Australian Cycling Forums - Bicycles Network Australia

Self-Driving Cars Could Make Humans Unhealthier Than Ever

https://www.wired.com/story/self-drivin ... ic-health/

This BBC article sums up the arguments for a change to autonomous electric powered vehicles and proposes that most people may have already purchased their last private vehicle and the changeover may be as short as 10 years, similar to the historical change from horse to car? A quoted "think tank" expects a decrease of 80% in the US vehicle fleet in that time and a banning of human drivers.

https://www.bbc.com/news/business-45786690

PS 3,000 comments so certainly seems to stir up opinions

LateStarter wrote:A quoted "think tank" expects ...a banning of human drivers.

He's off with Le Corbusier.

That is, they can't do simple math. Like Corbusier, who never quite clicked to the transport chaos The Internationalist dream would cause

Autonomous cars have a huge obstacle, but Autonomous trains and buses on dedicated lanes present easier challenges and is most probably more effective to solve the traffic jam.

And there is a huge incentive to do autonomous trains and buses as people are pulled into megacities and multistorey resident buildings which makes individual car use and ownership a nightmare for the residents, government and the builders.

opik_bidin wrote:Autonomous cars have a huge obstacle, but Autonomous trains and buses on dedicated lanes present easier challenges and is most probably more effective to solve the traffic jam.

Autonomous public transport won't solve traffic jams because people favour personal transport over public transport even when public transport is available. For example, in Perth public transport usage has fallen since the mining boom while service frequency has remained the same which means that there is currently spare capacity in the public transport system yet we have reliable traffic jams every weekday morning on the Mitchell Freeway immediately adjacent to a train line with the vast majority of traffic going to the CBD.

While autonomous trains will provide improved capacity on the existing train lines it won't address the majority of the reasons why people choose personal transport when public transport is already available. And I say this as someone who is currently employed working on a project proposal for autonomous passenger trains.

Scott_C wrote:

opik_bidin wrote:Autonomous cars have a huge obstacle, but Autonomous trains and buses on dedicated lanes present easier challenges and is most probably more effective to solve the traffic jam.

Autonomous public transport won't solve traffic jams because people favour personal transport over public transport even when public transport is available. For example, in Perth public transport usage has fallen since the mining boom while service frequency has remained the same which means that there is currently spare capacity in the public transport system yet we have reliable traffic jams every weekday morning on the Mitchell Freeway immediately adjacent to a train line with the vast majority of traffic going to the CBD.

While autonomous trains will provide improved capacity on the existing train lines it won't address the majority of the reasons why people choose personal transport when public transport is already available. And I say this as someone who is currently employed working on a project proposal for autonomous passenger trains.

The key here is land use. By concentrating and mixing residential and commercial spaces in one hub, then workplace, school and recreation etc on another transport hub, I think it is doable. If people live far away from each other covering a huge swath of land, the likehood to take public transport is small.

But by clever densifying mixing residential, commercial, services, schools and workplace plus leaving green spaces as belts to address environmental issues, I think it can be done.

Another thing is clever social engineering, narrow the freeway( for example by planting trees on it, taking 1-2 lanes from the freeway) or put barriers and a canopy to make it designated bus/bike lane , and double the frequency of the trains

The real solution to traffic congestion and safety is to provide good alternatives - effective public transport and [perceived] safe cycling. Having said that there are still some small advantages to driverless cars;
1. They can be cooridinated to improve the efficiency of the roads using different routes and speed timing to more efficiently use the road space
2. If they are pooled, ie not privately owned then parking space needs are considerbly reduced resulting in more road space being available

Scott_C wrote:...with the vast majority of traffic going to the CBD.

I agree with pretty much everything you've written but I'm not so sure about this one.

For a couple of reasons: firstly, the drop in public transport use has been attributed to the loss of jobs (office and retail) in the CBD. Secondly, having participated in the northbound carpark a couple of times recently I was aware of how many vehicles didn't get off in the CBD (or were non-CBD to start with - tradies; trucks etc.) Thirdly, Perth public transport is almost entirely focused on CBD and university commuters: any other sort of trip is a right pain in the ****. Try getting to Malaga from Munster before prestart, or to a beach (other than Cottesloe) from anywhere. Or to/from school at any time other than 'normal' time.

