Autonomous Vehicles

sesevans said:
@satchef1

Remember also that with AVs, vehicle design can change to accomodate multiple users on the same trip to the same "supermarket" in your example, thereby driving down the cost of operating the AV to the supermarket owner.
Click to expand...

Yes but no one will want to share a car with random people, especially late at night. Or if they are driving through bad areas.
 
Nasher said:
Yes but no one will want to share a car with random people, especially late at night. Or if they are driving through bad areas.
Click to expand...

I suggest most people who go to supermarkets do so during the day or early evening when produce is likely to be freshest. But if you think otherwise I am open to rethink that assumption.

But even if for example, work does not permit an earlier visit than "late at night", supermarkets could still instruct AV car design with great security features---perhaps compartments, cameras, etc? Just think outside the box.

Great economies of scale can be achieved through AVs where the cost of mobility can be reduced by replacing the human driver. With a taxi service today, a large percentage of the operating cost is the cost of the driver.
 
http://www.independent.co.uk/life-s...et-philip-hammond-town-planning-a8119551.html

Article and comments in today's Independent is similar to many of the same points we have discussed over the past month or so. Author notes UK Government, while spending £ 1 billion in support of autonomous driving (eg, £400 million on car chargers, £75 million on AI, etc) is not thinking through many of the bigger issues: how road space will be divided up, potential for hackers to cause disruption, etc Article points reader towards an unpleasant, dystopian future.

But little in the article talks about ride sharing and seems to automatically assume that AVs will be largely privately owned, which as I have argued seems the less likely path to travel.

Comments?
 
Living in a home with off street parking, I had not considered the dilemma for people wanting to own EV's who have no off-street parking. The FT discusses this problem today with following article:

" https://www.ft.com/content/b9cd3dbc-e406-11e7-97e2-916d4fbac0da

Five ways to solve UK’s electric car ‘charging conundrum’
Companies face challenge of 43% of households having no off-street parking

Almost 9m electric vehicles will one day have to be charged away from home because 43 per cent of British households do not have access to off-street parking, according to estimates by National Grid, operator of the country’s electricity system.

How to solve this problem will be one of the key challenges facing policymakers as the UK works towards a ban in 2040 on sales of new petrol and diesel cars (hybrid vehicles are not affected by the move). The Scottish government has set an earlier date of 2032 for the ban.

Several companies are working on possible solutions to the charging conundrum.
1. Lampposts One of the most obvious solutions is public chargers installed on pavements, of which there are an increasing number in the UK. Yet there may be limits to how much extra “pavement furniture” local authorities will allow. To overcome this problem, German company Ubitricity is converting lampposts in some London boroughs, such as Kensington and Chelsea, so that they double as chargers for electric cars. The process is relatively straightforward with the correct permissions, says **** Hechtfischer, Ubitricity’s founder. But there are limitations. The lamppost needs to be close to the kerb, or the chargers will create a trip hazard. Older “heritage” lampposts cannot be converted. Lampposts are also unable to accommodate fast chargers. The idea is to top up batteries “while you are working or while you are sleeping”, says Mr Hechtfischer. Similar to home chargers, if too many drivers try to plug into the power network at peak times — such as after work — then it could lead to local electricity shortages, and network reinforcements will therefore be required. Network operators have warned this year most chargers will need to be “smart” so they only start powering up batteries when the infrastructure can cope.

2. Induction pads Another possibility is induction charging, in which an electric car can power up its battery from a pad under the vehicle. This pad would be connected to the electricity network, and could potentially be located on the street. BMW and Daimler are both developing induction pads that charge a battery through the bottom of the vehicle in a conventional garage. Renault has gone a step further, demonstrating under-road charging for car batteries on a 100-metre test strip in France. Pressure points in the road detect a vehicle’s presence, and electricity passes upwards into the car, charging it as it moves. Gilles Normand, head of electric cars at Renault, says the charging technology can be effective at a distance of 4cm from the vehicle, compared to half a centimetre when the company began testing the arrangements three years ago.

