Welcome/Farewell displays…dynamic turn signals…road-projected turn signals…road-projected reversing lamps…road-projected lane guidance lines…autonomous-mode indicator lights…variable-message displays that can show a vehicle’s state of charge, warn following drivers of emergencies up ahead, and provide any number of other message content…numerous new interior lights…and that’s not even to mention perhaps the crown jewel of them all: ADB in its many, increasingly capable forms. It’s no secret that whole new kinds of vehicle lights are proliferating in the minds of engineers, designers, stylists, and safety practitioners. We give them heavy coverage here in DVN, because they represent a central wave of the future of vehicle lighting.
Aside from their functionalities, most all these new lights have significant styling punch and power. But at base, most car lights are safety items; even those not intended to directly provide a safety function have considerable power to improve or degrade traffic safety. Lights can inform or misinform. They can warn or distract. They can clarify or confuse. They can work in harmony with the other lights around them on the same car and others, or they can clash—overwhelm—drown out—contradict.
Most of a century ago when cars were getting their first serious lights, the questions of what to do and how to do it were, in many ways, quite a lot simpler. There was no established practice to worry about violating, what regulations existed were being written in time with the development of the lamps themselves, and the state of technology and technique constrained the possible implementations. The automotive world looked to broad conventions already in place for the likes of railways, to arrive at general guidelines. White light to the front and red light to the rear, for example, is a deeply ingrained, foundational plank. So much so that even though rigourously scientific research in the 1960s and ’70s found green is probably a better colour than red for the rear position (tail) light function, and found no problems caused by mixing red-taillight and green-taillight vehicles in traffic, there was practically no possibility to swerve from the red-taillight convention. It had simply been baked into the cake much too long ago, and so established practice won out over scientific rectitude.
Less seismic squabbles cropped up over the years, and they still haven’t been fully resolved—what colour should the rear turn signal be? The front position lamp? The rear side marker light?—but these kinds of questions are rapidly being dwarfed by the intricate questions of how to add new lights beneficially, without spoiling existing lights or making new problems. Theoretically, of course, the answer is “do the science!”. But what level of science is appropriate? American regulators want extensive, rigourous study demonstrating that a new kind of light doesn’t negatively affect safety. That’s a rational, reasonable approach, but it brings its own practical limitations: if a new kind of light doesn’t meet existing regulations, it can’t be deployed beyond a small scale, making it very difficult to gather realistic data about the new light’s safety effects.
But it’s easy to go too far the other direction, too, and argue for a new kind of light based on guesses, assumptions, hopes, wishes, dreams, commercial intents, and inadequate or improper kinds of science. Take sequential turn signals, for example: as long as there have been sweeping turn signals, there has been commercial promotion of the idea as a big safety improvement, and there has been grumbling in regulatory and technical-standards boards that the claims for better safety aren’t based on any real or apposite data. Which side is right? Maybe both, to some extent; almost every question raises at least one more question. If one car has turn signals that illuminate sequentially in a more-or-less horizontal line in the direction of intended movement, and the intensity and dwell time are appropriate, maybe that turn signal is at least as effective as a conventional on/off blinker. If one car out every ten has such signals, maybe that’s still okeh. But thorough scrutiny of real crash data has revealed that when most cars have a CHMSL, vehicles without one are at elevated crash risk. The same is true of DRLs. So at what point—if any—does the mix of dynamic and static turn signals begin to make problems? Does the overall safety benefit go negative for some vehicles? For all vehicles? We don’t know, and yet sequential turn signals are on the roads. And if turn signals that illuminate sequentially are at least as good as on/off blinkers, does that necessarily mean turn signals that extinguish sequentially are equally beneficial, or at least no worse? This very debate has just taken place in the SAE Lighting Systems Group, with decisions made but no new data presented or assessed—which means the decision was made based on conjecture, speculation, rationalisation, and opinion. Is it correct? Maybe or not, but these kinds of decisions can be very difficult to reverse once they’ve been made.
