John Deere To Display New Autonomous Concept at Agritechnica

[News]

John Deere new autonomous tractor. The agricultural revolution is here!!!!

 

[nick...]

I’d expect there is lots of things going on which we won’t hear about for years
Nick...

 

[Norfolk Olly]

And still nowhere to put your dinner bag ?

 

[ProDrive]

Be no good this year, nowhere to drop the pin through your chain when it gets stuck[emoji6][emoji6]

 

[Scribus]

More solutions looking for a problem.

 

[Scribus]

Some lad posted this a couple of months back in the 'New John deere Combine seen in states' thread.

At a guess I'd say that JD's big news this autumn will be that they intend running the machines from head office, it may not be a human at a screen, instead, they will have cooked up some super dooper algorithms to identify and 'correct' a wide range of issues in the field in real time. The combines (for example) will be in constant touch with HQ (when the systems are actually working) and so a great deal of control will pass out of the farmers hands if they elect to sign up to the deal, and it will be a deal, and JD will be doing their utmost to ensure customers opt in because they will be charging well for it. For a better idea of what they may have in store it could be worth looking at Caterpillar's Minestar system, the companies are not unknown to each other.

How are they going to sell the idea? We've already seen the puffery about staff shortages and no doubt the company will be crowing about how very futuristic they are and so on. But at the end of the day it will be all about JD's bottom line, it's a big American corporation after all.

We shall see how far off I was.

 

[Pheasant Surprise]

And still nowhere to put your dinner bag ?

No need for a dinner bag, with that “Alexa” will be doing the driving

 

[Scribus]

Automation of John Deere tractors will rely on skills that are vested only within the John Deere organisation, there will no opening for third party engineers or suppliers. Once you have bought into the JD scheme that's it, you pay the JD price or opt out entirely. Of course there will be a good number of farmers who will go with this for eventually it will move to the situation where landowners will basically pay JD to do the farming for them, and save the 'farmer' the hassle of running a business. The company is taking a step towards contract farming so this is not just about some claimed whizz bang technology, in the long run JD are making a move into expanding their entire business model.

But will it work? This is questionable. Yes, it could work on the large mono-cropped central plains of the Americas, Europe and Russia, but autonomous vehicles are a long long way from being acceptable for use on public roads. One of the reasons why the great push for autonomous vehicles (AV) foundered was because of the eye-watering cost of bringing them up to any sort of certifiable safety standard, we're talking billions, not a few million.

In addition to the restrictions on the use of AVs JD have also opted for battery power and appear to have fallen for the usual zero emission BS which suggests they are not so clever as they try to tell us. No mention of cobalt being mined by kiddies (but black and far away so what the hell) or the fact that it is a very finite resource. Perhaps they too are awaiting the magical battery fairy to deliver something which is actually practical.

 

[Clive]

Can't come soon enough !!

I think this concept looks good in that unlike DOT it looks like it would use existing equipment rather than machines tailored to the DOT platform

Full driverless automation of exiting "conventional" tractor is a more likely first step however I think and I guess all that is probably in the way of that is legislation?

 

[Pheasant Surprise]

Can't come soon enough !!

I think this concept looks good in that unlike DOT it looks like it would use existing equipment rather than machines tailored to the DOT platform

Full driverless automation of exiting "conventional" tractor is a more likely first step however I think and I guess all that is probably in the way of that is legislation?

Indeed. The first “horseless carriages” unsurprisingly looked very much like carriages.

I still think it’s properly a few years away yet. The reason I say that “boundary cases” (in software designer speak) on farm - as opposed to say autonomous driving in the public highway - are in many way quite difficult and unique and will take some solving to get right.

I expect it will graduate slowly from quite specialist (but reasonably well bounded) tasks, as we see today in hort. / fruit picking and be perfected there in a modular fashion, rather than a single “general purpose” machine as we have in today’s arable machines.

 

[Scribus]

Can't come soon enough !!

I think this concept looks good in that unlike DOT it looks like it would use existing equipment rather than machines tailored to the DOT platform

Full driverless automation of exiting "conventional" tractor is a more likely first step however I think and I guess all that is probably in the way of that is legislation?

Have you any idea what is involved in rendering AV's safe for the road?

Autonomous vehicles would have to be driven hundreds of millions of miles and, under some scenarios, hundreds of billions of miles to create enough data to clearly demonstrate their safety, according to a new RAND report.

Under even the most-aggressive test driving assumptions, it would take existing fleets of autonomous vehicles tens and even hundreds of years to log sufficient miles to adequately assess the safety of the vehicles when compared to human-driven vehicles, according to the analysis.

All this talk about mere legislation being the only bother which can be sorted at the swipe of a pen is the purest of poppycock.

 

[Scribus]

Indeed. The first “horseless carriages” unsurprisingly looked very much like carriages.

I still think it’s properly a few years away yet. The reason I say that “boundary cases” (in software designer speak) on farm - as opposed to say autonomous driving in the public highway - are in many way quite difficult and unique and will take some solving to get right.

I expect it will graduate slowly from quite specialist (but reasonably well bounded) tasks, as we see today in hort. / fruit picking and be perfected there in a modular fashion, rather than a single “general purpose” machine as we have in today’s arable machines.

Nothing exciting about that, robotic forklifts and trollies have been operating in warehouses for years.

