I don’t consider myself a technophobe and I’m not daft enough to dismiss all passive or active safety systems – I’ve relied on them on occasions. For me, though, systems such as lane-keeping assist exist in a grey area where the car decides it knows best when it doesn't necessarily.
Perhaps we’ll experience even more of this kind of nannying technology creeping in as we progress deeper into the age of self-driving cars. To my mind, we need to avoid any situation where both car and driver decide to be in control at once, because that could also give rise to an opposite scenario where each assumes the other is doing the work.
Opinion: why do people think autonomous cars are scary?
As the recent well publicised (albeit rare) incidents involving vehicles with various levels of self-driving capability show, there is little room for any kind of grey area when control over a two-tonne metal projectile is at stake.
Having said all of this, there is a time and a place when I wholeheartedly buy in to the benefits that lane-keeping assist can bring, and that is when it is working in conjunction with other technology to provide a genuinely useful function.
The Mercedes-Benz E220d estate I recently ran as a long-term test car was equipped with the Driver Assistance package, which combined adaptive cruise control with lane-keeping assist, blind-spot assist and other safety kit to enable the car to effectively maintain its own course.
It wasn’t a truly self-driving system - if you tried to take your hands off the wheel, you received a warning and the system cancelled itself - but it was a level of automation I felt totally comfortable with. On long motorway journeys I could trust the car to maintain speed and distance from other traffic, and to change lane automatically when I activated an indicator.
I guess ‘trust’ is the important factor here. The Merc’s system felt natural, but the important difference was I was telling car that I wanted it to do the bulk of the work, as opposed to it telling me that I was doing something that did not compute with its algorithms. As self-driving cars progress, this feels like an important definition, especially for those of us who enjoy driving when the mood takes us.
Which brings me, in a meandering way, onto the topic of a new book I’ve just finished reading. It’s called Autonomous driving – how the driverless revolution will change the world (Emerald Publishing, £19.99, ISBN: 9781787148345).
The book is written by Andreas Herrmann and Walter Brenner, both professors at the University of St Gallen in Switzerland, and Audi boss Rupert Stadler - who, you might surmise, knows a thing or two about the subject.
Audi A8 to beat Mercedes-Benz S-Class as first fully autonomous car
This book is an academic study, but is pleasingly easy to read and doesn’t get too mired in jargon that might stifle the reader. If you’re interested in the subject matter - regardless of whether you agree with the direction the industry is heading in – it’s a very decent primer, albeit one that’s likely to get overtaken by the pace of progress very quickly indeed, which the authors freely acknowledge.
Almost every facet of automated driving is broken down and analysed. On the topic of ‘trust’ - foremost in my mind, as detailed above – the authors write: ‘People often express the worry that they will be at the mercy of the car and will have no opportunity to influence the driving process… the interaction between the driver and the vehicle must be designed in such a way that the transition back and forth is clear and unequivocal’.