Naturally, any inventor wants the validation and remuneration of seeing their new light on every car. Likewise, a company proposing a new kind of car light stands to gain quite a bit if their new light is allowed or required: a product-differentiation lever, a revenue stream, other kinds of fame and fortune. The point of this article isn’t to accuse or judge any car company or expert group for the decisions and claims they’ve made, but to encourage discussion: just what kind and amount of science is necessary to really justify a new kind of car light?
For an exercise in point, let’s look at one particular proposal for front stop lights: one or two green forward-facing lights that illuminate together with the equipped vehicle’s conventional red rear stop lights. The photo here shows an equipped vehicle with its green front stop light lit as it slows to give way to a pedestrian.
There’s a well-polished website to promote the idea. The site has photos and videos, quotes from European Commission white papers, and what purports to be science and research. The site owners, through the tone and content of the website, make arguments that front stop lights are necessary, effective, and cost-effective; theirs is the right way to implement them, and they should be green.
Do the arguments have merit? Well…maybe. Some of them might, sort of. Certainly there are kernels of truth and reasonable propositions present: drivers have an easier time telling what pedestrians are doing than pedestrians have telling what an oncoming car intends to do. AVs will mean something has to replace the longstanding pedestrian/driver right-of-way negotiation via eye contact and hand gestures, and a light signal of some kind seems logical. The European Parliament are in favour of emergency braking display, and they base that favour on a position that “the deceleration of vehicles should be rendered easier for other road users to perceive by means of clear signal lights on vehicles”. A front stop light could give drivers more opportunity to take evasive manœuvres to avoid a rear-end collision. Those all sound like reasonable propositions.
But it is a long leap of faith to go from those general premises to an assertion that this outfit’s green front stop lights should be on all vehicles. Throughout the site and its documents, guesses and opinions and survey responses are overinflated into—and presented as—scientific findings. Phrases like “longitudinal field test” are used to make a loose collection of a few subjectively-reported anecdotes sound like a solid set of data. The argument for the front stop light’s cost-effectiveness amounts to “It costs only about €30 to equip a vehicle, so there’s no reason for the purchase price of the vehicle to increase”—anyone who has ever had even the remotest brush with a car company’s cost-control department knows that every cent is consequential; a €30 increase in build cost is certainly not accepted with a shrug, nor is it ever assumed the automaker will simply eat this cost and not raise the price of the car.
And then there is the matter of colour: why green? Well, because all the other possible colours are unavailable or unsuitable: white would get lost in the existing white DRLs and headlamps. Amber is for turn signals. Blue is for emergency vehicles. Red is for the rear of the car. And therefore: green, support for which takes the form of a paper musing on the psychological effects of red and green, which is then overtorqued into an argument that because green is associated with forward motion and words like ‘go’, it’s the appropriate colour for a front stop light.
Oops, but autonomous-mode indicator lights are turquoise (i.e., blue-green), a colour closely adjacent to green; how easy would it be to confuse the two? And green is already used in the traffic system to mean “go”—now this proposal wants to use it to mean “stopping”; what will be the effects on traffic signal clarity? On the speed and accuracy with which drivers react to green traffic lights? How many drivers will make a mistake when a green light comes on in their field of view? How many of those mistakes will result in crashes? And it seems likely, perhaps inevitably so, that some number of people would rely too heavily on the front stop light and enter the path of a vehicle that isn’t actually slowing or stopping enough to avoid a collision. Will any of these effects degrade, negate, or reverse any safety benefit that might come from the front stop lights as proposed? Anybody can guess, hope, wish, and opine, but we cannot know without rigourous, realistic science.
And the same probably goes for the other new kinds of light being proposed from all corners of the autosphere. Lovely ideas, most of them, but that’s not sufficient; their merit must be proven by reference to what actually happens in the real world as it actually exists. Scientists, start your engines!