Author Topic: Robo-Cab Predictions  (Read 8141 times)

0 Members and 1 Guest are viewing this topic.

Offline EEVblog

  • Administrator
  • *****
  • Posts: 31551
  • Country: au
    • EEVblog
Robo-Cab Predictions
« on: May 15, 2019, 06:49:19 am »
Just posted this on twitter, reposting here.
My predictions for Robo-Cabs / Robo-Taxi

PREDICTION #1:
Robocabs that can match a human driver in practicality are *many* years away. Perhaps even a decade or more.
Anyone talking next year is kidding themselves.
And they could still suck in many situations.

PREDICTION #2:
People taking Robo-cabs won't like or trust traditional looking cars with an empty front seat and steering wheel whizzing around.
Expect something different to win out.

PREDICTION #3:
Johnny Cab won't happen. People will detest humanoid robots trying to interact with them.

PREDICTION #4:
The "one bad experience" problem may seriously hamper initial Robo-Cab adoption.
And with social media how it is, all you'll hear about are the bad experiences.
The media & your Facebook friends will lap it up.
Although, Facebook could be dead by then...
 

Offline apis

  • Super Contributor
  • ***
  • Posts: 1667
  • Country: se
  • Hobbyist
Re: Robo-Cab Predictions
« Reply #1 on: May 15, 2019, 03:09:49 pm »
PREDICTION #1:
Robocabs that can match a human driver in practicality are *many* years away. Perhaps even a decade or more.
They don't have to. They have to be safer than human drivers, but in other respects it will be a question of price vs convenience. If a robocab is more convenient than the bus and cheaper than a normal cab (and possibly cheaper than the bus!) people will use them.

Anyone talking next year is kidding themselves.
Yes. Although last year Waymo said they would launch Waymo One without safety drivers in Phoenix this year. So far they have launched Waymo One but they are still using safety drivers. It's hard to tell how close they are to the goal because it's such a competitive market now that everything is kept secret.

But Phoenix is only their first pilot project. It will take a few more years before they launch in other cities, and then a few more years before they start to expand the operating area. It will take decades before they will be operating outside the US I suspect.


And they could still suck in many situations.
There will be many instances when the robocabs will seem stupid (because they are, it's not a general AI), especially in the beginning. But as long as it isn't safety related and doesn't cause too much inconvenience I don't think it will be a show stopper. People will use them as long as it is safe, cost effective and "good enough".

PREDICTION #2:
People taking Robo-cabs won't like or trust traditional looking cars with an empty front seat and steering wheel whizzing around.
Expect something different to win out.
I think something different will win out but mostly for other reasons. You want to use all seats for passengers, and the operating companies probably doesn't want the passengers to try and interfere with the driving controls. So the steering wheel, etc, will be removed (or at least hidden).

There will probably also be more specialised vehicles. You use a phone app to order a vehicle, so I assume you will fill in how many you are, where you are going and if you have any luggage or special needs, and then they send an appropriate car to handle your specific request.

If a lot of trips are short with only one passenger for example, then there will probably be small 1-2 seated cars that can handle such requests. One can also imagine there will be mini busses that pick up people during rush hours for example. Time will tell.

PREDICTION #3:
Johnny Cab won't happen. People will detest humanoid robots trying to interact with them.
Agree. A stupid animatronic puppet that annoys you and takes up precious space. Remember Clippy? People doesn't wan't a "human" interface in between them and the machine.

PREDICTION #4:
The "one bad experience" problem may seriously hamper initial Robo-Cab adoption.
And with social media how it is, all you'll hear about are the bad experiences.
The media & your Facebook friends will lap it up.
Although, Facebook could be dead by then...
Yes, if they rush it and launch big before the service is ready it will probably set them back many years because of the publicity backlash. But they have already invested so much in this, they won't abandon this idea easily.
« Last Edit: May 15, 2019, 03:11:52 pm by apis »
 

Online Bud

  • Super Contributor
  • ***
  • Posts: 4213
  • Country: ca
Re: Robo-Cab Predictions
« Reply #2 on: May 15, 2019, 04:23:28 pm »
Quote
Anyone talking next year is kidding themselves.
Weed smoking Musk has promised exactly that, putting them on the streets next year. Yeah, another visionary thing. With cannabis the term "visionary" takes a new meaning.
Facebook-free life and Rigol-free shack.
 

