At the end of August, Honda unveiled a new system it's been working on to improve the safety of pedestrians who have to cope not only with badly driven vehicles in their vicinity but also their own tendency to wander into roads without looking.
This experimental technology, says Honda, "determines if the pedestrian is in danger of being struck by an oncoming car"; it does this by utilising GPS and DSRC (dedicated short-range communications) in the car and the pedestrian's smartphone to trigger warnings that an impact may be imminent. Loud alarms and onscreen alerts appear on both the phone and in the car, raising the immediate question of whether our reaction to an electronic alert is going to be any quicker or more effective than our reaction to the sound of an oncoming vehicle, or the sight of a pedestrian appearing from behind a parked car.
The announcement came as scientists at Carnegie Mellon University unveiled what they term a "crash-proof car", a modified Cadillac SRX that uses radar and infrared cameras to detect and avoid obstacles. It reflects what we've consistently been told about driverless cars: that their safety record is extraordinarily good.
Google's driverless cars were once revealed to have only had two crashes, both of them when humans were behind the wheel. So for all the ingenuity within Honda's alert system, you can't help but feel that the best solution to safety issues is to remove humans from the equation completely. "The car's electronics are simply more reliable than people," said Professor Raj Rajkumar, at Carnegie Mellon University's transportation research centre, "and will protect drivers from their own bad behaviour."