 

[Clive]

Have you any idea what is involved in rendering AV's safe for the road?

Autonomous vehicles would have to be driven hundreds of millions of miles and, under some scenarios, hundreds of billions of miles to create enough data to clearly demonstrate their safety, according to a new RAND report.

Under even the most-aggressive test driving assumptions, it would take existing fleets of autonomous vehicles tens and even hundreds of years to log sufficient miles to adequately assess the safety of the vehicles when compared to human-driven vehicles, according to the analysis.

All this talk about mere legislation being the only bother which can be sorted at the swipe of a pen is the purest of poppycock.

Yes I have a very good idea of what involved

Uber, google, Tesla and others are all on it - some of them have ambition to put men on Mar so I doubt making cars drive autonomously will trouble their budget when developing the tech

but I don't recall mentioning autonomous tractors on the road at all ? - I agree that’s a very long time away yet

 

[Scribus]

Yes I have a very good idea of what involved

Uber, google, Tesla and others are all on it - some of them have ambition to put men on Mar so I doubt making cars drive autonomously will trouble their budget when developing the tech

but I don't recall mentioning autonomous tractors on the road at all ? - I agree that’s a very long time away yet

Actually Clive I don't think you do, certainly not if you rely on the bunch of chancers you mention above. Uber have already killed one pedestrian, Tesla cars, at the the last count, have killed at least three occupants while on auto pilot in between crashing into firetrucks, and Waymo (Google) who started boasting of actually having cars without drivers on the open road have quietly reversed that policy.

So do we take the word of big tech with products to sell or the analysis of these two ladies at the Rand Corporation?

Nidhi Kalra is a senior information scientist at the RAND Corporation, a codirector of RAND’s Center for Decision Making under Uncertainty, and a professor at the Pardee RAND Graduate School. Her research addresses energy, environment, and science and technology policy.

Susan M. Paddock is a senior statistician at the RAND Corporation and a professor at the Pardee RAND Graduate School. Her research includes developing innovative statistical methods, with a focus on Bayesian methods, hierarchical (multilevel) modeling, longitudinal data analysis, and missing data techniques.

Who conclude that -

Our results confirm and quantify that developers of this technology and third-party testers cannot drive their way to safety.

Autonomous Vehicles Cannot Be Test-Driven Enough Miles to Demonstrate Their Safety; Alternative Testing Methods Needed

Under even the most-aggressive test driving assumptions, it would take existing fleets of autonomous vehicles tens and even hundreds of years to log sufficient miles to adequately assess the safety of the vehicles when compared to human-driven vehicles.

www.rand.org

I know who I would rather trust

 

[Chris F]

@Scribus when in the past have a few deaths slowed the progress of big business? 1 billion test miles could be covered by 10,000 test cars in a single year.

However, given 1770 people were killed on UK roads in the 2017/2018 year - I don't see that even now there would be that many deaths with autonomous vehicles.

 

[Scribus]

@Scribus when in the past have a few deaths slowed the progress of big business? 1 billion test miles could be covered by 10,000 test cars in a single year.

However, given 1770 people were killed on UK roads in the 2017/2018 year - I don't see that even now there would be that many deaths with autonomous vehicles.

May I suggest you read the report I linked to where they answer your very point in the preamble -

Given that current traffic fatalities and injuries are rare events compared with vehicle miles traveled, we show that fully autonomous vehicles would have to be driven hundreds of millions of miles and sometimes hundreds of billions of miles to demonstrate their safety in terms of fatalities and injuries.

It's not that the cars have to drive those miles, it's that they have to do so without intervention from the safety driver.

What I find grimly amusing about you post is that AV's were once hailed as reducing road deaths to miniscule proportions, but we don't hear that any more for some reason, instead the argument is now what the fluck if a few people are killed by big tech's toys?

 

[Scribus]

Oh, and who is going to pay for these 10,000 test cars, the 10,000 safety drivers and the infrastructure to maintain this 10,000 strong fleet?

 

[Chris F]

May I suggest you read the report I linked to where they answer your very point in the preamble -

Given that current traffic fatalities and injuries are rare events compared with vehicle miles traveled, we show that fully autonomous vehicles would have to be driven hundreds of millions of miles and sometimes hundreds of billions of miles to demonstrate their safety in terms of fatalities and injuries.

It's not that the cars have to drive those miles, it's that they have to do so without intervention from the safety driver.

What I find grimly amusing about you post is that AV's were once hailed as reducing road deaths to miniscule proportions, but we don't hear that any more for some reason, instead the argument is now what the fluck if a few people are killed by big tech's toys?

I didn't say that at all. I said when has big business ever cared about a few deaths.

We have a death every 200 million miles driven in the UK. I wouldn't call that rare - in fact i'd bet that AV's have already done more than that and in their infancy have only killed 3 people?

Also I doubt we will see unmanned AVs on the road. After all the London undergound has been capable of being autonomous for years now and it still has a driven.

AV tech though has the ability to make the roads safer - half the drivers I see on the road could easily be replaced by an AV!

 

[Chris F]

Oh, and who is going to pay for these 10,000 test cars, the 10,000 safety drivers and the infrastructure to maintain this 10,000 strong fleet?

Have you driven a tesla?

 

[Scribus]

Have you driven a tesla?

What on earth has that got to do with it?