Online Bud

  • Super Contributor
  • ***
  • Posts: 4213
  • Country: ca
Re: Robo-Cab Predictions
« Reply #3 on: May 15, 2019, 04:26:22 pm »
People will use them for sex, drinking, and leave all sort of garbage behind.
Facebook-free life and Rigol-free shack.
 

Online 0xdeadbeef

  • Super Contributor
  • ***
  • Posts: 1493
  • Country: de
Re: Robo-Cab Predictions
« Reply #4 on: May 15, 2019, 04:36:05 pm »
I would argue that real autonomous driving (in a real world traffic situation you could encounter today in a car) could only work reliably with real artificial intelligence (and not some brainless neuronal network giving somewhat nondeterministic results) so it could react to unexpected situations at least as good as a human driver.
So even ten years from now sounds pretty much optimistic unless there is some unexpected breakthrough in AI. While I would really look forward to have autonomous driving in twenty years or so when I'm retired and half-blind, I would not entrust a car with my life that will decapitate me when mixing up a traffic sign with a truck.
It's somewhat shocking to see how easily today's image recognition systems can be tricked into totally overlooking traffic signs, persons and what not.
Trying is the first step towards failure - Homer J. Simpson
 

Offline Red Squirrel

  • Super Contributor
  • ***
  • Posts: 2480
  • Country: ca
Re: Robo-Cab Predictions
« Reply #5 on: May 15, 2019, 04:50:19 pm »
I still want to see how well a self driving car would do on a typical road in the winter, especially side roads.  No lines (covered in snow) to act as guides and no real distinguishable road edges, other than snow banks.  Any kind of radar/lidar will have a hard time with that texture I think. 

In a southern city where the roads are super smooth, and have nice crisp lines it should not be too hard, but it's not like that everywhere.

I think driving assist technologies will continue to improve, where the driver can do a bit less of the work, but driverless is a long ways away I think.  Then again it's easy to underestimate just how advanced technology can get sometimes.
 

Offline apis

  • Super Contributor
  • ***
  • Posts: 1667
  • Country: se
  • Hobbyist
Re: Robo-Cab Predictions
« Reply #6 on: May 15, 2019, 05:37:46 pm »
People will use them for sex, drinking, and leave all sort of garbage behind.
Well, they will be charged for the cleanup if they do. Expect inwards facing cameras.

It's somewhat shocking to see how easily today's image recognition systems can be tricked into totally overlooking traffic signs, persons and what not.
Only Tesla are relying that heavily on neural networks and they haven't demonstrated the same level of self driving as the others. The Tesla autopilot is basically advanced cruise control. Everyone else is using very good sensors and deterministic algorithms for the basic steering and collision avoidance. They extend here and there with neural networks when it makes sense and is safe. It's ironic that people first think of Tesla when self driving cars are mentioned, based on what they have shown so far they are way behind Waymo and the others.

I still want to see how well a self driving car would do on a typical road in the winter, especially side roads.  No lines (covered in snow) to act as guides and no real distinguishable road edges, other than snow banks.  Any kind of radar/lidar will have a hard time with that texture I think. 

In a southern city where the roads are super smooth, and have nice crisp lines it should not be too hard, but it's not like that everywhere.
I don't think it matters if the roads are super smooth or not, it all began when the Stanford team won the 2005 DARPA grand challenge and that was a 212 km (132 mi) course through desert. What they have problems with today are things like figuring out the best place to stop and pick up passengers at a crowded shopping center parking lot, etc.

