In the first installment of this story, I described an accident in which a motorcycle was rear-ended by a Tesla operating in “Autopilot” mode, a political scandal in Germany as that country’s equivalent of NHTSA told the German transport minister that there was no place for “Autopilot” on German roads, and independent research here in the United States that suggests Forward Collision Warning (FCW) systems will often fail to warn drivers that they’re about to hit us.
I pointed out that when the Department of Transportation and the National Highway Traffic Safety Administration released a sweeping Federal Highly Automated Vehicles Policy document last year, the word “motorcycle” didn’t even appear in it. Around the same time, the National Safety Council and DoT announced the “Road to Zero,” — a plan to virtually eliminate road traffic deaths, and which places a lot of emphasis on the improved safety promised by HAVs. The American Motorcyclist Association was peeved that there’d been no motorcycle industry input. More than a few bikers speculated that one “zero” on the Road to Zero was the number of motorcycles on the road. Even AMA President Rob Dingman has written about “…the possibility that in the rush to deploy this technology, we will see motorcycles excluded from certain roadways, or worse, banned… altogether.”
Are HAVs really that much of a threat to us? Either on the road or politically?
It’s true that the Advanced Driver Assist Systems currently on the market aren’t doing a great job of helping drivers to avoid collisions with motorcycles, but there’s a ton of research and development work that will probably make us safer in the future. And organizations like the AMA and Motorcycle Industry Council are now alert to HAVs and working to ensure that policies and technology will accommodate motorcycles.
Of course some motorcyclists think that the current systems are not ready for public roads. Why, they wonder, are we approving vehicles with FCW systems that don’t warn of motorcycles? To answer that question, I need to explain that U.S. federal agencies have adopted the Society of Automotive Engineers’ six-level standard for defining the degree of automation of the driving role.
- At SAE Level 0, the human driver does everything;
- At SAE Level 1, an automated system on the vehicle can sometimes assist the human driver conduct some parts of the driving task
- At SAE Level 2, an automated system on the vehicle can actually conduct some parts of the driving task, while the human continues to monitor the driving environment and performs the rest of the driving task;
- At SAE Level 3, an automated system can both actually conduct some parts of the driving task and monitor the driving environment in some instances, but the human driver must be ready to take back control when the automated system requests;
- At SAE Level 4, an automated system can conduct the driving task and monitor the driving environment, and the human need not take back control, but the automated system can operate only in certain environments and under certain conditions; and
- At SAE Level 5, the automated system can perform all driving tasks, under all conditions that a human driver could perform them.
It’s basically up to manufacturers to declare what level they’re operating at. According to Tesla, its cars are Level 2 for now. Per DoT and NHTSA, that means the driver basically “must” keep his or her hands on the wheel and pay attention.
As far as federal agencies are concerned, those Teslas and dozens of other cars that will stay in lane and operate on their own in stop and go traffic have “advanced driver assistance systems” (ADAS). They’re not technically or legally considered to be highly automated vehicles. HAVs are defined as vehicles operating at SAE Level 3 or higher.
Clearly, some Tesla drivers didn’t get the memo and operate them as if they were capable of Level 3 or 4. Maybe they get that idea from Tesla’s own promotion of features like “summon,” in which the car operates without a driver at all — it’s a feature you’re supposed to do only on your own property, but… c’mon.
So technically, you can’t just go and buy a HAV from any major car maker just yet. You’ll be able to buy one soon; perhaps as early as the 2020 or 2021 model year. That’s not to say you won’t share the road with them even sooner.
If you live in Silicon Valley — or indeed anywhere in California or Nevada, or Pittsburgh, or Tempe, Arizona — there is an excellent chance you’ve already shared the road with HAV cars being tested on public roads by Waymo (aka Google), Apple, Uber and every major car maker, as well as heavy trucks built or adapted by Otto, Volvo and others. Those vehicles, which so far are operating with engineers and/or test drivers ready to take over the controls, have obviously encountered many motorcycles out in the real world, so there’s lots of data about the way current autonomous vehicles deal with us — but manufacturers aren’t usually willing to share it with Mark Gardiner (or you.)
One way developments become public is through patent filings. Ford recently filed a patent on a system that uses a mixture of cameras and microphones to detect motorcycles approaching from behind while lane-splitting. I’m not sure whether Ford’s engineers realize that by the time that system’s available to the public, half the bikes on the road may be electric.
