If that many people were killed and injured in a war anywhere in the world, it would definitely make headline news. As it is, the carnage is swept up, the bodies buried, and everyone tries not to notice. We get in our cars, try not to think about the risk, and continue our driving, sometimes well, sometimes inattentively, and always assuming that we will arrive at our destination in comfort, and safety.
The one time you can be sure that a fatal accident reaches the news headlines is one involving an autonomous vehicle. I say autonomous, but so far, only semi-autonomous vehicles have been involved in any fatal incidents. In the fatal incidents involving semi-autonomous electric vehicles, the common element has been a human operative relying on the semi-autonomous system as if it were fully autonomous.
The 6 Levels of Autonomy
While writers and readers at CleanTechnica are probably fully aware of the different levels of autonomy in electric vehicles, the general public is less so, and probably doesn’t make the distinctions between fully autonomous (level 5), almost fully autonomous (level 4), and only semi-autonomous (levels 1 to 3). (See the table at the end). They might very well already be using the level of autonomy provided by level 0, where automatic systems aid them in their day-to-day driving, such as anti-lock braking systems, traction control, reversing cameras, obstacle warnings, and cruise control, but have no experience of an automated system completely taking over the function of driving to any extent.
They are probably also unaware that in the fatal crash involving an Uber vehicle, Uber might have partially caused it by cutting corners to save money on its autonomous-system’s sensors. For the public, all autonomous vehicles were created equal, and so a failure of one is an indictment of them all.
Built in Failure
It seems to me that levels 1 to 3 of autonomy have failure built into them. The autonomous system relies on the human driver to complete its functions, and the driver is relying on the autonomous system to complete the driver’s functions. In other words, neither the driver nor the autonomous system has full responsibility for control of the vehicle. Anyone with any experience of life will know that when nobody has full responsibility for something, then sooner or later, something is going to go badly wrong. It seems inevitable that any semi-autonomous system will be inherently unsafe, and that the only safe system will be either complete autonomous control by the system, or complete control by the driver with only minimal assistance from an automatic system.
Safety in Autonomy
One thing that also seems to be true is that a fully autonomous system is likely to be a lot safer than a human driver in almost all cases. Artificial intelligence does not fall asleep, never gets bored, inattentive, or distracted by something more interesting than the road ahead. Artificial intelligence is not panicked and taken by surprise by events, but is calm and attentive at all times, and reacts instantaneously and logically in every situation.
Artificial intelligence is not hampered by failing eyesight or hearing, and can even see in the dark, using radar and infrared sensors. It might be the case that a fully autonomous vehicle, unlike a human driver, would no longer require headlights or fog lights, but just running lights to make it more visible in the dark. Artificial intelligence does not become impaired by drinking, or drugs, or the desire to seek thrills and excitement by aggressive driving. Autonomous vehicles do not experience road rage. All this being so, the speed limits could be higher for autonomous vehicles driving closer together, and so allowing better traffic flow and higher density.
The Hive Mind
In addition to the systems that current autonomous vehicles rely on, they could also communicate with each other on the road, so that every vehicle is aware of every other vehicle in its immediate vicinity, knowing their speed, position, and intended next maneuver or heading, making a traffic collision even less likely. Where human drivers are provided with traffic lights and traffic signs to control driving behavior on the road, an autonomous vehicle could receive information from roadside transmitters, and so all the traffic lights and traffic signs could be done away with once all vehicles are autonomous.
It might even be the case that nothing equivalent to traffic lights would be needed. At a junction, all the autonomous vehicles approaching that junction could agree on speed and road position with each other to allow all the vehicles to go over the crossing without stopping, and just pass through the gaps between vehicles. Humans could not calculate that, but computers could. All the autonomous cars in the city could adopt a hive mind, and flow through the roads, driving very close together, with ease, and rapidity, like a murmuration of starlings in the sky, a nest of ants, or a shoal of fish.
Where human drivers sometimes ignore traffic lights or break the speed limit, autonomous vehicles would obey the rules perfectly without exception. All of these possibilities could lead to an end to the huge number of road deaths once all vehicles are autonomous. Traffic collisions would be reduced to system failures, or where pedestrians and cyclists, have been at fault. In the future, cities and roads could be better designed to keep vehicles, cycles, and pedestrians separated. We would not need traffic policemen, and insurance would be very cheap.
We assume that autonomous vehicles will bring many benefits, but it is not easy to predict. The obvious benefit is that it can combine the benefits of private and public transport in one.
City Bus Blues
Even though I know the theoretical benefits of public transport, such as reducing CO2 and pollution, and easing congestion on roads, I rarely use it. At my previous house I had a bus stop almost at the bottom of my front garden, just around the corner from my hedge, and if I wanted to use the bus, I only needed to step out of the front door, go down my garden path to the front gate, and from there it was literally 20 yards walk, passing my own front garden, to the bus stop. All very convenient, right? But it was all downhill from there.
The service was somewhat sporadic, so sometimes I would get a bus almost straight away, and sometimes wait and wait, which is something I hate. Then when I got on the bus, it wasn’t that comfortable; bumps in the road would have me bouncing up and down on the relatively hard seat.
The bus was a big double-decker diesel affair, and was quite noisy. It went along briskly enough, but seemed forever to be stopping at bus stops, or in traffic, and then there was the guided tour of suburbia. From my house, there was a wide and straight main road about a mile away, which went directly into the city, but the bus wanted more fares than just mine, so we would wind our way through suburban roads, picking people up, before finally making it to the city, jam-packed with passengers. Believe me, the bus was the last resort.