They don't pretend to be able to handle heavy snow yet, but there are plenty of places without snow. The cars use the detailed maps to determine where it is to within a few centimeters (without gps) by comparing the map with the sensor data. If everything is covered in a meter of snow the maps won't be of much use though. Even humans can have problems determining where the road is in such conditions. Might be a while before they solve that.

From the DARPA grand challenge course:
 

Online 0xdeadbeef

  • Super Contributor
  • ***
  • Posts: 1493
  • Country: de
Re: Robo-Cab Predictions
« Reply #7 on: May 15, 2019, 06:13:13 pm »
That DARPA challenge course looks like a piece of cake compared to driving through most European cities in the rush hour.

And regarding Tesla's "autopilot": It's what other manufacturers had for years but called it something like "lane departure warning system" or "adaptive cruise control".
Actually I read a pretty balanced article lately about driving experience with a model 3 and the author stated the "auto pilot" worked worse than similar systems of other (gasoline/diesel) cars in the same price range.
Tesla's achievement is mainly to create the impression to be ahead of anybody else regarding autonomous driving even though they drag behind in ACC etc.
« Last Edit: May 15, 2019, 06:20:41 pm by 0xdeadbeef »
Trying is the first step towards failure - Homer J. Simpson
 

Offline apis

  • Super Contributor
  • ***
  • Posts: 1667
  • Country: se
  • Hobbyist
Re: Robo-Cab Predictions
« Reply #8 on: May 15, 2019, 06:39:22 pm »
That DARPA challenge course looks like a piece of cake compared to driving through most European cities in the rush hour.
Yes, driving on dirt roads isn't the difficult part. What might be difficult for humans can be easy for a self driving car and vice versa. After the grand challenge there was the urban challenge in 2007 which the Stanford team also won. Then they managed to convince Alphabet to invest in a self driving car project which later became Waymo. Since then they have focused on driving in all kinds of traffic.

https://waymo.com/360experience/
 

Online 0xdeadbeef

  • Super Contributor
  • ***
  • Posts: 1493
  • Country: de
Re: Robo-Cab Predictions
« Reply #9 on: May 15, 2019, 07:05:17 pm »
My main issue with the common approach of training a system in a well defined scenario is that this implicitly means that anything outside that scenario will cause a more or less unpredictable and thus potentially catastrophic behavior.
Like in 2016 there were reports of Google's self-driving car getting confused by cyclists executing a track stand. Now I guess they trained their cars in the meantime to cope with that but what happens if a clown on a unicycle appears next to the car? Or a bear on a unicycle? Or an elephant that escaped from a circus? What if there is a hailstorm at the same time or an eclipse or the road is undermined by water and somebody in a penguin costume is trying to warn people about a rampage?
There are just too many unexpected situations that you can't train you system for so you need a system that can actually judge a complex situation just as a human could. Even if it involved elephants or people in penguin costumes. And that is just not going to happen soon.
Trying is the first step towards failure - Homer J. Simpson
 

Offline apis

  • Super Contributor
  • ***
  • Posts: 1667
  • Country: se
  • Hobbyist
Re: Robo-Cab Predictions
« Reply #10 on: May 15, 2019, 07:14:50 pm »
Unpredictable situations are naturally the most difficult. The basic parts of the drive system is deterministic though, it will avoid driving into anything the lidar sees whether it's a bear or someone wearing a penguin costume. Self driving cars don't have to be 100% safe, nothing is. Human drivers certainly are not: "In 2010, there were an estimated 5,419,000 crashes, 30,296 deadly, killing 32,999, and injuring 2,239,000." source. If robocabs are much safer than that it's acceptable if they fail in some very rare cases.
 