Anyway, microphones are optional. Christian Lauterbach — an engineer working on Google’s Waymo autonomous vehicle project who happens to also be an AMA member — recently told American Motorcyclist magazine that he has no qualms at all about riding his BMW K 1600 GT on the same roads where Waymo’s testing vehicles, and that Waymo’s robotic drivers already move over to allow lane-splitting motorcycles to pass. So if you’re one of the conspiracy theorists who suspect that once cars drive themselves, “the government” will just ban motorcycles, maybe you should take heart, because the people designing the HAVs of the future imagine sharing the road with motorcycles.
Autonomous versus connected
The most important safety advance available to motorcyclists is, thus far, mostly confined to mock road environments in special test facilities like the University of Michigan’s Mcity near Ann Arbor, Michigan, and Gomentum Station, which is built on the site of an abandoned naval weapons center near Walnut Creek, California. Gomentum bills itself as “the largest secure Connected Vehicle (CV) and Autonomous Vehicle (AV) proving grounds in the U.S.”
“Connected” is a key word for motorcyclists, because right now, the autonomous vehicles on the road are really autonomous — as in “operating by themselves.” Like human drivers, they look around and struggle to interpret what they’re seeing. As smart as those Teslas already are, they’re still just trying to guess the intentions of all the other drivers around them. And of course, those of us on motorcycles are smaller and even less predictable; we accelerate faster, and often don’t limit ourselves to marked lanes.
What if cars and their drivers didn’t have to guess where we might show up, or what we plan to do in the next few seconds? Luckily for us, pretty much everyone who is actually shaping the HAV future expects the auto industry to standardize on vehicle-to-vehicle (V2V) systems. So future HAVs will have all same cameras, radars, lidars, etc., but they’ll also all carry a wifi “beacon” that will communicate with nearby vehicles.
The Intelligent Transportation Society of America was incorporated way back in 1991. The Car 2 Car Communications Consortium was formed in Europe in 2005. General Motors demonstrated a V2V safety system using Cadillac cars in 2006 — and some high-end 2017 CTS models have a V2V system (although right now, they can only communicate with other CTSs.) Last year, the DoT proposed a rule, Federal Motor Vehicle Safety Standard No. 150, that would mandate such systems on all four-wheeled passenger vehicles as well as cars and light trucks.
FMVSS No. 150 is a big bowl of chunky alphabet soup. The important gist of it is that the feds have called for passenger cars, light trucks, and heavy trucks to receive and transmit basic safety messages (BSM) on a dedicated short range channel (DSRC), using Public Key Infrastructure- (PKI-) based Security Credential Management System (SCMS).
Every existing car manufacturer is rushing ahead with R&D into HAVs in general and V2V systems in particular. So are Google, Apple, and IBM. But there’s not universal agreement about the timeline for adoption, how or even if such systems should be mandated and regulated, or on what information exactly should be transmitted on the 5.9 gigahertz spectrum that the FCC set aside for DSRC.
The DoT wants 100 percent of new cars and trucks equipped with DRSC-PKI systems by 2023. By 2028, the feds want all such vehicles to also carry a suite of safety apps that include Intersection Movement Assist (IMA) and Left Turn Assist (LTA). Although the proposed rule mentions the potential to improve motorcycle safety, motorcycles aren’t currently included amongst the vehicles that must have DSRC-PKI systems.
We should want them. The motorcycle industry was slow to come around to the idea that such systems are, first, inevitably going to be adopted by the auto industry, and, second, will dramatically improve rider safety.
Not surprisingly, BMW was one of the first car companies to research the safety implications of V2V systems that included motorcycles, demonstrating a system that prevented a BMW 5-series car from turning left across the path of an oncoming R 1200 GS in 2011.
But, it was not until 2015 that BMW Motorrad, Honda, and Yamaha announced the creation of the Connected Motorcycles Consortium. Since then, Kawasaki and KTM have joined as full members, but pretty much every major moto brand has at least indirect input through ACEM — the European Association of Motorcycle Manufacturers — representing Ducati, MV Agusta, Peugot Scooters, Piaggio, Triumph, as well as Bombardier Recreational Products, Harley-Davidson Europe, Polaris, Suzuki International Europe and Yamaha Motor Europe.
It’s not clear how, exactly, motorcycles will operate in the V2V future, but it’s already impossible to overstate the safety implications of V2V systems for motorcyclists. It’s inevitable that human riders and our motorcycles will become active participants — with robotic drivers — in smart traffic.
What about HAM: Highly autonomous motorcycles?