3. Battery swapping One unconventional idea is to replace a flat battery with an entirely new, fully-charged one. The old battery can then be recharged at leisure and reused in other vehicles. This solution is potentially quicker than waiting for a vehicle to charge, allowing motorists to pull up to a swap site and drive off with another battery within minutes. This concept has been floated by Elon Musk, chief executive of electric carmaker Tesla. The practicalities of such a scheme, including large costs to create a network of swap stations complete with full batteries, have prevented the idea from coming to fruition. But it has not gone away. Nio, the Chinese electric vehicle start-up backed by technology groups Baidu and Tencent, plans to use battery swapping as a key selling point when the company launches its electric sport utility vehicle in Beijing. Building a limited number of swap stations initially, rather than embarking on a nationwide rollout, will allow Nio to keep costs down.

4. Petrol stations Energy companies such as Royal Dutch Shell are already installing rapid chargers for electric car batteries at petrol stations. Some carmakers, such as Tesla, have also made fast chargers available at service stations and airports. Supermarkets are also expected to become charging hotspots. Shell will have fitted chargers at 10 of its UK service stations by the end of this year. These can power up a flat battery to 80 per cent of its capacity in about 30 minutes. The downside is availability: Shell is starting off with one charger per filling station in the UK. It cannot be reserved, although drivers can check if it is free via a smartphone app. As electric vehicles become more prevalent, National Grid is considering how hundreds of cars could be rebooted simultaneously without causing problems for the local electricity networks. It has suggested a fleet of superfast charging points could be installed at service stations that are connected directly to the high voltage transmission network, avoiding the risk of local power shortages.

5. At work Plugging in electric cars during office hours is a no-brainer, says Juliet Davenport, chief executive of Good Energy, an electricity supplier. “Charging en route can be tricky to plan and time-consuming while in general, our cars are parked at work for up to eight hours,” she points out. But there are cost implications for employers. Good Energy is running a trial where employees pay for a parking space at work in return for the ability to power up their batteries for free. “The opportunity is hugely exciting, but it is essential that costs are carefully considered,” says Ms Davenport."
 
Nasher said:
Yes but no one will want to share a car with random people, especially late at night. Or if they are driving through bad areas.
Click to expand...

This is a thing that already happens, albeit not really in the UK. In the US when you get an Uber, the cheapest of the options is to get in an Uber with some random person who happens to be going the same direction as you. It actually works really well and you can have some fun conversations as long as you're not socially inept.
 
I'll miss driving, but I'd hope that if everything is automated, speedlimits could be raised as human error/reaction time would no longer need to play a part. Faster journeys are definitely worth it.
 
FishFluff said:
This is a thing that already happens, albeit not really in the UK. In the US when you get an Uber, the cheapest of the options is to get in an Uber with some random person who happens to be going the same direction as you. It actually works really well and you can have some fun conversations as long as you're not socially inept.
Click to expand...

Yes like most Brits.
 
FishFluff said:
This is a thing that already happens, albeit not really in the UK. In the US when you get an Uber, the cheapest of the options is to get in an Uber with some random person who happens to be going the same direction as you. It actually works really well and you can have some fun conversations as long as you're not socially inept.
Click to expand...

I think you might be pleasantly surprised with the possibilities for design of interior of vehicles in AVs. Whether the AVs are for a small number of passengers or many more that can be accomodated in a van or bus, you should probably think of the interiors more in terms of "living rooms on wheels". There will be so much more possibility for "passenger distraction" such as reclining seats, fold up seats, seats that rotate 180 degrees, head-up displays on windshields, home theatre, containers/ice chests for beverages, other displays on glass and more.

The passenger can interact with other passengers quite comfortably or set up a table to work or watch a film, etc. The side view mirrors of today could become cameras that display road information such as speed or traffic for passengers interested in keeping up with what is going on outside.

If the vehicle is used for ride sharing, I anticipate cameras to monitor security and buttons that will allow for a passenger to signal to the "driver" a wish to pull over to the side of the road. The interior of the ride sharing vehicle can be remotely monitored.

Think outside the box.
 
FishFluff said:
This is a thing that already happens, albeit not really in the UK. In the US when you get an Uber, the cheapest of the options is to get in an Uber with some random person who happens to be going the same direction as you. It actually works really well and you can have some fun conversations as long as you're not socially inept.
Click to expand...

The leading China based ride sharing company is Didi Chuxing. They are a near monopoly in China, having taken over Uber's ride hailing business there. They have a fleet of 21 million vehicles on Chinese roads and will turn the fleet into data collectors, esp with venture capital backing of $ 10 billion, competing locally with Baidu, another Chinese firm. Their CEO's goal is to take on Google, Tesla and Uber in self driving vehicles. The CEO admits that Google is in the lead.

