Translate

Sunday 21 February 2016

Part 5 - Where is here anyway?

Now seems to be a good time to have a look at where some manufacturers are at the moment.

As always money talks, so we will start at the top end of town where development usually starts and then flows down.

Mercedes 2016/17 E-Class, Teslas, Volvo, BMW and Audi are probably the closest we are yet to fully autonomous vehicles but I would suggest that Mercedes probably leads the field in mature tested technology.

 The new Mercedes E Class offers adaptive cruise control, keeping the correct distance from the vehicle in front and adjusting speed accordingly.

Interestingly, "speed limit pilot" doesn't rely on maps or GPS to determine the correct speed but actually reads the road signs and adjusts speed accordingly.

"Steering pilot" uses other vehicle and buildings to determine correct position when no lane markings are present.

Cross traffic detection at intersections with auto braking if the driver doesn't respond.

A real life saver is the detection of a hazard at the tail end of traffic where there is no room to maneuver and can brake earlier and avoid accidents totally and automatically up to 70 Kph in that situation.

Evasive active steering where a driver swerves the vehicle it can detect the hazard such as a pedestrian and control the direction and speed allowing easier recovery.

"Remote parking pilot" allows the car to park itself and get itself out and ready to go when requested from a phone app.

Vehicle to infrastructure or vehicle to vehicle communications (V2X) will be built in to the new series and accessed via mobile phone so events ahead such as collisions or roadworks can be known in advance.

Auto lane change is another feature. When the driver puts on the indicator to change lanes the vehicle checks if it is clear to do so and if safe will change lanes. At this point it isn't clear if it reads the road markings or uses look ahead radar to determine the legality and safety of the maneuver or relies on the driver's skill to determine that but it will detect oncoming vehicles or overtaking vehicles from behind and will execute the maneuver only when safe.

In compliance with the current laws in most countries it insists on the driver having hands on the wheel ready to assume control and if it doesn't detect that, it slows to a controlled halt.

In Nevada in the US Mercedes claims to have the first autonomous testing license plate for a full production vehicle.
These are just some of the safety features as we are primarily interested in the autonomous side. Luxury is of course way up there and the tech isn't visible apart from the configurable dash display screens, which is a good thing in a luxury car.

Tesla S. This is the newer kid on the block, an electric sports car with all the bells and whistles with a price tag to match.
Electric cars are the path to the future and in this regard Tesla has a distinct edge with its tech and with 0-100 Kph in 3 seconds it also has a lot of fans.

Autopilot, as Tesla calls their automation allows lane holding, lane change with indicator activation, adaptive cruise control, ESC and auto braking. It can also scan for parking space and parallel park.

These features are updated by software.
These are already on the road in Australia and are being tested using autonomous mode as most people would if they had access to them:
 Audi has  lot of technology in their concept vehicles and have done a lot of testing but their driver assist features appear to be more in the range of warning systems rather than direct action. That is it tells the driver rather than taking control directly.

From what I can gather they are taking a more transitional approach, that is building in the tech but not yet enabling it until the laws change and focusing more on the driver being warned rather than assisted.
 It appears that traffic jam control is a standard feature but prompts for driver control once the jam is cleared and speed increases. Probably a more conservative approach than some others.

Volvo, long know for driver safety has stated that by 2020 no one will be killed or seriously injured in a new Volvo car or SUV.

They are really really serious about safety and so confident in their technology for autonomous vehicles that they are the first that I know of to put their money where their mouth is:

Volvo CEO: We will accept all liability when our cars are in autonomous mode.

This of course is one of the big questions looming over the widespread use of autonomous vehicles, who has liability in the event of a collision?
But if you consider that in the long term there will be a huge reduction in the number of collisions then it starts to make economic sense for the manufacturers to wear this, but more of that in a later article.

As stated in an earlier blog, Volvo will have 100 autonomous cars on the road in normal drivers hands (or not if in autonomous mode but I digress) in the "Drive me" project in Gothenberg commencing in 2017 utilising Volvos IntelliSafe Autopilot-equipped XC90s.
This a fairly useful test as they will be used daily on commuter routes that will be at average speeds of 70 Kph. This daily use in real world conditions will go a long way to gaining acceptance of these vehicles on the road. The story on Volvos web site makes interesting reading and lays out their reasoning and why they want to do this now.

My read in this blog on the level of automation is derived from what information is available publicly, and the level of  automation on the road already in production vehicles. Industryweek.com had a summary of a report from navigant research in October 2015, based on not only current but future potential and is based on this criteria:

The analysis looked at the strategy and execution of autonomous vehicle technology for 18 automakers. The OEMs were rated on 12 criteria: vision, go-to market strategy, partnerships, production strategy, technology, geographic reach, sales, marketing and distribution, product capability, product quality and reliability, product portfolio, pricing and company commitment.