If you’ve been following the news about autonomous vehicles, you’re aware that last year, Waymo (aka Google) sued Uber. Google’s claim was that the ex-head of Google’s HAV project, an engineer named Anthony Levandowski, took proprietary information with him when he left Google to join Otto, a company that has since been acquired by Uber. Uber subsequently fired Levandowski, but not before the kerfuffle made him the world’s most famous HAV engineer.
You know what Levandowski’s first HAV was? An autonomous dirt bike he called “Ghost Rider.” It was the only motorcycle entered into one of the early Defense Advanced Research Projects Agency (DARPA) Grand Challenge contests for self-driving vehicles. (You may be familiar with another widespread technology that got its start as a DARPA project: The Internet.)
Although Ghost Rider didn’t come close to winning DARPA’s $2 million first prize, a search of the Institute of Electrical and Electronics Engineers’ web site brings up several papers devoted to autonomous motorcycles. Honda recently demonstrated a motorcycle that balances itself at slow speeds — the first “Honda Riding Assist” feature. And of course Yamaha’s already demonstrated Motobot, a motorcycle-riding robot they built in conjunction with SRI International.
AutoRD is a British start-up run by a guy named Torquil Ross-Martin. He designed Heathrow Airport’s Terminal 5 autonomous transport pods and ran the Tata truck company’s HAV research. But since his first love has always been motorcycles, it was inevitable that he’d circle back to two-wheelers. He currently has a BMW C1 scooter that’s been retrofitted with a suite of advanced rider aids.
Ross-Martin’s about to put up a new web site and seek investors. He was not eager to leak much information about his project for this story, but he foresees the day when your motorcycle will react by itself when the car in front of you suddenly brakes, or even countersteer by itself in order to prevent you from running wide on Angeles Crest.
I think that would be a good thing for most riders, but when I talk to American motorcyclists about autonomous motorcycles, they’re dismissive. “We ride motorcycles precisely for the joy of controlling them,” they tell me.
Skeptics point out that many American riders still don’t even want ABS, in spite of overwhelming evidence that it makes them safer on the street. When the DoT and NHTSA decided HAVs were the cornerstone of future road safety, the AMA went as far as to release an official statement that basically said, “You can take my handlebar away when you can pry it from my cold, dead hand.”
That’s tough talk until motorcyclists look over at dedicated HAV lanes in which traffic moves more efficiently than human drivers (and riders) can manage. In such a future, I think some riders would be willing to temporarily cede control, in exchange for getting to their destination quickly and almost effortlessly.
Will HAVs expand motorcycling?
I didn’t write all this to convince you to welcome our robot overlords. But I will say that V2V technology will revolutionize more than motorcycle safety, it’ll reboot motorcycle sales — especially in the developed world — if the motorcycle industry can steer the discussion.
In my 50-or-so years of riding, I couldn’t possibly count the number of conversations I’ve had with people who would love to ride, but who are scared of bikes.
I’ve given up trying to rationally discuss motorcycle safety in casual conversation, so I usually laugh and say something like, “Yeah, everything my mother told me about motorcycles was true. And to be honest, most motorcyclists really shouldn’t ride on the street. But if you’re an expert, and vigilant, and wear all the gear, you can be about as safe as the average car driver.”
Then the I’d-love-to-ride-but guys hit me with their closing argument: “It’s not me I’m worried about, it’s the other guy.”
As far as I’m concerned, anticipating “the other guy” is the single most important phase of the road-riding game, and treating other drivers as a completely uncontrollable variable is naïve at best and fatalistic at worst. But most non-riders implicitly believe that even motorcyclists who do everything right run a high risk of simply being taken out, out of the blue, by an inattentive driver.
This is the thing about fears: even an exaggerated or irrational fear is still a real fear. In fact, those are the hardest ones to allay.
Even though it’s going to be decades before most cars on the road are HAVs, the best part about V2V warning systems is that they can be retrofit. It’s easy to imagine that even a decade from now, most cars and bikes will at least have the ability to warn of, if not actively prevent, collisions between inattentive drivers and riders.
There are millions of potential riders who crave the sense of freedom motorcycles offer, who’d like nothing more than to filter through stop and go traffic on their commute, and who would love to always find a parking spot near their destination — to say nothing of being able to rock a bitchin’ black leather jacket and actually have an answer when admiring strangers ask, “What do you ride?”
You know what prevents those people from even entering a motorcycle dealership? The idea that sharing the road with cars and trucks is akin to playing Russian Roulette. The motorcycle industry may have more misgivings about HAVs than the car industry, but it should welcome our dashboard nannies, because we need the millions of new riders who will emerge the moment they feel safe on two wheels.