How safe are ‘safe driving aids’?

The BBC ran a story on Tuesday this week (25 January) entitled “Motorbikes ‘to get safe driving aids'”.

Rather predictably, the news raised howls of protest when motorcyclists read that:

“Motorbikes could soon be sporting collision detection and other safety features more usually found on cars.”

And that according to researchers:

“the first bike-based safety systems could be appearing on motorbikes within two years.”

The ‘Saferider’ project is a collaborative R&D effort that is part of the European Commission’s Seventh Framework Programme. Participants include Mira, Yamaha, Porsche Engineering and FEMA, among others.

The drive for the project comes from the high and still-rising rate of motorcycle casualties, quoted in the BBC report as 22% of all road accident fatalities, though no details are given about that figure – it’s presumably Europe-wide.

So what’s it all about?

“Saferider takes the driver safety systems that are becoming standard on cars and tries to adapt them to the unique needs of motorcyclists,” said Jonathan Moore, an ITS consultant at Mira, involved in the Saferider project.

“One of the most difficult things is getting the rider’s attention,” he said. “There’s a high level of ambient noise and vibration to deal with and we really don’t want motorcycle riders looking down at the handlebars any more than they need to.”

The project has been looking at a range of systems to get the attention of riders and feed them information.

Many systems revolve around so-called ‘haptic’ feedback, which in plain english relies on the sense of touch. For instance, information can be fed to the rider via aids like vibrating seats and cheekpads in the helmet, or though a throttle system that is not simply passive but gives feedback to the rider.

To most riders this all sounds a bit ‘Tron’, but in fact these kind of systems are in everyday use in other fields.

An example is the “stick shaker” that warns a pilot that the plane he’s flying is approaching a stall condition. Haptic feedback has also been applied to digger technology, so that the operator can ‘feel’ harder objects such as large rocks when excavating.

And of course, virtually anyone who’s played a computer game will be familiar with force-feedback from the joystick, steering wheel or other controller.

In any case, we already use haptic feedback as we ride. We feel the motion and resistance of the throttle, we feel the degree of ‘squeeze’ on the brake levers, and we sense deceleration and acceleration by the apparent weight on our arms and in how we want to slide forward against the tank or back on the seat. We feel the vibration of the motor and the sense of the machine riding over the road surface.

But against the idea that we can feed extra information to the rider, we need to balance that against the fact that it’s possible to overload the rider with input.

We’re already processing a huge amount of information with our eyes as we watch the road around us, and feeding more information via other sensory systems can get tuned out.

Ever driven past a speed camera with the GPS warning “bonging” away madly and simply not heard it?

It’s difficult to train people to receive multi-sensory input. Visual information as dials and lights, as well as heads-up displays in aircraft are likely there for a reason – the info comes in via the eyes.

Nevertheless, there’s no good reason we can’t learn to use alternative feedback systems to gain more information, provided they add to, rather than interfere with, what we already do.

So what are the data-collecting systems involved? What’s the technology that feed us the information?

This is where it all seems to get a little more pie-in-the-sky.

One of the simplest uses GPS. If the rider is travelling too fast to negotiate upcoming bends, then the software alerts them. In theory, it’s little more complex than looking at the GPS as you ride, something I tend to do already as I approach an unfamiliar bend.

The obvious questions are how the software can interpret what speed is appropriate for the bend in the prevailing conditions, and if the ‘approved speed’ will come to be treated as a target speed when the rider should actually be going more slowly.

Plus, of course, you’d have to be 100% confident in the veracity of the mapping system. Judging by the number of times my GPS routes me across green lanes or thinks I’m travelling over green fields because the mapping is out of date, I can see a huge problem with keeping current with even minor changes to road layouts.

Other systems include laser scanners, smart helmet-cameras and radar.

“One system under test based around radar constantly monitors the blind spots around and behind riders,” said Mr Moore. “Vehicles behind or to one side of a bike can be hard to spot because the helmet restricts visibility and riders must remember to move their head regularly to check. We put a motor in the cheek pad of the helmet so if you do not notice the object it will vibrate and give you a tactile warning that there’s something to the right or left.”

If there’s one thing that you notice if you ride a bike first THEN switch to cars and vans, then it’s the lack of permanent blindspots on a bike and the fact that with a half-decent pair of mirrors there’s virtually nothing you can’t see if you only turn your head to look.

Whilst I won’t take the “nothing beats the Mark One Eyeball” line, this really does strike me as massively overcomplicating what is a relatively trivial problem on a motorcycle.

In fact, it’s the reverse that’s the problem. Motorcycles are far more at risk of being sideswiped when they sit in other vehicles blindspots. Educating the rider who is usually ignorant of the car driver’s problems in seeing round the various pillars and headrests is the solution here.

Another system being trialled is a collision detection system, developed by Mira.

Here, a laser detects an imminent impact with an object in the rider’s path, such as an emerging car. At the moment, the automatic braking system ONLY steps in when a collision is inevitable, and only applies 0.3g’s worth of brakes. That’s firm but hardly emergency braking.

The theory is that by using some brakes, the likely severity of the injury will be reduced, but I don’t follow the logic. If the collision is inevitable, then the rider is going to crash anyway, so surely the best way to use the system would be to progressively (so the rider can hold on!) apply the brakes harder right up to the point where the ABS kicks in.

I’ve seen estimates that average riders use 60-70% of available braking force when braking hard in an emergency. In fact, we can achieve over 1g on a good surface, 0.3g seems peculiarly low.

Nor can I see how this will ‘reduce injuries’ except in the odd cases where the rider when confronted by a car in their path fails to brake at all or possibly brakes so clumsily as to lock up and fall off, although as we’re all going to have ABS/combined brakes fitted in the next couple of years that problem should recede with time. It also switches off at more than 10% of lean, yet with that minimal lean angle you can still use a very significant amount of brakes.

The braking system is apparently augmented by electronically activated anti-dive forks, apparently because it improves stability under braking.

I think we’re reinventing a square wheel here.

Anti-dive was all the rage on GP bikes for a year or two in the very early 80s and fed through into road bikes a year or two later. Riding those anti-dive bikes showed me that the fork dive is actually the feedback mechanism which tells you how hard you’re braking. I tried it, turned it down to the minimum setting and promptly forgot about it.

The fad for anti-dive vanished as fast as it appeared as it had no benefits, but added more weight and mechanical complication. There are also far better alternative front suspension arrangements around that separate the forces of braking and steering, as BMW are aware.

The system I did see as potentially useful, though in a way you’d have to be a pretty poor rider to need it, was the distance sensing system, which uses a haptic throttle which gets progressively more difficult to hold the throttle open as you get too close to the vehicle in front – it could teach a few riders to follow at more sensible distances when planning overtakes!

But currently all this technology requires a pannier-full of electronics to analyse the data plus the mechanical aids. Various electronic driver aids are already fitted to cars, but by definition they have the size and space to fit them although the more electronic gizmos fitted to a car, the less reliable it is, in my experience.

Whilst I don’t doubt that miniaturisation will be brought to bear on such technology and the electronics will end up on a circuit board the size of a pin-head, there’s only so far you can go with making mechanical devices smaller – the trade-off with cost is a steep one.

For example, ABS is still a heavy addition to any bike, the adaptive braking system described below will require a servo system and even the electronic anti-dive will need a motor.

Let’s not lose sight of the fact that motorcycles are best left light and nimble, and the more technology you add to a machine, the bulkier it gets and the more you need the technology to control the machine.

Catch 22?

Personally, I’d rather have some research into the effectiveness of a twin rear LED light over the totally inadequate single incandescent bulb my 2009 model bike came fitted with.