One interesting comment by the CEO: "The victorious ride-hailing player will eventually compete against—and sometimes cooperate with—Google and Tesla in self-driving vehicles, with the winner going on to “build a transportation and vehicle system for the future,” he said."

https://qz.com/1167945/didi-is-now-ready-to-take-on-google-goog-and-tesla-tsla-in-self-driving-cars/
 
Chuk_Chuk said:
Unfortunately the box is called weight, cost, and fuel efficiency (petrol or electric) and they won't let you stray too far.

I don't think such a taxi service would be affordable.
Click to expand...

The box for many in this thread is “it’s different to now, therefore it can’t possibly work”. That’s more what he means.

That and “I don’t want it to actually work as I like tagging my car around and am worried I can’t drive dangerously any more of more vehicles are automated”.
 
Chuk_Chuk said:
Unfortunately the box is called weight, cost, and fuel efficiency (petrol or electric) and they won't let you stray too far.

I don't think such a taxi service would be affordable.
Click to expand...

@Chuk_Chuk @Amp34
I continue to believe that a ride hailing AV service is likely to be the first major commercial deployment, ahead of individually owned AVs.

What we do know is that Uber has been running a ride-sharing service for a number of years now and continues to lose billions of dollars per year. Yet both Uber and Didi Chuxing are both the two most highly valued start-ups (ie, venture-backed, non-public companies). Uber hopes to IPO in 2019.

The main cost of today's ride hailing service is human. Take out the human driver and the cost potentially drops from an estimated $ 3 per mile to $ 1 or less per mile. At less than $ 1 per mile, the potential to succeed expands exponentially.

This assumes that sufficient progress is made with reducing the cost and increasing the functionality of sensors and in particular LIDAR. And having the ability to extract the data that you need from this array of sensors and discarding what is not necessary, yet retaining the ability to work out the "edge cases". I appreciate this is a big ask and what leading companies are working on: developing and manufacturing a system that can be deployed at scale with cost effective and maintainable hardware. Solving the first 95-99% is relatively easy compared to the last 1%.

The prize of winning this race is clearly worth it in the minds of creators and venture capitalists. Intel estimates that self driving cars will add $ 7 trillion to the global economy by 2050, and $ 2 trillion in the US alone. And that excludes the contribution this technology could make to trucking and other areas.

Today the technology is too expensive to sell direct to consumers and why I think the approach being taken by Waymo seems sensible: a ride hailing taxi service in carefully selected regions (meticulously mapped), owned by a large fleet operator able to defray costs across a large number of AVs. And one operated in partnership with others: Lyft (an Uber competitor in the US), Avis (to maintain Waymo's fleet), Autonation (to service Waymo's fleet), Truvo (an insurance start up funded by Munich Re to insure passengers in Waymo's fleet). I expect more partnerships to be announced by Waymo in 2018.

Will Waymo's partnership approach guarantee commercial success? Will the vertically integrated approach of GM make more sense? Will Tesla make a big breakthrough based on its engineering skills in space exploration or other areas (by the way if you are a person who owns a Tesla and paid for their Tesla Enhanced Autopilot feature promised six months ago, you are still waiting)?

All valid questions.
 
The Guardian (30/12/17): "Britain could miss out on autonomous vehicle revolution owing to the shortage of charging bays and National Grid limitations"

"Amanda Blanc, CEO of Axa, said a lack of rapid charging bays and pressure on the National Grid have overtaken questions about accident liability as the biggest barriers to autonomous vehicles entering the transport mainstream."

"According to the National Grid, growth in electric vehicles on Britain’s roads could see peak electricity demand jump by more than the capacity of the Hinkley Point C nuclear power station by 2030."

"Until recently, the question of who is liable in an accident involving a self-driving car was seen as a major hurdle in the development of autonomous vehicles – but Blanc said a government draft bill had brought clarity. If it becomes law, insurers will automatically pay out if motorists crash a car in self-driving mode. The question of whether man or machine are to blame will then be debated by the insurance company and the carmaker."

"Others point to progress being made. Waymo, a company that started as part of Google, launched tests of fully driverless taxis on the streets of Phoenix, Arizona, last month".

What can Britain do better to plan for this future?

https://www.theguardian.com/technol...in-obstacle-to-self-driving-car-rise-says-axa
 
You must log in or register to reply here.
Back
Top Bottom