The following users thanked this post: nctnico

Online 0xdeadbeef

  • Super Contributor
  • ***
  • Posts: 1493
  • Country: de
Re: Robo-Cab Predictions
« Reply #11 on: May 15, 2019, 07:32:30 pm »
Currently, autonomous cars just stop when something unexpected happens - but that is not a behavior that can be accepted in any situation. Actually stopping or blocking a street can kill people in certain situations.
Besides, not judging a situation correctly can also result in driving into a truck that was mistaken for a traffic sign or whatever.
And yes, human drivers make mistakes as well, but not the same idiotic category of mistakes unless they are drunk or copulating or both.
Now coming up with numbers of accidents caused by humans is somewhat pointless as obviously there is no way to compare that to issues caused by autonomous cars given the very few of them that only drive in the brightest sunshine after the track was examined in each and every detail. Actually, taking into account the ridiculous small number, slow speed, ideal conditions and possibility of human intervention, the accidents caused by prototypes are somewhat alarming. Admittedly, most accidents were caused by unexpected braking (i.e. another car hit the dumb autonomous car which was confused and hit the brakes without any warning). But there was already at least one fatality despite of the somewhat ideal conditions and low number of self driving prototypes (which most likely would not have happened if they hadn't disabled Volvo's emergency brake system).
Trying is the first step towards failure - Homer J. Simpson
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 19839
  • Country: nl
    • NCT Developments
Re: Robo-Cab Predictions
« Reply #12 on: May 15, 2019, 07:53:10 pm »
People will use them for sex, drinking, and leave all sort of garbage behind.
That is the first thing I told my wife: if we buy a self driving car we can have fun in the back seat while driving around (after a stop at Ikea to buy some curtains and sticker to put onto any camera to avoid getting lawsuits from people being traumatised from seeing my hairy buttocks).
« Last Edit: May 15, 2019, 07:55:09 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 2338
  • Country: pl
Re: Robo-Cab Predictions
« Reply #13 on: May 15, 2019, 07:54:39 pm »
People will use them for sex, drinking, and leave all sort of garbage behind.
People will fill them with TNT suitcases in some back alley and dispatch them on their last journey >:D
Not even kidding, I'm sure genuinely criminal uses will be found for dumb machines which blindly deliver arbitrary payloads to arbitrary destination.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 19839
  • Country: nl
    • NCT Developments
Re: Robo-Cab Predictions
« Reply #14 on: May 15, 2019, 07:56:20 pm »
Quote
Anyone talking next year is kidding themselves.
Weed smoking Musk has promised exactly that, putting them on the streets next year. Yeah, another visionary thing. With cannabis the term "visionary" takes a new meaning.
Don't expect anything from Tesla. They probably won't exist in a couple of years. Either bankcrupted or borged by a Chinese firm.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline apis

  • Super Contributor
  • ***
  • Posts: 1667
  • Country: se
  • Hobbyist
Re: Robo-Cab Predictions
« Reply #15 on: May 15, 2019, 09:16:02 pm »
Currently, autonomous cars just stop when something unexpected happens
How do you know that? In fact, the last thing I heard from a Waymo official was that the car would pull over and stop if there was a situation it couldn't deal with. That's the same thing a human would do. (And in some occasions stepping on the brakes is the right thing to do.)

Besides, not judging a situation correctly can also result in driving into a truck that was mistaken for a traffic sign or whatever.
Yes, and the same applies for humans. What matters is how often something like that happens, if it happens less often with robocars then they are better obviously.

And yes, human drivers make mistakes as well, but not the same idiotic category of mistakes unless they are drunk or copulating or both.
Humans both drive when drunk and when copulating (and probably both), they also drive while using their smart phone or when they are too tired or distracted by other things. They also cause accidents when at their best. But in the end it just comes down to who killes fewest people. I don't care if I get run over by a human or a self driving car, what I care about is how likely it is to happen.

