The Future of Trucking

California, Arizona and Florida. Google has been testing prototype vehicles on public roads in those states for several years now. Several foreign manufacturers are testing similar designs on public roads in parts of Europe.

 

What happens to the autobots when it snows or rains.
It's pretty straight forward to program a robot for inside work and driving a car from point A to B on dry roads.
The crux of programming robots is working in adverse conditions since it requires a real person with experience to judge and respond to the varying conditions.

A good example of this would be why real world automation hasn't come to ski areas. They still need lifties to help people load a automatic lift, they still need real drivers to drive the groomers despite being done a night within a mapped area.

Automation will come but they will still need someone in the seat to drive when the going gets tough.

 

I think you can expect in the years ahead technology applied first to new safety systems--for trucks more lane control, etc. Cars will have features we haven't envisioned. This morning crossing Indy it seemed the 4-wheelers were determined to bunch up too close to react to anything. An automated lane control feature would prevent that, require a stoppable distance unless able to safely pass. A car approaching another from behind would auto-slow or pass, not close in and ride bumper.

 

I'm confident they will handle snow and slippery roads better than any human could as they're be able to sense loss of traction quicker and with more accuracy than an human could.

When I skied in France 10 years ago, there were many un-manned automatic lifts.

This is because the market for grooming machines is incredibly small. I doubt more than 100 new units are sold per year world wide. And these are machines that are generally run by a labor pool that is willing to work for nearly nothing (cost of lift tickets and a bag lunch in many cases). So the initial high cost of automation isn't justified for this application at this time. Once the price of automation comes down, the groomers will run themselves and the ski bums will have to find another way to earn their lift tickets.

 

But can they anticipate slippery conditions?. Temps close to freezing and approaching a bridge or overpass that is likely to freeze first?.

I am sure they can modulate brakes and steering better than a human, but will that make any difference when the truck is already halfway across a patch of black ice?.

 

Commercial airliners seem to be able to all but fly themselves but the plane still needs to HUMAN pilot and co-pilot. I saw something once on the Discovery channel about how one airline had installed in its planes an auto pilot system that was able to completely fly the plane with no human.

But they still had a pilot and co-pilot on the plane in case anything happened to the system. Turns out something did happen. For some unknown reason the auto pilot though something was wrong with one of the systems and the plane was out of control. Thankfully, the pilot had a button he could push for such malfunctions and he and the co-pilot were then able to take control of the plane and land it safely.

Bottom line...It may be a long while before an 18-wheeler can run down a highway with no human driver in it just in case the auto system acts up for whatever reason.

 

https://www.youtube.com/watch?v=RIe9gSRBQTk

 

Well the trucks already monitor outside air temp constantly so no miracle required there. The GPS already knows where the bridges are so no miracle required there either. The autonomous vehicles being tested today have forward looking cameras and radar along with software to analyze what it sees. It wouldn't be difficult or expensive to add a forward looking thermal camera that could detect the temperature of the road ahead. Combine that with sensor that tell the truck how wet or dry the road are and boom, you've got a truck that can anticipate where slippery roads might be quicker and with more accuracy than a seasoned pro.

 

Apples to oranges comparison. Airliners have been able to fly and land themselves for years. But they deal in 3 axis' not 2 and they are much more affected by weather than ground based vehicles. But most importantly, they cannot simply go into a fail safe mode and pull over and stop if/when a critical system fails. They have to get to a runway which is capable of handling the aircraft at its current weight and land safely on that runway with critical systems failed. And we probably could make airplanes that could do that. But every additional system is additional weight. And in aircraft design, every ounce counts. We want the trucks light but we don't bat an eyelash if they end up being a couple hundred pounds heavier.

A truck, can pull itself over to the shoulder if one of its critical systems goes offline. And if there is no shoulder, that truck can stop where it is, put the 4-ways on, and if need be, have a system that automatically warns other autonomous vehicles that there is a vehicle stopped ahead. That same system could easily be made to automatically trigger digital road signs to give manual drivers a warning that there is a disabled vehicle in the road ahead. Very easy to do with existing off the shelf technology. None of that is feasible with transport category aircraft.

 

I think this country is to litigious and that seperates us from some of the automation that goes on in Europe. It's okay to make certain comparisons between Europe and the US but I've seen many areas where it just wont work because the businesses involved fear the legal consequences when things go wrong. I've been on automated lifts skiing in Austria but those ski areas puts the responsibility to avoid danger on the skier. Here in the US ski areas have to be roped off so gumbys don't go skiing off a cliff and still need liftys to help people get on a slow moving chair.

The thing is no matter how reasonable an idea sounds and how it works on paper it doesn't mean it will work in the real world especially one that would mix professional automation with amateurs driving manually. There are lots of manufacturers out there with self parking cars and there has been a lot of resistance to just that simple feature. It's going to take a lot of trial and error and trust before people give up control of the wheel. I think once we reach a point where there is at least a generation or two raised with personal cars being fully automated will we see the commercial side start to be trusted to go fully auto.