Many decisions in everyday driving are tradeoffs. One wouldn't swerve off the road for a pigeon, but in the case of a similar sized rock would be compelled to swerve or simply duck. Too many people are killed every year swerving to avoid small animals like birds, dogs, cats, racoons or skunks. The decision is more difficult as animals get larger - wolves, deer, moose or bears. And of course the biggest problem of all for auto-piloted cars is pedestrians of all sizes, especially small children.
I once witnessed a number of cars on a limited access highway spinning out on the road and off the guardrails immediately ahead of me. Without consciously thinking, in my mind I quickly calculated their paths, down shifted and threaded my way unscathed. Vehicles following me panic braked and joined the pileup. While a computer could in theory be programmed to do what I did, it is highly unlikely. Of course if they were all on autopilot, with ABS and stability control, this type of occurrence would be unlikely.
The problem with a public road system is the large variety of vehicles, unlike a limited access rail system. The other problem is the time-frame for implementation. Inevitably, there would be a time of both auto-piloted and human driven vehicles on the road. At some point, one would only allow auto-piloted vehicles on the faster roadways. Can you imagine the political infighting for such decisions in light of what just went on? Is a dictatorship on the horizon?
The U.S. air traffic control automation program is a likely scenario. Billions have been spent on this comparatively simple system. Inevitably, as the prototypes were becoming promising, the technology became obsolete, unavailable and unserviceable, so a new development contract was let. This cycle repeated many times over many decades.
So there's a great deal of politics in implementation at all levels, from designers to legislators.
I would be thrilled to be able to do office work in my auto-piloted car, just a many farmers do today in their auto-piloted tractors. The question is will I live long enough to see the day? Hopefully the auto-piloted car will have a better future than the flying cars promised decades ago.
I would certainly agree that the driverless car will have to depend on more than just radar. The obstacles you mention would mostly be undetectable by radar. That's why I think that truly autonomous mode will require optical sensing, assistance from V2I, and of course algortihms to make sense of the information.
How does one develop autmatic control algorithms? A good starting point is, you ask yourself how you would do this manually, and then you have the algorithm emulate that behavior.
So for example, the shredded truck tire. Optical sensors will notice that. The algorithm then needs to determine whether it can swerve left or right, ar whether it has to slow down drastically (which would be the last choice).
If all cars on this segment of road are being driven autonomously, a suuden swerve will be a whole lot safer than it would be with emotional, mostly untrained humans in the mix.
A well executed V2I system would work even better, noticing the obstacle and signaling to cars upstream to get in another lane.
I grant you that on highways with very controlled access, autopilot in cars might perform nearly as well as an autopilot airport train. But automatic trains run few and far between and are not of different manufacturer and performance. Even in a very controlled highway, how would an autopiloted car react to transport trucks' shredded tires, stray moose or even sheets of ice lifting off the roofs of truck trailers and other debris let alone stray pedestrians?
The permutaions on the average road are orders higher than on a protected rail system.
To counter that point, there is a way much safer emrgency mode for the driverless car: stop. Airplanes don't have that luxury.
I'm sure that before driverless trains were installed in airports all over the world, there was similar skepticism. And yet, here we are. I'm not sure why engineers would be so skeptical, since one would think that engineers have seen the march of controls automation moving in this direction for decades. Even in cars, for heaven's sake, where the only manual control left to the driver is SOME steering, SOME braking, as sometimes the throttle. (I'm referring to ABS, yaw control, and cruise control as steering, braking, and throttle that has already automated at least parts of these critical functions, always in ways that certainly the average driver can't hope to equal.)
Would an engineer really believe that there's some law of physics or nature or whatever, that absolutely forbids automatic control to get beyond this point? Seems odd, right?
There's a big difference between an autopilot in a plane and in a car. In the air, when autopilot is used, there are literally miles between aricraft. And despite all the training, autopilot still guides planes straight into mountainsides.
In aircraft situations, distance and timing become more critical around airports and landings - comparable to roadways. In such time distance critical situations, your average pilot would normally have his autopilot off and is unlikely to be distracted reading or texting. Even with landing automation, the pilot will no doubt have undivided attention.
So it appears nothing can save us from the perils of distracted driving, except less rather than more distractions available in cars.
To say that it is a driverless car and still expect the human driver to be present , and not just be present ,but be alert enough to take over the control of the car in case the system is not able to handle a situation makes the whole concept meaningless.
It is better to have a driver assistance system where the control always lies with the human driver.
"So, that says that true autonomous driving should only occur on roads that have been set up for such contingencies, e.g. roads which have some shoulder or emergency lane available. (The alternative is that traffic will more often come to a complete standstill, if drivers aren't ready to take over quickly.)"
I totally agree with the comment (for automated islands) but would add that you should not mix automated and manual driven vehicles. Imagine an automated car slowing down because of say sensor oe driver failure on a six lane highay with all the traffic doing 60mph. Not pretty.
If you mix manual and automated vehicles, then how do you deal with the manual driver not letting the automated one in (for some unexplained emergency). Is there such a thing as being rude to an automated car?
"It would be interesting to understand what the Google car algorithms would do if it prompted for the driver to take control and he/she did not?"
Don't know about Google, but if I wrote the algorithm, I'd always default to "slow down, move to emergency lane, stop." Whatever situation arises for which the automatic controls can't handle it and the human driver is unavailable. As this is happening, either V2V and/or just the proximity sensors in cars behind it, will slow down traffic.
So, that says that true autonomous driving should only occur on roads that have been set up for such contingencies, e.g. roads which have some shoulder or emergency lane available. (The alternative is that traffic will more often come to a complete standstill, if drivers aren't ready to take over quickly.)
Sensors will also need redundancy. For example, it would be nice for the car to have more than one way of determining where the emergency lane is. If not with an onboard sensor, then with a V2I signal.
I don't really think any of this is terribly futuristic. We have airplanes flying and landing on autopilot, we have trains with no driver, why not cars? For planes and trains, the path they can travel is well determined. So we have to do the same with autonomous cars.
>> "Drivers being asked to take more driver education courses and stay alert all the time behind the wheel inside a self-driving car, in a way, defeats the whole purpose of autonomous cars."
Sure, it is autonomous but you cannot have a nap in the car. You cannot even be distracted. Yet, it is an autonomous self-driving car. The main challenge is that the economics of replacing humans in driving cars is not that huge. You are going to part with four guys. Suddenly, no one needs to drive as the car can drive itself. What have you truly saved economically?
>> "Can you draw a parallel between a pilot flying an airplane on autopilot and a driver in a self-driving car?"
Considering that you can have a maximum of say 8 passengers, I do not see the reason why we need people in these self-driving cars. If the thesis is to drive someone to the mall, when people are seated, why not make one of the people have a driving license. But if the idea is to send the car to pick things in the mall without people in it, then it makes a lot of sense. While you can be in a car and be reading expecting it to drive itself, it does not offer a lot of productivity when compared to airplanes.