Skip to content. | Skip to navigation


Personal tools
You are here: Home / Blog / Is Autonomy The End Of The Road For The Truck Industry? Don't Bet On It!

Is Autonomy The End Of The Road For The Truck Industry? Don't Bet On It!

Every so often, a random commentator will proclaim the end of the truck industry as we know it. Some new technology will come along that will allow us to transport goods from A to B faster and cheaper.

Haven’t We Been Here Before? 

You might think that automation is the only threat that the industry has ever faced, but we’ve been here before. Throughout much of the 20th century, the rail network was supposedly going to take up the slack, putting one of America’s biggest industries to bed. 

Pixabay - CC0 License

But of course, that never happened. Despite all the promises of higher efficiency and lower unit costs, rail freight failed to replace its road-based counterparts. In most countries, upwards of 90 percent of all transportation occurs on highways. 

The reasons for this are long and complicated and take the expertise of an economic historian to explain fully. But the gist of it has to do with flexibility. 

Rail makes it very cheap to transport large consignments of goods, like coal, for a powerplant. But it is hopelessly impractical for modern warehouses. You can’t simply build your facilities next to the train track and help yourself to cargo as it passes by. It doesn’t work like that. 

Then there’s the problem that most of your suppliers don’t integrate with the rail network anyway. So even though it is cheaper to transport goods that way, it is so impractical to make it all but impossible for most enterprises. 

Automation And The Truck Industry

The latest idea is that automation will sweep away the truck industry as we know it, replacing it with wheeled drones that will zip across the country without driver assistance. They will reduce truck accidents, “free up” labor and improve overall efficiency. 

But whether computers can replace human drivers remains a matter of contention, even today, many years into the technology. It is nearly fifteen years since the University of Carnegie Mellon first demonstrated the concept. And yet, nobody is ready with a product that can actually navigate the roads. Not even Google. Not even Tesla. 

Pixabay - CC0 License

There are actually good philosophical reasons to suspect that current technological paradigms will not be able to solve autonomy. The main problem is that machines don’t understand context - only statistics. Thus, they can see something completely new and unexpected and treat it as if it is just another version of something else. They can’t connect ideas. They don’t have common sense. They’re just identifying objects and calculating probabilities, without any understanding or intuition. A self-driving car has no idea that that is what it is. 

At the moment, there doesn’t seem like there is a path for giving vehicles a human-like subjective experience. Nobody has a clue how to do that. Yet it may be necessary for fleets of driverless cars to get off the ground.

While this view might sound pessimistic, it is not. Nobody is ruling out a paradigm shift that will eventually make widespread autonomy possible. But it just hasn’t happened yet. And it probably won’t for a while.

Add comment

You can add a comment by filling out the form below. Plain text formatting. Comments are moderated.