Editor’s Note: In recent days there have been two events in the world of self-driving cars that should make everyone take notice. The first was the announcement by Ford Motors that it is planning to have a fleet of autonomous vehicles on the road by 2021. These will be revolutionary vehicles. No steering wheel, no brake pedal, no gas pedal, and no individual ownership. Ford envisions on demand car service that will not require human intervention.
The second, and more proximate development is the announcement that Über is starting a self-driving car experiment in Pittsburgh with specially equipped Volvo XC90’s that will pick up passengers, but will have both a driver who can intervene if needed and an observer to gauge the passenger experience. The Principal Analyst for Molilocity, LLC., Gerry Purdy, PhD., has been one of the nation’s leading experts in mobile technology for more than a generation. He’s now turning his attention to self-driving cars and wrote this piece which he’s graciously agreed to share with Tech50+ readers:
Earlier this year (January 13), I wrote about the coming full autonomous vehicle that could be summoned as needed and drive us without a steering wheel as, perhaps, a decade away, possibly further. Now, this past week, Uber announced it is planning to do a trial of self-driving taxis in downtown Pittsburgh starting in the next few weeks. The test will use Volvo’s XC90, which has been configured to drive autonomously. 
Ford also plans to join the test using modified Ford Fusions (as shown in the lead photo to the left). Has my rather conservative prediction been terribly off or is something else going on?
What struck me about this announcement compared with all the other announcements recently, however, is that in the Uber test in Pittsburg, there are real passengers being taken to their desired location for the very first time. Right now, it may seem inconsequential, but I can assure you that this is a watershed moment for telematics.
It turns out, there are a number of different stages of self-driving cars and trucks, and this trial by Uber is really happening but with a lot of oversight. But, it’s still a real watershed moment that should get you to stop and realize that all this talk about autonomous vehicles is real and is happening much fast than anyone expected, including yours truly. Let’s put this all into perspective.
You read almost every day about an announcement in telematics – the integration of technology and vehicles (both cars and trucks). Every major car and truck company is working on ways to bring technology to support a safer driving experience into the market. But, just adding Super Cruise Control (like what’s available as Autopilot in the Tesla cars) doesn’t mean you can give control over to the car, have it safely take you where you need to go and let you ‘tune out’ from driving to do other things such as reading, doing work on your laptop, or taking a nap.
The reason that Tesla owner Joshua Brown died on May 7 while driving in Florida was because he did just that – he ‘tuned out’ from the responsibility of driving, and the Tesla look ahead software got ‘blinded’ from the reflection off the truck in front of the car and didn’t automatically slow down. Sure, those kinds of things will likely not happen again. Software that looks for a blinding situation will be updated along with the use of better sensors that will work correctly in such situations.
The Society of Automotive Engineers (SAE) has published a great graphic on the different levels of autonomous vehicle operation. It is similar to the graphs created by the National Highway Traffic & Safety Administration (NHTSA) but extended from five to six levels to better define the operation of these self-driving vehicles:
Note the two bracketed areas at the top of the diagram: they delineate the difference between driving that is monitored by the driver and driving that is not monitored. Going from one step to the next is not accomplished in linear fashion. Rather, making that transition takes a lot of work, testing and passing government regulations to ensure that the truly autonomous vehicles will work all the time, not just some of the time.
In particular, as you move from Level 3 to Level 4, you’ll notice that the icon for the drivers eyes being off (as in asleep) shows up. That’s the demarcation that is going to take the most government oversight and regulatory approval.
Thus, going from Level 3 (where Tesla is today) to Level 4 is going to take a lot of time and governmental oversight. And, further, Level 3 does not mean you can take your eyes off the road or the steering wheel. Watch this video of a driver using Level 3 driving assist called Autopilot in a Tesla and see what happens when a service vehicle suddenly moves from the left to pass right in front of the car. The system did adjust and prevent an accident, but you can also hear that the driver is very aware that it was a dangerous situation and was able to take back control if necessary.
The current forecast for our being able to have cars and trucks that drive autonomously is at least five years off, with the more likely date to see this stuff (as I said before) around 2025. It might happen earlier but it just as likely might happen later.
Now, re-read the announcement by Uber about their testing self-driving vehicles in downtown Pittsburgh. They are not cars that come by, let you in, and take you to the address you entered on your smartphone. Rather, this is a test environment in which each car will have two people in the vehicle besides the passenger: one to sit behind the wheel and take over when the situation is unsafe and the other to observe and make notes to help give engineers feedback so they can improve the software in a rapid manner.
There were a number of other announcements in the past 10 days about the future of autonomous driving. Perhaps they are all just coincidental, but it seems that the auto manufacturers are each predicting that they will have some form of autonomous vehicle in the market by 2021 which most likely will slip at least a year or two. Also, notice that these companies are not saying that you’ll be able to buy a fully autonomous car by then. But, rather, they will sell cars that can only go in certain areas in which have been marked such as in a shuttle or small form of a bus going on a certain, predefined route.
Uber also announced it was buying Otto, an autonomous vehicle software development company focused on automatic driving of trucks rather than cars. And Uber also recently announced a partnership with Volvo to invest $300 million in autonomous vehicle development.
Ford said on August 16 that it will also roll out a fleet of driverless cars within five years. These are the vehicles that will be deployed in designated areas marked to define the route of these autonomous cars. The vehicles will operate in only a geo-fenced area that has previously been laid out and mapped. And, Ford also announced an investment in Velodyne, a company that makes LIDAR sensors for autonomous self-driving cars. LIDAR uses laser light beams to sense things nearby much like radar works.
In summary, while these developments in self-driving cars are impressive, it is still many years before you should take your eyes off the road and hands off the steering wheel. The major auto manufacturers along with firms like Uber are bringing the world of self-driving cars into reality in our lifetime.
If you left the Earth and returned in, say, 25 or 50 years, you’d likely immediately notice that no one is driving cars or trucks. They all will be driving themselves and controlled by on-board systems that are safe and allow you to do other things while going from point A to point B. And the accidents and fatalities from such operation will be a very unusual event.
In the meantime, if you use Autopilot in a Tesla or other similar Super Cruise Control, don’t take your eyes off the road or your hands off the steering wheel. You’ll thank me later.
About the author
J. Gerry Purdy, Ph.D. is the Principal Analyst with Mobilocity LLC. A nationally recognized industry authority, Dr. Purdy focuses on monitoring and analyzing emerging trends, technologies, and market behavior in mobile computing and wireless data communications devices, software, and services. Dr. Purdy is an ‘edge of network’ analyst looking at devices, applications, and services as well as wireless connectivity to those devices.
Dr. Purdy provides critical insights regarding mobile and wireless devices, wireless data communications, and connection to the infrastructure that powers the data in the wireless handhelds.
The post A Watershed Event For Self-Driving Cars appeared first on Tech50+.