As announced at the end of April, it’s time again for our popular annual feature: Seattle’s Most Dangerous Intersection. Last year, five options received over 15 percent of the vote, and many of our readers shared their own personal horror stories about the eventual winner: Denny Way & Terry Avenue. This year’s nominees for worst intersection in the city are:
- 5th & Stewart
- N 50th St & Phinney Ave N
- N 50th St / Stone Way N / Green Lake Way N
- Leary Way NW & 20th Ave NW
- Montlake Blvd & 520
- Montlake Blvd & NE Pacific St
- Rainier Ave S & I-90
- Rainier Ave S & S Henderson St
- Ravenna Ave NE & NE 54th St
- Sand Point Way & 50th Ave NE
Many of these sites have hazards in common, such as wide lanes, minimal traffic control devices, multiple lane changes, outdated layouts, blind corners, and long wait times for walk lights. Several are within busy neighborhoods that are typically teeming with tourists, bicyclists, and people walking their dogs as well as locals heading to their favorite restaurant, checking out the latest bar, or rushing to catch a bus. Other challenges include impatient rush hour drivers, speeding near schools, and frequent construction that results in rerouting foot or bike traffic.
With a national average of one crash-related pedestrian death every two hours and more than 150,000 pedestrians seeking treatment in emergency departments for non-fatal crash-related injuries, it’s obvious that there’s room for improvement in pedestrian safety. One of the newer questions revolves around driverless cars and the effect they might have on the public health issue. Cars are getting increasingly automated, and there are some big brands behind that development — including Google, Tesla, Toyota, and Apple.
Although some people like the idea of a car that does all the work, not everyone wants their ride to be controlled by sensors and software. A survey conducted by Insurance.com found that 22.4 percent of people would buy a vehicle with fully autonomous capabilities, but an almost equal number — 24.5 percent — said they would never consider it. However, when saving money on insurance entered the hypothetical, so did acceptance. An 80 percent discount on car insurance saw a rise in the “very likely to buy” rate to 37.6 percent and a fall in the “never” rate to 13.7 percent.
Truth is, computers have been slowly taking over our cars whether we realized it (or liked it) or not. Anti-lock brakes, back-up cameras and collision warning systems are just a few of the computer-controlled features widely available on today’s vehicles. In fact, the National Highway Traffic Safety Administration has a system that categorizes cars based on how much automation they have:
- No-Automation (Level 0): The driver is in complete control of the vehicle at all times.
- Function-specific Automation (Level 1): Automation of one or more specific control functions. Examples include electronic stability control or pre-charged brakes.
- Combined Function Automation (Level 2): At least two primary control functions are automated at the same time (e.g., cruise control in combination with lane centering).
- Limited Self-Driving Automation (Level 3): The driver can cede full control of all safety-critical functions under certain traffic and environmental conditions, but there must be sufficient transition time for the driver to take over.
- Full Self-Driving Automation (Level 4): The vehicle performs all safety-critical driving functions and monitors roadway conditions for an entire trip, including occupied and unoccupied vehicles.
While perhaps useful for helping people transition into embracing the technology, these five categories also demonstrate just how complicated it is to have an automated car handle traffic situations and to assign fault in an accident involving one. For example, if a car’s computer fails to give a lane-departure warning when a driver drifts out of his lane, and the car hits someone standing on the corner, who is responsible? The car designer? The software programmer? The car manufacturer? The driver? The owner who didn’t maintain/service the computer?
Of course, it may seem a tad pointless to discuss liability in relation to self-driving cars, since data has shown that human error is responsible for 95 percent of all car accidents. It has been estimated that that if just 10 percent of the cars on the road were self-driving, accidents would be cut by 211,000 a year, deaths would drop by 1,100 and the economic savings would reach $22.7 billion. When the percentages flip and 90 percent of the cars are autonomous, it would avoid 4.2 million accidents, spare 21,700 lives and save $450 billion.
As an example of how autonomous vehicles (AV) are faring, Google’s self-driving cars have logged more than 1.3 million miles since 2009 and, as of January 2016, have been involved in only 17 crashes – all of which were caused by human error. All that changed in February when a Google AV caused a crash with a bus after it had trouble navigating sandbags that prevented it from making a turn. Although the damage was minor and no one was injured, the crash was an important lesson about the issues autonomous cars may experience when there are unpredicted, random changes in the driving environment (a point to be carefully considered where pedestrians are concerned).
However, perfection isn’t realistic, and the small number of crashes that might be caused by self-driving cars may be worth the large number they would prevent. Self-autonomous vehicles offer improved safety to pedestrians by eliminating the human factor. Their driving skills aren’t affected by distraction, intoxication, fatigue, illness, or road rage. They are fitted with cameras, lasers, and radar that put a complete picture together at all times, in all directions, anticipating others’ movements and identifying problems on the road ahead.
The prospect of self-driving vehicles is a likely near-term reality, yet cities in the U.S. are woefully unprepared for their arrival. Despite the fact that federal law requires metropolitan planning organizations to produce regional plans every four years that look at least 20 years into the future, a researcher who recently examined the plans for America’s 25 biggest metropolitan areas found that none of them had substantially dealt with autonomous vehicles in the areas of liability, transportation planning, or engineering standards. It seems that while autonomous vehicles could improve pedestrian safety, this future promise does not change today’s risk – or that of any time soon, if those in charge don’t start tackling the impact of self-driving cars on the urban landscape. Just one more reason to continue with our efforts for improved sidewalks and the pedestrian master plan.