The full article is available but at a cost but the above is a summary that shows the relative rankings.

Obviously more detail than I have access to and makes a good read.

Interestingly Honda, Toyota and General Motors from mid price range manufacturers are all touted as possible leaders in this technology in the near future.

If nothing shiny comes along to distract me, next week I will look at the cars aimed at the average car owner.

Until then.

Saturday 13 February 2016

Part 4 - Up up and away.

Today I want to start with a couple of interesting developments that have occurred this week.

Firstly the US National Highway Traffic Safety Administration (NHTSA) has replied to a Google submission that the "self driving system" can be driver of the vehicle. This is in the situation where there are no manual controls and the car itself is the driver.

There are other issues surrounding that but the NHTSDA may change their rules based on submissions from manufacturers regarding some regulations on auto safety feature including foot operated braking systems. The full story is here .

A video of the statement is here 






This is also interesting in that California is formulating laws that require a licensed driver in a vehicle ready to assume control at all times when running in autonomous mode. It will be interesting to see where this goes now.

 This new NHTSDA statement brings up a whole new range of options.

Google argues that having controls within an autonomous vehicle is actually a risk as drivers may try to take control of a situation that the computer can handle better.

Given that the car may legally be the driver, does that mean that the driver assist development going on as a transition to fully autonomous vehicles can be bypassed and move directly to the end goal?

If this is the case then Google, Lyft, Uber and others that want to bring in automated taxis will suddenly be in an arms race to be the first to market and the whole process could happen far faster than originally anticipated. Google has stated that it wants them on road by 2020, this new statement may even accelerate that.

I think that the development of autonomous taxis is the catalyst that will launch the entire world wide move to autonomous vehicle for a very simple reason. The cost to manufacture anything new before economy of scale kicks in is very expensive. To build a consumer vehicle with all these features would put it initially out of the reach of the average driver.
To a company like Google, Lyft or Uber it makes a great deal of sense to be the first to market and grab a huge slice before anyone else can. When you can afford to order these vehicles in taxi fleet quantities the unit cost is far lower than it would be for a normal consumer. Once these are mass produced to a standard model the cost of custom produced parts becomes less as they become standard parts produced in bulk. 

Back off the soapbox and back to the news.

Secondly Right here in Perth Western Australia we will soon have an autonomous bus appearing on the streets. 
Purchased by the RAC this vehicle will be tested first on the RAC private testing area before being considered for use in public.The electric bus, built in France can carry up to 18 passengers at up to 45 kph. The state government will be using this as a catalyst for the legal framework required in Western Australia to allow autonomous vehicles on the road. Interestingly enough the law in WA requires a licensed driver with at least one hand on the wheel at all times while the vehicle is in motion.
The official media statement from the Minister for Transport is here.
An interesting quote from that media statement that is being heard more frequently just about everywhere is: "It is not a matter of if this technology will come to WA, but when it will, and that time is fast approaching. " 

This is the type of bus being trialed.



Now that I have started of with current news I might carry on in that vein this week, mainly because I am too busy to write something truly meaningful.

Google, arguably the leader in development of autonomous cars and one of the few I haven't for some reason referred to frequently (possibly because the Mercedes and Teslas are way cooler) has plans now for its electric autonomous car to be charged while driving.


Meanwhile in other news, the military has taken a keen interest, considering that the basically kicked of the whole process with their original challenges in 2004 but according to this article are only looking at the simpler part of driverless convoys.

An interesting debate is whether autonomous cars will largely replace private ownership and become a car as a service. An interesting article sees the US automakers backing the thought that people will want a car like a mobile phone, not hired but owned.

On the other hand Uber is openly discussing a driverless world and its plans are squarely aimed in that direction and as the Uber CEO stated "When there's no other dude in the car, the cost of Uber becomes cheaper than owning a car" 

I would welcome feedback and possible debate as this is all new ground and, as yet, there is no "True Path"as to what will happen in the future and all options are still open.

So help me out here I want to start some discussion that might help to get this whole thing more widely known.

Feel free to contribute.




Saturday 6 February 2016

Part 3 – The good, the bad and the ugly.

This week I want to talk about the social impact of automated vehicles and how they will change the world as we know it.



The impact of autonomous vehicles and the way it will change roles and processes within society will result in this being the most disruptive technology to date, up to and including the Internet. That is a bold statement but the implications for society are huge, and that’s no exaggeration.

