I’d made the turn when a push from behind momentarily sped my bike up, then the front wheel twisted and the bike fell, sending me over the handlebars. It’s a terrifying feeling, getting thrown from your bike, and there’s a moment in mid-air when you accept that you’re going down. I caught myself with my hands, skinning my palms, elbow, and knees, then looked back at the mid-1980s dilapidated station wagon that hit me.
The damage ended up being only a bent wheel, some blood on my business-casual outfit, and a few weeks with a sore wrist. It could’ve been worse. In 2012, the most recent year of the National Highway Traffic Safety Administration’s comprehensive data crunching, 726 bicyclists died in motor vehicle accidents.
Crashes like mine are now avoidable—if your run into the right car. Big changes are coming to the auto industry: advanced sensor suites capable of detecting cyclists and pedestrians, mass-adoption of bike-avoiding software, and ultimately, the rise of the self-driving car. While the fully robotic car is still a ways off, small advances combined together are poised to reduce car-on-bike fatalities within the next year.
At the moment, Tesla is leading the charge of incremental safety advances. Forget the scary fast 3.2-second zero-to-60 time of the Model S P85D, and look instead at Tesla's new self-driving system called Autopilot. It’s a myriad of sensors and data-gathering processors, including a 360-degree sonic radar to detect blind spots around the car and a special camera system that can read speed limit signs and lane markers.
These systems aren’t new. What's new is the systems integration and consumer-ready roll-out. Nod off behind the wheel, and the car takes over, exiting on the right turnoff, stopping at stop signs, and pulling into your driveway.
If you're on the highway, you can flip the turn signal to merge into the right lane. Your Tesla will notice the car in your blind spot, wait for it to pass, and then change lanes. That’s great for car-on-car safety. It's even better for cyclists who often get caught between turning and through traffic, with cars madly dashing to change lanes, says Dr. Jeffrey Miller, associate professor of Engineering Practices at USC and IEEE member.
Tesla isn't alone. Volvo’s cyclist-specific safety mechanism, which is already on the market, incorporates a radar in the radiator grille and a camera set in the rear-view mirror. The radar senses objects ahead and measures the distance between the object and the car, while the camera picks up on the objects’ pattern of movement and predicts which way the pedestrian, cyclist, or fellow vehicle might go. The system can recognize a cyclist based on speed and profile. Should the cyclist swerve or brake suddenly, the system recognizes the move and automatically stops the car.
While pedestrian avoidance tech has long been a feature of Volvo's cars, spotting an obstacle and automatically stopping the car before impact is a skill vehicles are just starting to get really good at. Even if they appear to be incremental, these advances will soon start making a big difference for cyclists, says Dr. Alberto Broggi, professor of computer engineering at the University of Parma in Italy, who worked with VisLabs to engineer a self-driving car’s successful trip from Parma to Shanghai. “Everyone will be safer with this kind of automatic detection and avoidance systems.”
The incremental advances are poised to stat saving lives immediately, but Google has announced an even loftier goal: a world without distracted drivers where robotic cars navigate without humans.
The company, which has been researching automated driving in earnest since 2009, has its modified Lexus SUV outfitted with suites of sensors, including some that can recognize a cyclist’s hand signal for a right or left turn, and steer the vehicle accordingly to avoid a collision. And in May, the company unveiled its prototype without a steering wheel, brake pedal, or accelerator pedal. The company will need to sort through regulatory hurdles before mass-adoption can ever take place, but some experts are predicting we'll see the first full self-driving and cyclist-avoiding cars in 2020.
Granted, even if autonomous cars become the norm, it will take years before they're widely enough used to have a game-changing effect on road biking. Until then, the advent of self-driving vehicles shouldn't serve as an excuse to stop building bike-specific infrastructure in cities, says Mikael Colville-Andersen, CEO of Copenhagenize Design Company. "There are many good reasons that over 150 cities in Europe have implemented broad areas of 20 mph zones for cars—they save lives, reduce injury, and make cities feel safer for the inhabitants. If autonomous cars show up, fine. As long as they roll around at a speed that fits the urban theatre."
Yes, all cities should strive to look more like Portland. But in the meantime, cars that move in predictable, coordinated patterns while stopping early enough to avoid collisions will make an immediate impact on individual lives. A distracted driver that might’ve led to a fatality will soon result in—at worst—some abrupt braking and a scare for both parties.
“Ultimately, we should think about the road as a place where accidents will be extremely reduced,” Dr. Broggi says. “Accidents between a cyclist and a car will occur only due to strong and unpredictable misbehavior of the cyclist.”
Those of us on two wheels, with human-powered braking and steering, now just need to hold up our end of the bargain. Or at least signal clearly enough for Google's sensors to pick us up.