I think there's opportunity for someone to start tracking or interviewing the drivers in the twice-daily car-park and find out where they're actually going. A second question would be: are in the car because you had to pick up/drop off school age kids somewhere pre/post work?

Finally, while public usage may have dropped off, it still pretty damn'd busy, particularly on the trains. It's still sardine-o-clock at 4:30pm out off the city. Last train I caught at that time still had no spare seats until Joondalup.

I listened to a podcast on the weekend. The lady on the podcast made a very good point. She had spent her life trying to not get in cars with strange me. The chances of her getting in a ride share car without a driver is nil. Ever.

So who is going to buy these things? Why?

Scott_C wrote:

opik_bidin wrote:Autonomous cars have a huge obstacle, but Autonomous trains and buses on dedicated lanes present easier challenges and is most probably more effective to solve the traffic jam.

Autonomous public transport won't solve traffic jams because people favour personal transport over public transport even when public transport is available. For example, in Perth public transport usage has fallen since the mining boom while service frequency has remained the same which means that there is currently spare capacity in the public transport system yet we have reliable traffic jams every weekday morning on the Mitchell Freeway immediately adjacent to a train line with the vast majority of traffic going to the CBD.

While autonomous trains will provide improved capacity on the existing train lines it won't address the majority of the reasons why people choose personal transport when public transport is already available. And I say this as someone who is currently employed working on a project proposal for autonomous passenger trains.

The biggest public transport barrier is simply the way other passengers behave. Like the guy I observed one morning blow snot all over his hand, and then wipe it off that hand onto the hand railing in the train carriage. This was after he'd spent a good 20 minutes sniffling and snorting away on this fairly silent early morning train. Absolutely disgusting.

g-boaf wrote:
The biggest public transport barrier is simply the way other passengers behave. Like the guy I observed one morning blow snot all over his hand, and then wipe it off that hand onto the hand railing in the train carriage. This was after he'd spent a good 20 minutes sniffling and snorting away on this fairly silent early morning train. Absolutely disgusting.

LOL i think a bit of snot is the least of your worries if public hygiene is your concern

Scott_C wrote:
While autonomous trains will provide improved capacity on the existing train lines

Not doubting you, but curious how this would be the case ?. As an autonomous noob, i would of thought greater safety margins would be inbuilt, at least in these early days, resulting in less frequency ?

Hyundai to give autonomous cars "human intuition"

https://www.carsguide.com.au/car-news/h ... -intuition

Hyundai wants to teach its autonomous cars to think like humans, investing in AI specialists Perceptive Automata to develop new software for self-driving vehicles that will mean your car is as good at spotting unpredictable hazards as you are.

The solution is to make the driverless cars bigger so they can carry say 40 people then set them along fixed routes with specific places to stop. For a security an employee can sit in the front the the vehicle

Ross wrote:Hyundai to give autonomous cars "human intuition"

https://www.carsguide.com.au/car-news/h ... -intuition

Hyundai wants to teach its autonomous cars to think like humans, investing in AI specialists Perceptive Automata to develop new software for self-driving vehicles that will mean your car is as good at spotting unpredictable hazards as you are.

I thought the whole point of autonomous cars is that they would be better at spotting unpredictable hazards than humans, who aren't very good at that

find_bruce wrote:

Ross wrote:Hyundai to give autonomous cars "human intuition"

https://www.carsguide.com.au/car-news/h ... -intuition

Hyundai wants to teach its autonomous cars to think like humans, investing in AI specialists Perceptive Automata to develop new software for self-driving vehicles that will mean your car is as good at spotting unpredictable hazards as you are.

I thought the whole point of autonomous cars is that they would be better at spotting unpredictable hazards than humans, who aren't very good at that

It's a bit of a pointless exercise, isn't it? Maybe it's their 'get out of jail free' card. Whenever an automated vehicle hits a cyclist, the last entry in the vehicle's log will be 'sorry mate, I didn't see you'.

I think I've just solved all the legal liability issues regarding automated vehicles having accidents.

find_bruce wrote:

Ross wrote:Hyundai to give autonomous cars "human intuition"

https://www.carsguide.com.au/car-news/h ... -intuition

Hyundai wants to teach its autonomous cars to think like humans, investing in AI specialists Perceptive Automata to develop new software for self-driving vehicles that will mean your car is as good at spotting unpredictable hazards as you are.