Normally I would cycle in, which took me about 15 minutes straight to my office with no walking, where the bus would take at least 30 minutes, cost me a bus fare, and leave me with a mile to walk to the office. The bike route even took me along a traffic-free path along the local river, which was a nice start to the morning, watching little plumes of mist held in the golden sunbeams, while majestic swans paddled unhurriedly by.
How different it would be if I was able to make an entry in a smartphone app, say when and where I want to go, and a few minutes later, get a little “ding” on my phone to say my robo-taxi is waiting. I get in, carry on reading what I was reading, and the car zooms off, taking the quickest route to my office. After 10 minutes, I get out at the end of my journey, with no worries about parking, as the car goes off to its next assignment, and my account is automatically debited with the affordable fare. Now that might have been more tempting than cycling, especially in the rain.
The Key to Congestion?
Some people believe that such systems would make car ownership a thing of the past, and so greatly reduce congestion and the need for car-parks at supermarkets and places of work. I am inclined to agree to some extent. For a lot of people, that would be cheaper than car ownership, and without the responsibilities such as maintenance, parking, annual tests, vehicle tax, insurance, and charging-up.
For some people, their car is a symbol of their status: the make and model is a statement to the world of how wealthy, and important they are. It gives a boost to their flagging ego to be seen driving by in their brand new, shiny, autonomous vehicle. They are not about to forego having a car of their own.
When they arrive at their destination they can tell their autonomous car to go and park itself, or find a charging station while they do their shopping or whatever business they want to attend to. Some people think this might make the roads even more congested, as autonomous vehicles drive around empty, heading for a parking space, or just going round and round in circles waiting for their owners to need them again. We would go from vehicles with only one occupant taking up space on the road to vehicles with no occupant at all. It is hard to predict, and I am not about to try. Only time will tell.
A further possibility is that electric robo-taxis will be so cheap that everyone will want to use them in preference to buses and cycling. That would create even more congestion, as one bus can carry about 40 people, whereas, if each person was in an autonomous vehicle, that is potentially 40 vehicles instead of the one bus. Cycling is also more healthy for people, (sans traffic-fumes), so people swapping the bike for the robo-taxi would not be such a good thing, either.
This increase in congestion does seem likely, as even before automation, Uber ride-hailing services have created the situation where many of the vehicles driving about the city causing congestion are Uber drivers, looking for passengers.
One possibility I have heard about recently is that when autonomous vehicles go to the car park, they will take up a lot less space. Have you ever parked in the supermarket car park, and on your return, can’t get back in the car because someone has parked so close that you can’t open the door, or you have been blocked in by someone parking where they shouldn’t? I did see one unfortunate fellow who had to resort to climbing through the back, to clamber over the seats to get himself behind the wheel. Now, if all vehicles were autonomous and parking with nobody inside, then nobody needs to open the doors, so they can park a lot closer together, thereby getting more cars in each row.
In a normal car park there might be a row of cars with an access road behind them, and then a row of cars in front of them, nose to nose, with an access road behind them too. With autonomous cars, provided they are able to communicate with each other, there could be another row of cars directly behind the first row, and another row of cars behind the second row. Because these are autonomous cars, all connected, and cooperating with each other, cars at the back of the row will move out of the way to allow cars at the front to get out. They could even rearrange themselves afterwards to make the empty spaces more accessible.
There are many other advantages, such as enabling people who might not otherwise be able to drive to either own or use an autonomous vehicle: perhaps those too young, those with medical conditions, and those with age-related problems, or who are otherwise disabled to drive. It would also make long journeys much more restful or productive, as instead of driving, one could work, read, or watch a movie, take refreshments, sleep, or a combination of these.
Depending on the range of batteries, by the time autonomous vehicles are the norm, it might also be possible to complete long journeys without stopping at all, bar “comfort breaks.” This would make journeys by autonomous vehicles quicker, as well as less stressful.
There being no driver to pay, one would assume that the robo-taxi would be relatively cheap to make use of, and enabling people who would not be able to afford their own car or to make much use of conventional taxis to enjoy the convenience of door-to-door transport in the comfort of an autonomous vehicle.
Of course there are downsides too. Automation and artificial intelligence are likely to create a situation in the near future where at least 50% of the population will be unemployed, and indeed, unemployable. Automation simply cannot work in the current free-market system. Yes, we will have robots capable of producing everything and anything anyone could think of, but there’ll be few people able to buy all of these products, if 50% of the population is unemployed.
Under the current system, unemployed people do not have the financial resources to participate in society, so we can look forward to a dystopian future when 50% of the population neither contributes to, nor benefits from, this wonderful automated high-tech society. This will start with all the people currently employed in driving being replaced with autonomous vehicles, and in warehouses where selecting orders will be fully automated. Along with the redundancy of drivers would come the demise of traffic cops, motoring insurance clerks, and all those driving-related jobs.
Even without autonomy, EVs will spell the end for service-and-repair shops in general, and businesses making, selling, and fitting all the soon-to-be-redundant spare parts, like oil filters, spark plugs, fan belts, exhausts, batteries, brakes, and clutches. With the spread of automation and AI, the job losses would start in relatively unskilled jobs, but soon extend to the jobs of many highly knowledgeable, skilled decision-makers, so consequently, no-one would be immune.
The only solution to this is to introduce a national basic income at the earliest possible time to replace the current system of welfare benefits and pensions, and for this to be sufficient to allow everyone to fully participate in our society. People will be able to find much more fulfilling ways of using their time than working in full-time jobs. Since I retired, I am having a great life, and don’t know how I ever found the time for a full-time job.