The day when our highways are filled with cars whose drivers can leisurely watch the latest blockbusters, or even take a nap, still hasn’t come. Nevertheless, it’s getting closer and closer, at least judging by the latest self-driving car statistics.
The self-driving car market is extremely competitive, and there are more levels of car autonomy than one may originally assume. At the moment, our roads only have partially independent driving cars. However, soon that will change. The autonomous vehicle global market is expected to reach $36 billion by 2025, with North America owning 29% of all the self-driving vehicles in the world.
But even greater investment and meticulous research are going into data on self-driving car accidents. Sometimes, these accidents are caused because a system reacted improperly, but mostly they’re influenced by the humans outside the vehicle. There have even been instances of people intentionally pointing guns at a passing autonomous vehicle. Moreover, 94% of all road accidents happen due to human error, as reported by the National Highway Traffic Safety Administration.
Must-Know Stats About Self-Driving Cars
- According to one survey, 56% of Americans wouldn’t feel safe in a driverless vehicle.
- There are 5 different levels of self-driving.
- The self-driving Tesla Model S caused a pedestrian fatality in Florida in April 2019.
- The most common accident involving self-driving cars is getting hit from the rear, according to recent self-driving car statistics.
- At the moment, none of the self-driving cars operating on US roads are completely autonomous.
- The Tesla cars with an Autopilot mode may have a tendency toward smashing into emergency response vehicles.
- Uber’s self-driving cars require the most human intervention.
- The more complicated the route it follows, the better an AV’s self-driving system becomes.
- 76% of Americans want Congress to have brakes installed in driverless cars.
- Waymo’s car has excellent test drive results regarding human intervention.
Now, let’s take a closer look at some of the most recent self-driving car statistics.
People Also Ask
How many self-driving cars have crashed?
It depends on how you interpret the question. Based on the sources we’ve mentioned, it could be said that there were 13 serious crashes (including 6 fatalities). On the other hand, the total number of all the crashes and collisions all over the US must be much higher when we consider that there were 28 rear-end crashes in 2018 in California alone. With this in mind, the data on autonomous vehicle accidents indicates that collisions and crashes aren’t uncommon.
How many deaths have self-driving cars caused?
So far, there have been two pedestrian fatalities. One was caused by an Uber vehicle, and the other by a Tesla AV. Additionally, four AV drivers died in self-driving car accidents.
Do we need insurance for self-driving cars?
Self-driving cars need insurance as much as any other type of vehicle. However, at the moment, you can count on the insurance rate being a bit higher, mostly due to their high-tech profile.
What percentage of cars are self-driving?
The exact percentage isn’t known, but it’s expected that the US will be the global leader in the industry. For starters, self-driving car statistics also seem to say that next year there will be a few thousand robotic cars on the roads. That figure is still negligible compared to the millions of regular cars. However, by the year 2035, the US streets will have about 4.5 million self-driving cars roaming around.
Are self-driving cars safer?
At the moment, self-driving cars have a higher rate of accidents compared to human-driven cars, but the injuries are less serious. On average, there are 9.1 self-driving car accidents per million miles driven, while the same rate is 4.1 crashes per million miles for regular vehicles.
Driverless Car Statistics
There are 5 different levels of self-driving.
In general, the public is less concerned with these levels, but they’re very important for the manufacturers of self-driving cars. These vehicles range from level 1, where the car can independently accelerate or steer, to level 5, which implies complete automation in all imaginable conditions.
The rise of autonomous cars may drastically alter the car insurance industry.
On the one hand, the pricier and more technologically advanced the vehicle, the higher the insurance. Nevertheless, if self-driving cars improve drivers’ general safety in traffic, the insurance rates may go down. Moreover, it’s quite probable that car manufacturers will start their own insurance companies.
As for the autonomous vehicles industry, statistics show it rising by 16% globally every year.
The faith—and subsequent investment—in robotic cars is so strong that by 2025 the global value of this market should be worth an incredible $1 trillion or more!
