The recent incident involving Uber and a pedestrian has the media buzzing. Just about everyone, and anyone has something to say about it. The over-analysis has been going on for days with both the “if it bleeds, it leads,” media, as well as the more respected technical media.
Truth be told, I am surprised this is getting so much attention. It is not as if this is the first incident of this type. In fact, there have been a number of accidents with cars in autonomous mode, from minor bumps to deaths. One of the more notable was in 2017 when a Tesla killed a driver. Another incident, again involving a Tesla, happened when it ran into a stationary fire engine at 65 mph. Earlier in 2017, an autonomous vehicle owned by Cruise (acquired by GM last year), struck a motorcyclist on a San Francisco street. The list goes on and on.
Statistics on autonomous vehicle accidents are a bit rare (perhaps because of the industry’s efforts to keep them hushed?) However, one I did find, although a bit dated, is a report from late 2015 (https://www.cnbc.com/2015/10/29/crash-data-for-self-driving-cars-may-not-tell-whole-story.html) that notes self-driving cars are involved in more accidents that their driver counterparts.
That being said, there are a number of metrics at work here that need to be addressed in the autonomous vehicle segment. The first, and IMHO, most visible, is the hype. This is the same issue that 5G faces. In the madness to monetize emerging technologies, proper cautions are tossed aside and rose-colored glasses are donned.
Why that is, has to do with what is at stake, and it is astronomical. The price of entry is very high, but so are the rewards. Therefore, the tendency is to push the envelope. When that push goes wrong, there is an inclination to either downplay, justify or excuse it.
For example, early reports on the Uber incident by the media say that Uber is not to blame because the woman was not in a crosswalk DUH! They are so missing the point. It should not matter where the woman was. Avoiding such accidents is precisely what autonomous vehicles should be good at – she was a slow-moving target and there was plenty of recognition time. This was an epic fail on the part of the vehicle because it did not recognize the situation. Crosswalk or no, the technology should have been able to recognize that it was a pedestrian or bicycle, wherever it was located on the street.
Tell me that the database and algorithms are not that stupid, and contain enough data and intelligence to realize that there is an object in front it, and a collision is imminent. This is, probably, one of the most common scenarios – pedestrians, or other objects (animals, debris, various stopped vehicles, etc.) on a street. At a minimum, this should be a classic case, easily programmed. Even if such is not the situation, the vehicle should stop, or take avoidance measures, whenever it sees a stopped, or slow moving object in front of it. That is not rocket science!
Next issue is technology. In my rhetoric on autonomous vehicles, I have said, more than once, sensor-only technology will NEVER be reliable enough to enable fully autonomous cars. There is a lot of noise being produced that vehicles will reach level five – totally driverless and intervention-free, by 2021. I think that is sheer hyperbole.
The third issue is perception. Unlike 5G, autonomous vehicles are the next generation of a platform that everyone knows about, and interacts with. The deployment of 5G will go largely unnoticed by the vast majority of the populous. Autonomous vehicles, however, will be exactly the opposite. There is better than 100 years of automotive history that, except for the technology, has remained largely unchanged; you get in, you accelerate, you steer, and you brake. Self-parking autonomous vehicles are one thing. Vehicles sans a steering wheel, go, and stop pedals are something else, entirely.
Therefore, flipping the mental switch that says it is ok to let your car drive itself has to overcome not only tradition, but also history, familiarity and trust.
The flap over the Uber incident is ridiculous. Fallout is from “it is the end of driverless cars” to “it is part of the learning curve.” The real ramifications are somewhere in the middle. Autonomous vehicle technology will continue to evolve. There will be mishaps along the way. It is no different with this technology from any other. However, pushing the edge of this envelope, for the sake of hitting forecasted goals, in this case, becomes an issue of life-safety.
The key is to be realistic. Stop over-hyping, hand wringing, over-expecting, over-promising and all the rest of the “overs.” All this does is cause a hard stop, such as Uber halting all autonomous car testing, when a problem or situation arises, and makes everyone nervous. Autonomous cars will continue to evolve as the technology evolves. However long it takes to reach the different levels should be presented, pragmatically and realistically. It will go much better that way.
That is, however, not our competitive mentality. In this case, perhaps it should be.
Executive Editor/Applied Wireless Technology
His 20-plus years of editorial experience includes being the Editorial Director of Wireless Design and Development and Fiber Optic Technology, the Editor of RF Design, the Technical Editor of Communications Magazine, Cellular Business, Global Communications and a Contributing Technical Editor to Mobile Radio Technology, Satellite Communications, as well as computer-related periodicals such as Windows NT. His technical writing practice client list includes RF Industries, GLOBALFOUNDRIES, Agilent Technologies, Advanced Linear Devices, Ceitec, SA, and others. Before becoming exclusive to publishing, he was a computer consultant and regularly taught courses and seminars in applications software, hardware technology, operating systems, and electronics. Ernest’s client list has included Lucent Technologies, Jones Intercable, Qwest, City and County of Denver, TCI, Sandia National Labs, Goldman Sachs, and other businesses. His credentials include a BS, Electronic Engineering Technology; A.A.S, Electronic Digital Technology. He has held a Colorado Post-Secondary/Adult teaching credential, member of IBM’s Software Developers Assistance Program and Independent Vendor League, a Microsoft Solutions Provider Partner, and a life member of the IEEE. He has been certified as an IBM Certified OS2 consultant and trainer; WordPerfect Corporation Developer/Consultant and Lotus Development Corporation Developer/Consultant. He was also a first-class FCC technician in the early days of radio.