I thought the whole point of autonomous cars is that they would be better at spotting unpredictable hazards than humans, who aren't very good at that

It's different

drivers make errors. This can be helped by autonomous vehicles. Trains, cars, buses, trucks. No, They don't mistake and switch gas-brakes around, being tired and sleepy, etc.

so if it's on a dedicated track with no obstacle, it's fine. Just like trains running on their own rails and no human and animal crossings.

problem is, it's not only vehicles that use the street. Bicycles, crossings, markets, etc. They present that challenge, which in my opinion, cannot be solved, as living beings are unpredictable.

my take is AVs are good on dedicated lanes, but not when living beings are involved. and with this, a private single occupancy vehicle dream will be erasaed, because it isn't practical to get dedicated lanes for each one of them, althugh even now many countries including Australia try. But it opens up cheap mass transit,

mikesbytes wrote:The solution is to make the driverless cars bigger so they can carry say 40 people then set them along fixed routes with specific places to stop. For a security an employee can sit in the front the the vehicle

Already exist. Also known as a bus ?

Arbuckle23 wrote:

mikesbytes wrote:The solution is to make the driverless cars bigger so they can carry say 40 people then set them along fixed routes with specific places to stop. For a security an employee can sit in the front the the vehicle

Already exist. Also known as a bus ?

Peter Newman calls them trackless trams

NASHIE wrote:

Scott_C wrote:
While autonomous trains will provide improved capacity on the existing train lines

Not doubting you, but curious how this would be the case ?. As an autonomous noob, i would of thought greater safety margins would be inbuilt, at least in these early days, resulting in less frequency ?

In trains automatic braking is an old and proven technology that is proven through decades of use to be more reliable than human operators, the WA metro network has had automated braking since the 1990's (in fact the old system is end-of-life which is part of why there is interest in a new automatic driving system).

At the moment the majority of Australian suburban passenger trains operate based on fixed lineside signal locations. For the most common system (3-aspect signalling) the signals are spaced at least braking distance apart for the worst case train that may use that line (so when a train passes a signal at yellow and starts braking it will stop before the next signal which will be at red) as the signals are braking distance apart the distance between the back of one train and a following train on green signals needs to be twice worst case braking distance plus a safety margin (generally 180 to 200m, called the overlap).

When you move to an automated system you need continuous communications with the trains which allows you to get rid of the fixed lineside signals and give continuously variable limits of authority via the comms system (i.e. you can drive up to the 2.050km mark, now the 2.051km mark etc.) this allows you to operate "on green" at a distance of 1 braking distance, for the particular following train, plus a safety margin, in addition as the speed of the following train decreases its braking distance decreases which allows it to come up behind the train in front to a minimum separation equal to the safety margin, in 3 aspect signalling the minimum separation is always worst case braking distance plus safety margin. Overall going from a "green" separation of 2 worst case braking distances to 1 actual braking distance allows you to run trains ~1.5 to twice as often.

You can do a communications based signalling system without the automation but you would need larger safety margins as human drivers are considerably less reliable than the automated braking systems and you have all the information you need for automation once you have the communications based signalling system in place.

Still think it's a real pipe dream.

I mean, the Eastern seaboard can't even switch over to bringing their own plastic bags to the shops, so how are they gonna manage self driving cars .....

Scott_C wrote:
WA metro network has had automated braking since the 1990's (in fact the old system is end-of-life which is part of why there is interest in a new automatic driving system).

Overall going from a "green" separation of 2 worst case braking distances to 1 actual braking distance allows you to run trains ~1.5 to twice as often.

Thanks for reply makes sense, especially if the system needs upgrading as im sure autonomous passenger trains will be a given in the not to distant future.......not sure about cars and buses.

NASHIE wrote: Thanks for reply makes sense, especially if the system needs upgrading as im sure autonomous passenger trains will be a given in the not to distant future.......not sure about cars and buses.

I rode my first autonomous passenger train 12 years ago!

Sydney is completing a train now.

piledhigher wrote: I rode my first autonomous passenger train 12 years ago!

Sydney is completing a train now.

I wasnt very clear, but i was commenting on our backwater mining town Perth

NASHIE wrote:I wasnt very clear, but i was commenting on our backwater mining town Perth

But we have an autonomous bus. Or so the RAC WA claim!