Motor vehicle collisions today are a daily occurrence and estimates put 95% of these down to driver error.
Road accidents are a major cause of death in many countries. Estimates indicate that approximately 400,000 people die every year in road accidents around the world. Deaths and injuries caused by road accidents result in significant social and economic costs and it has been estimated that in OECD countries, approximately 1 - 2 per cent of GNP is lost every year due to road traffic accidents.” (Report from Monash University)


Let’s jump forward 10-20 years and see what we have.

By this time autonomous vehicles will be the vast majority of vehicles on the road and a few things will have changed. Manual vehicles, probably not outlawed, but most likely would be required to have a transmitter to alert autonomous vehicles and other manual vehicles as a warning that they can be unpredictable as they have people driving, much like learner plates in Australia today warn other drivers to expect possible erratic behaviour.
Collisions will have reduced dramatically and the effect of this are worth examining in more detail. as the flow on effects are huge.

To start with, let’s have a look at the police and how their role has changed.

The good

Huge resources have been tied up in traffic enforcement since motorised vehicles first appeared.

Now, for the first time, police can be released to do real policing.

No (or minimal) traffic patrolling, no road rage, speeding or drunk driving. No random breath test, license plate scanning or driver license enforcement. No fine follow up and license testing for most users. No hoons, burnouts or high speed pursuits. No traffic duty when lights fail or responding to road collisions and best of all...... Not having to deliver bad news to family that there has been the death of a loved one in a vehicle, and as any police officer anywhere will tell you, is the worst possible part of policing today.

Apart from the toll on the people involved in accidents, the trauma also affects all emergency personal as well, particularly where kids are involved.
Imagine the number of officers freed from these duties and the reduction in cost of the police vehicle fleet.

Now, imagine the flow on effect even further with regard to hospital and ambulance services, trauma teams and reduction in ongoing intensive rehabilitation, prosthetic limbs, wheelchairs and access, and trauma from losing families. Huge resources within the medical system freed up, reduced or redirected to more productive areas.

Main Roads will be affected in a large way as it is forecast that traffic will decrease dramatically due to less personal ownership, therefore no or greatly reduced traffic lights. No more roundabouts, Speed humps or, hopefully, congestion, although that is debatable.

Here, in Western Australia the only government document I have seen so far is in this interesting paper on the whole subject by Main Roads themselves, but even this carries a disclaimer "The views and opinions expressed in this report are those of the authors and do not necessarily reflect the official policy or position of Main Roads Western Australia".

The development of warning systems will need to be done where conditions change due to scheduled or remedial roadworks occurring by Main roads and other authorities.

The flow on effects continue, to suppliers of police vehicles, speed cameras, traffic light manufacturers and ambulances, repair services, suppliers and contractors involved in these systems.

The bad
Every silver cloud has a black lining or something like that.

Many people who work in transport and related industries will be out of work or doing jobs totally different to what they do now.

I believe the first major hit, and this is probably closer than you may think, will be taxi drivers. With the ongoing war with Uber in Australia and other companies such as lyft overseas the taxi drivers are feeling the pinch now. With these ride share companies inking deals with car manufacturers to produce autonomous taxis, the writing is on the wall.
Unsuspecting Uber drivers will be hit as well.

Truck drivers won't be a long way behind with automated delivery systems coming into play and long distance deliveries fully automated. That leads to truck stops, motels and service stations losing large amounts of trade.

 
The flow on effects mentioned earlier will all have impact on jobs, people building roads, traffic lights, service vehicle sales,fleet suppliers and lease companies will all have to change the way they do business. Insurance companies will need to reassess their whole industry and determining the legal liability is a minefield all of its own and will be the subject of a later article.

The Ugly
Driving habit as automated vehicles start to appear will create a whole heap of issues.

Initially do we need to mark autonomous vehicles someway? If so there will be idiots who force their way in front because they know the car will automatically give way to avoid collision. This could cause collisions if the autonomous car is being driven manually.

Laws determining when and where they can be used will be a major issue and many battles will be fought in this arena.

Car parks will have to be redeveloped and councils will lose the current huge revenue they receive through parking fees and fines. Governments will lose income from drivers licenses, vehicle registration and the state government insurance office will lose a large part of its business if the number of vehicles on the road reduce as much as is being predicted.

Panel beaters will be another greatly reduced industry as will car accessory makers and resellers.

This, and the previous posts are just an overview of what is to come and the depth of detail out there is immense and growing daily.

Next week I want to look at some of the tech involved and how it works, then, I may look at differences in outlook between manufacturers.

 
Stay tuned....