Road safety is one of the main concern of public health. In 1998 WHO (World Health Organization) estimated that over 1 million of people died because of road accidents. Studies and analysis found that 20% to 40% of road fatalities are caused or co-caused by mental and physical health conditions of drivers who, despite tiredness and drowsiness, keep driving with a consequent reduction in attention and lucidity. Human factor is estimated to be the sole cause (51%) and co-cause (38%) of road accidents. Actual on board safety systems are exclusively based on monitored conditions in motion of vehicle and its subsystems like tire grip (ABS, ESP and traction control systems). On the contrary, no system for monitoring driver’ watchfulness is now available.
The research focuses road accidents caused by the decreased level of attention of drivers. Under this definition different mental and physical conditions are reassumed: carelessness due to lack of concentration determined by events/objects inside or outside the vehicle, level of attention of reflexes due to lack of rest or to the use of alcohol, medicines or drugs.
The general aim of the research is to investigate new areas to enlarge active safety of vehicles, operating on the relationship between vehicles and drivers, according to analysis of different nature. The project aimed to assess the feasibility of an intelligent and adaptive system able to identify “lack of security” conditions due to a substantial inadequacy between driver mental and physical conditions and vehicle motion conditions.
The project was built up around following phases:
- Enhancement and improvement of a driving simulator in order to take into considerations some other relevant aspects;
- First set of statistically meaningful trials to develop the system for monitoring the driving behaviour;
- Developing of an experienced system for driver’s behaviour self-learning, taking into account also motion conditions;
- Critical analysis and validation of developed diagnostic strategies, through a second set of trials by simulating driving conditions with a low level of attention.
- Preliminary project of a prototype system (with CAN interface) and including all the developed diagnostic strategies that can be transferred into a vehicle.
The developed simulator has a fixed post equipped with on-board interfaces, retro-action systems and it is managed by four PCs working in a network.
The system proved to have satisfying performances, as some parameters confirmed (time for executing one step of calculation).