Now coming up with numbers of accidents caused by humans is somewhat pointless as obviously there is no way to compare that to issues caused by autonomous cars given the very few of them that only drive in the brightest sunshine after the track was examined in each and every detail. Actually, taking into account the ridiculous small number, slow speed, ideal conditions and possibility of human intervention, the accidents caused by prototypes are somewhat alarming. Admittedly, most accidents were caused by unexpected braking (i.e. another car hit the dumb autonomous car which was confused and hit the brakes without any warning). But there was already at least one fatality despite of the somewhat ideal conditions and low number of self driving prototypes (which most likely would not have happened if they hadn't disabled Volvo's emergency brake system).
Only one of Uber's prototype cars have caused a fatality (not counting Tesla since we seems to agree it's not a self driving car).

Waymo has driven autonomously over 16 million km (10 million mi) now without causing any accidents. As you rightly point out they have sometimes been hit by tailgaters when the Waymo car did something "stupid" like stop for a red light or a stop sign, but I would argue that is a problem with human drivers not the self driving cars. So the stats are looking pretty good for Waymo. Still, we won't get any real data until they begin driving without safety driver completely, which should happen within the next few years.

Some companies should't have been allowed to test their cars on public roads. Uber had not only disabled Volvo's crash avoidance system, but also their own built in emergency brake system (which actually detected the pedestrian in time and would likely have prevented the accident if it was enabled). Why? Because it had so many false positives it was basically broken. Instead they relied on the safety driver who was watching tv on the phone. Other companies normally use two highly trained safety drivers in a car that has so little autonomy.

Btw, Uber is now working on autonomous flying cars instead:  :scared:
 
The following users thanked this post: nctnico

Online 0xdeadbeef

  • Super Contributor
  • ***
  • Posts: 1493
  • Country: de
Re: Robo-Cab Predictions
« Reply #16 on: May 15, 2019, 10:43:14 pm »
Pah, Uber. These two geniuses will bring air taxis to Germany ;)

I guess only Germans will fully understand this, but let's say if if we are competitive in one field, than it's moron politicians.
Trying is the first step towards failure - Homer J. Simpson
 

Offline Electro Detective

  • Super Contributor
  • ***
  • Posts: 2713
  • Country: au
Re: Robo-Cab Predictions
« Reply #17 on: May 15, 2019, 11:24:17 pm »

All good on EEVblog host's "Robo-Cab Predictions"  :-+

BUT...  ???

The last part of PREDICTION #4: may be a bit of a fail unfortunately

Quote: "Although, Facebook could be dead by then..."


I'd bet most of the addicted hoards will cash out first, before FB goes up in digital smoke
(an event already 10 years overdue btw  ;D)

 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 5399
  • Country: au
Re: Robo-Cab Predictions
« Reply #18 on: May 15, 2019, 11:49:41 pm »
I would argue that real autonomous driving (in a real world traffic situation you could encounter today in a car) could only work reliably with real artificial intelligence (and not some brainless neuronal network giving somewhat nondeterministic results) so it could react to unexpected situations at least as good as a human driver.
So even ten years from now sounds pretty much optimistic unless there is some unexpected breakthrough in AI. While I would really look forward to have autonomous driving in twenty years or so when I'm retired and half-blind, I would not entrust a car with my life that will decapitate me when mixing up a traffic sign with a truck.
It's somewhat shocking to see how easily today's image recognition systems can be tricked into totally overlooking traffic signs, persons and what not.

I am not a robot, but I remember many years ago, at night, following a "ute" which had a warning pattern painted upon its tailgate, identical to that used to indicate that a road is terminated at a "T" junction.

It came to just such a junction, with just such a sign, & the ute virtually disappeared as it slowed & the two patterns merged together.
Any optical recognition system would have been confused by this-------hell! I knew he was there, but the only thing that really stood out was his tail lights.

And this was a nice clear night!
 

Online 0xdeadbeef

  • Super Contributor
  • ***
  • Posts: 1493
  • Country: de
Re: Robo-Cab Predictions
« Reply #19 on: May 16, 2019, 12:44:19 am »
Yeah, e.g. good luck with that, Uber:

There is a certain creative energy in some German cities like here:

Or why not use another color?

