information and warnings also has such effects [GRE 06]. These are therefore two
essential issues for the design of cooperative devices for collision avoidance and/or
keeping the vehicle in a secure trajectory.
The real time analysis of driving performance in terms of lateral and longitudinal
control could give rise to inter-device mutualizations (sensors, analysis algorithms,
or diagnosis functions). The three examples of cooperative devices therefore
illustrate that we can not only improve device's respective performances (or to
“make them more robust”), but also reduce the costs of implantation on the vehicle.
There is therefore every reason, ultimately, to try and combine these different
analyses and/or assistance functions. It is the potential mutualization of embedded
“intelligence” as a means of acquisition (measure instruments, for example) required
by these different devices, as well as on the “integrative” purpose of a future
automobile copilot, that we will now discuss.
5.4. Discussion: towards an “intelligent” and “integrated” copilot
The different research projects we have presented only constitute part of the
necessary functions that an automobile copilot should have. We could add numerous
other functions, going far beyond stakes of strict security, such as favoring
economic driving [BAR 06] or increasing the comfort of the driver. As it stands,
most assistance devices are designed, developed and then implanted in the vehicle
