Driver Distraction Detection

Alain Dunoyer (Jaguar Landrover)

Driver distraction is thought to be a significant factor in road accidents.
It would therefore be very beneficial if a real-time estimation of driver
attention could be derived. The aim of this project is to investigate
methods of achieving such a goal by monitoring already available driver
inputs such as steering wheel angle position, vehicle speed, throttle
depression, lateral acceleration etc…

It is envisaged that the initial phase of the project would consist of a
literature search to investigate potential methods. The second phase will
cover implementation and validation. It is quite possible that a number of
concurrent methods will have to be implemented. Also, it is anticipated that
there is large variation in driver patterns, depending upon age, experience
and driving conditions. It may be that any distraction system would need to
‘learn’ the driving nature of a given driver at the beginning of a journey
to minimise false alerts which would detract from its usefulness.

For this project JLR will provide real-time data for the analysis phase and
will also offer the opportunity to validate any developed methodology
directly into a vehicle if the implementation is feasible within the project
timescale.