Or let's try metal:


Even the good old "German Autobahn" holds some surprises for autonomous cars:



Now imagine that at night with some snow and this yellow marking foil tends to detach creating funny new direction proposals.
Also these metal walls tend to be bent back and forth and are hard to see at night and sometimes the yellow markings are actually on the wall.
Trying is the first step towards failure - Homer J. Simpson
 

Offline BravoV

  • Super Contributor
  • ***
  • Posts: 6849
  • Country: 00
  • +++ ATH1
Re: Robo-Cab Predictions
« Reply #20 on: May 16, 2019, 12:57:13 am »
My "fundamentalist" view only, why bother with the "vehicle" ?

Don't see this going anywhere until the AI is matured as human, period.

Fact, a motorized vehicles that have been used for centuries, matured technology, running on a "fixed & isolated road" with no ordinary human ever use it normally, and with today's technology advancements like telemetry, wireless communication, sensor technologies, powerful computer and etc, still we have these ...

-> Latest in 2019  :palm:

And further references :
-> List of accidents for the past century

Now, are you sure you want this robo thingy runs in the road where humans still freely roam at it's vicinity after reading above facts ?  :palm:


PS : Word "road" above is deliberately substituted for "rail", and "vehicle" for "train".  >:D
« Last Edit: May 16, 2019, 01:13:43 am by BravoV »
 

Offline Homer J Simpson

  • Super Contributor
  • ***
  • Posts: 1120
  • Country: us
Re: Robo-Cab Predictions
« Reply #21 on: May 16, 2019, 01:16:57 am »
 

Offline splin

  • Frequent Contributor
  • **
  • Posts: 998
  • Country: gb
Re: Robo-Cab Predictions
« Reply #22 on: May 16, 2019, 03:09:28 am »
I reckon it will be a *lot* longer than 10 years before we see autonomous cars - if ever. Don't forget the engineering maxim "the first 90% of a project takes 10% of the time. The last 10% takes the other 90% of the time". IE. The easiest parts are tackled first and the remaining issues get increasingly harder - autonomous driving being a perfect example.

Hell, in the UK the 'smart' meter development programme has been running for over 10 years and the costs could rise to £20B - for a bloody meter that only has to count kW hours and send the numbers to a data centre periodically. And it's only just started rolling out meters that meet the current standard. That should be way simpler than autonomous driving - though like most developments benefitting from government direction (interference) it's way more complicated than it should have been.

Double hell - I'm still waiting for a house robot that will fetch me a beer and find the TV remote after 50+ years of promises from 'Tomorrows World' etc. How hard can that be?


The basics of driving, navigation and situational awareness are relatively straightforward (but not easy). The problem is the myriad of corner cases and unexpected situations that have to be dealt with - the traffic sign that has slipped and is now upside down; the washing machine that fell offf the back of a truck; the breakdown requiring a short reverse down a one-way street; etc. ad-infinitum. If a car stops everytime it encounters a situation it doesn't know how to deal with gridlock will quickly follow.

You either have to have a team of very clever people considering how to design the hardware and software to reliably deal with every conceivable possibility and then have some way to handle the unconceivable ones (or much more likely, the conceivable ones that didn't occur to the developers because being hip and eco only use electric scooters and public transport and don't drive). Alternatively you have to come up with something that can self determine the best course of action in every situation that arises. The first would require vast engineering resources, the latter extremely good AI. That may be the answer and I don't know how good the state of the art is but I'm not holding my breath.

Some of the more interesting problems will arise from the non-autonomous vehicles co-occupying the roads. Human drivers will have a great deal of fun exploiting the safety first characteristics of self driving cars by, for example, forcing their way into traffic from a side street knowing that the autonomous car will be programmed to give way to avoid a collision. Or pranksters competing to come up with the funniest/cleverest scheme to halt or misdirect autonomous cars by the subtle application of bits of tape to road signs etc. Changes to the law will probably be the answer along with automated reporting of dash-cam footage showing aggressive human driving behaviour etc. but these will take time to implement and both sides will adapt and evolve; but how long will it take?