At the moment, no self-driving car operating on US roads is completely autonomous.
Right now there are still no AVs (autonomous vehicles) operating without a human driver. For now, there will always be one, or even two, people present to react to any of the car’s potential warnings.
Official statistics on self-driving cars report that 41 states (and Washington DC) have taken steps to consider legislation on self-driving cars.
So far, 29 states have prepared laws about autonomous vehicles, and in 2018, 18 bills about self-driving cars were enacted throughout 15 states.
Self-Driving Car Accident Statistics 2012–2019
Most self-driving car companies hire two safety drivers for their tests.
While one of them pays attention to the road, the other one is supposed to monitor the system. Uber is the only company that doesn’t use two drivers in its tests. This choice may have played a role in an incident that ended in a fatality.
The most common accident involving self-driving cars is getting hit from the rear, according to recent autonomous vehicle statistics.
This may serve as an argument in favor of self-driving cars since a driver rear-ending another vehicle is often caused because they weren’t giving the right level of attention to the road.
Sideswipes are the second most common accident involving an autonomous vehicle.
Again, these accidents often happen because of mistakes made by the other vehicles’ human drivers failing to properly overtake the self-driving car.
Once driverless cars are four or five times safer than they are now, people will trust them as much as human-driven vehicles.
Due to the current statistics and news reports about accidents with self-driving cars, people are still wary of buying or driving (in) self-driving vehicles.
Based on 2018’s self-driving cars stats, 76% of Americans want Congress to have brakes installed in driverless cars.
(US Public Policy Polling)
The main reason for this decision is safety—but also privacy, as drivers aren’t fully comfortable with a machine being solely responsible for where they go.
Accident Statistics on Google’s Self-Driving Cars
Waymo’s self-driving car has excellent test drive results regarding human intervention.
Waymo started out as a Google project, but the company became its own subsidiary in 2016. And the company has made fast progress. In 2018, Waymo’s test drivers had to react and take control of the car only once every 11,018 miles. That was twice the distance from 2017, and it’s an excellent result regarding the AV competition, too.
Waymo needs to enhance its alarm system to prevent more self-driving cars from having crashes.
During a test in 2018, a Waymo driver fell asleep at the wheel. Despite the alarm signaling that the car was in manual mode, and an extra audio alert, the driver failed to wake up. The car had been in auto mode, but when the driver dozed off, he accidentally pressed the gas pedal, which automatically switched the system to manual. The vehicle crashed into a highway median, but there were no serious injuries.
Google’s self-driving car was hit by a human-driven car.
Waymo isn’t the only company that’s had some autonomous car accidents in its tests. During one of its self-driving tests in San Francisco, a Google van was hit by another car, when the human-operated car swerved to avoid another vehicle. There were no fatalities.
The first accident that was undoubtedly caused by the new Google self-driving car happened on February 14, 2019.
Google’s car (and the driver) estimated the situation wrongly as it believed the bus would stop to let the Google AV continue on its way. The Google car was maneuvering near some sandbags in the street and hit the bus in the next lane. There were no victims.
Stats on autonomous car accidents confirm that Google’s self-driving car has had 24 minor accidents.
Since it started operating on US roads, Google’s self-driving car had only 24 minor accidents in the 2009–2016 period, which is quite praise-worthy.
Accident Statistics on Tesla’s Self-Driving Cars
A Tesla driver was killed in 2018 when the Model X SUV accelerated and crashed into a safety barrier on the highway.
According to his family, Walter Huang, a 38-year-old Apple engineer, thought that Tesla’s self-driving cars were more reliable and safer than regular cars. The company is being sued for the car system not recognizing the obstacle and speeding up, as well as the California Department of Transportation for not replacing a crash attenuator guard.
Tesla’s self-driving car statistics suggest that when their cars are in the Autopilot mode, they tend to smash into emergency response vehicles.
A Tesla S in Autopilot crashed into a parked fire truck in Utah, and its driver even injured her foot from trying to press the brake repeatedly. Tesla is being sued for the accident. Similarly, a Tesla sedan smashed into a parked police SUV. Fortunately, the police car was empty at the time of the incident, and Tesla’s driver suffered minor injuries.
In April 2019, the self-driving Tesla Model S caused a pedestrian fatality in Florida.
This is one of the most disturbing Tesla autonomous vehicle accidents. The investigation is still ongoing, and it’s not clear if the car was on Autopilot when the incident happened. The car entered a three-way stop sign intersection without stopping, hitting a parked pickup truck, which started spinning and then hit the woman.
Tesla often blames its drivers for Autopilot crashes.
Tesla is defending its Autopilot system by stating that the drivers involved in these accidents should have been more attentive to the autonomous vehicle’s signals. In most of the cases, the steering wheel did not confirm human touch before the accident.
Accident Statistics on Uber’s Self-Driving Cars
Self-driving car statistics indicate that Uber’s self-driving cars need human intervention the most.
At the moment, Uber is losing to the competition, as Google, General Motors, Toyota, Mercedes-Benz, and the like are showing better results in their test drives. While driving, Uber’s test driver has to react three times per mile.
Uber’s self-driving vehicle killed a pedestrian in Arizona.
This drastically slowed its autonomous car research and erased Uber from most self-driving cars stats for a while. In fact, they fired all their drivers and called all the cars off the roads. The final verdict states that the company won’t be charged for the death of Elaine Herzberg, 49. However, the Uber driver who was watching The Voice on their phone when the car hit the woman may be facing charges.
Uber’s self-driving car software may recognize people as an unimportant obstacle.
In short, what happened in the Arizona incident may be followed by even more self-driving car accidents. The car’s software algorithms interpreted the woman as a false positive—an object that can be ignored, much like a wind-blown plastic bag.
After a long break following the Arizona incident, Uber’s self-driving cars are again on America’s streets.
Uber will start off testing its self-driving cars in Pittsburgh, particularly the Strip District, where a few other companies are testing their vehicles as well.
This year Uber received $1 billion to continue its research and improve its car safety.
The company got its new funding mostly from Toyota and Saudi Arabia.
Self-Driving Car Safety Statistics
Further autonomous vehicle statistics indicate that 56% of Americans would not feel safe in a driverless vehicle.
(Pew Research Center)
According to Pew’s survey, until the rates on crashes and accidents improve, Americans won’t be so keen on taking a relaxing ride in an AV.
55% of small business owners believe their fleets will be fully autonomous in 20 years.
(Nissan Motor Corporation)
While this statistic may seem incompatible with the previous one, it just confirms that people have more confidence in AI when only goods are being transported safely. Moreover, 66% share the opinion that vehicles will be fully electric by then.
The more complicated the car route, the better the self-driving system, as shown by statistics on self-driving cars.
According to General Motors, their Cruise learns much faster in the complicated, knotty streets of San Francisco than in simpler streets. For example, construction work is encountered 39 times more often in SF than in most other cities.
Self-driving cars will cause more traffic jams.
If more people are using self-driving cars, they can send them to cruise the city until they ask the cars to return to their pick-up place. In this way, there will be more vehicles in traffic.
Most self-driving car statistics indicate that pedestrian behavior is one of the software’s greatest challenges.
Meanwhile, human drivers can predict with some certainty whether a pedestrian on the sidewalk will cross the street or not (especially outside a pedestrian crossing). Self-driving systems and software still have a long way to go.
Self-driving cars have 5 levels of autonomy, which vary greatly in regard to how much a driver can sit back during the journey. And it won’t be too long before we reach the final level. However, to get there, self-driving car manufacturers need to invest more resources into enhancing their software and scanning technology in order to ensure overall traffic safety. After all, the self-driving car statistics seem to show that drivers still aren’t very confident about driving an autonomous car, and perhaps for good reason.