For as long as I can remember there has been no shortage of headlines trumpeting major scientific breakthroughs every few months. But almost invariably, after the initial excitement, it turns out that they've only developed the first 90% and all it requires for the last 10% are a few minor wrinkles to be ironed out, somebody to decide the best shade of blue to paint it, a solution to the Goldbach conjecture and the discovery of some new materials which can withstand 5000C whilst being a bit stronger than graphene at half the weight.

Humans both drive when drunk and when copulating (and probably both), they also drive while using their smart phone or when they are too tired or distracted by other things. They also cause accidents when at their best. But in the end it just comes down to who killes fewest people. I don't care if I get run over by a human or a self driving car, what I care about is how likely it is to happen.

You might not care but almost everybody else will. People tend to get very excited about deaths caused by things out of people's control and very quickly start baying for blood/something to be done/heads to roll. Rational consideration of cost - benefit tradeoffs won't be at the fore-front of most observer's minds (unless they happen to be engineers). When a human f**ks up and kills someone it's an accident. When something manufactured f**ks up and kills someone because of a design defect it's a company/corporation head that has to roll. Laws will have to change radically to allow autonomous vehicles to operate without the threat of company destroying liability costs arising from every incident involving human injury or death.

Case in point: it's my suspicion that railways in the UK are much too safe. The death rate on the railways is extremely low compared to the roads - which is a result of public demands for change after every major accident. That almost invariably means more cost and as a result we have a fabulously safe, but hideously expensive railway network that hardly anyone uses because it's too expensive. Eg. a child gets killed at an uncontrolled crossing resulting in all such crossings to be closed and fences to be erected alongside every inch of track to prevent any inadvertent access. When I say hardly anyone, that's relative to road usage of course - our trains are regularly overcrowded but only account for less than 9% of passenger miles. (Yes I do understand that many of the most important safety improvements don't incur significant costs - but many do.)

There are plenty of youtube videos showing the difficulties and problems that self driving cars get into for those interested.
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 10099
  • Country: au
Re: Robo-Cab Predictions
« Reply #23 on: May 16, 2019, 03:34:00 am »
Humans both drive when drunk and when copulating (and probably both), they also drive while using their smart phone or when they are too tired or distracted by other things. They also cause accidents when at their best. But in the end it just comes down to who killes fewest people. I don't care if I get run over by a human or a self driving car, what I care about is how likely it is to happen.

You might not care but almost everybody else will. People tend to get very excited about deaths caused by things out of people's control and very quickly start baying for blood/something to be done/heads to roll.

This.  Most definitely this.  Add the Media into the mix and you are guaranteed to have a reaction a thousand fold greater for one death from autonomous vehicles than a hundred from human driven ones.

It's sad to say, but this will only start waning when we get to the point that autonomous vehicles become more pervasive and we as a population can start getting into the statistical comparison mindset that apis has already expressed.  We are several decades away from that, IMHO.
 

Offline Red Squirrel

  • Super Contributor
  • ***
  • Posts: 2480
  • Country: ca
Re: Robo-Cab Predictions
« Reply #24 on: May 16, 2019, 03:39:26 am »
Quote
Anyone talking next year is kidding themselves.
Weed smoking Musk has promised exactly that, putting them on the streets next year. Yeah, another visionary thing. With cannabis the term "visionary" takes a new meaning.
Don't expect anything from Tesla. They probably won't exist in a couple of years. Either bankcrupted or borged by a Chinese firm.

I hope they stick around.  Love or hate them, I think they are pretty much driving the EV market right now.  If they die, the EV market goes with it.  Teslas are basically the apple of EVs and everyone else is trying to have a similar offering.  Except unlike phones, there is not necessarily a huge demand for them so if they go away, that's that.   Of course I could be wrong, the other companies could still continue with their own EVs but I feel they are just doing it because they want to have something they can have to compare with Tesla.  They can tell a customer "It's electric just like a Tesla, and those are cool right?".
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf