After a recent article about an electric tractor, one of our readers suggested we look at autonomous tractors more, and that turned out to be a very good suggestion. There’s a whole world of autonomy going on both on farms and construction sites that tends to fall through the cracks while we tend to get focused on cars. Machine learning is going to transform not just the road, but how it’s built and how the people who drive on it eat.
Kubota lays out a pretty expansive vision with their Kubota X concept from last year. It would be cool enough if you could give a tractor one task to do at a time, but Kubota wants their futuristic tank-crab tractor to do a whole lot more. They show a farmer controlling what the tractor does while not being anywhere near the farm. It starts up, attaches itself to implements, and goes to work tending the field.
The tractor is also going to be part of a more expansive ecosystem of autonomous products. They show an autonomous drone that checks the field before the tractor goes out, giving it the go ahead. Kubota also depicts a drone checking the crops themselves to control the tractor’s fertilizer output, as well as to fine tune the watering system, which also tends to itself. They hope to optimize crops for flavors and a variety of other factors to make each farm unique and/or tailored to buyers’ wants or needs.
Kubota also hopes to help farmers find direct buyers for their crops instead of selling vegetables en masse to middlemen. Technically Kubota would be a middleman, but they seem to only be interested in arranging for people to meet and forge business relationships.
Kubota’s video focuses on farmers, but the company has to know that tractors are used in a variety of other industries. Earth moving, landscaping, vacant lot maintenance, and a variety of other people are using tractors on the daily. An autonomous tractor could do a lot for any of these industries. In many cases, a human will probably have to supervise the machine and transport it to work sites, but that may change in the future, too.
There’s no reason this technology couldn’t be scaled down for home and other small-scale users. There are already the equivalents to a Roomba for people’s lawns, but a replacement for a larger ride-on mower that can perform other large lawn tasks would be very useful. I’d imagine anyone who owns a large lawn would prefer to let the robots take care of it once the price drops to the point where it’s affordable.
Construction and Mining
Farming is only the tip of the proverbial iceberg for what autonomous large equipment will be doing.
At first glance, it might seem like off-road uses of autonomous vehicles would be easier. After all, there aren’t traffic lights, other drivers, and idiotic pedestrians who dart into the road or jaywalk. On the other hand, construction tasks are equally complex in their own way.
Like many autonomous vehicle programs, Caterpillar’s expertise started to develop during the DARPA AV program, but unlike the people who went on to work on street programs, Caterpillar focused on work. Now, it’s starting to pay off. They’re offering mining trucks that can pilot themselves not only around the job site, but to perform specific tasks that have been carried out by drivers in the past. They stop in just the right spot to get loaded, and then drive across the site to places where they are supposed to dump the load.
I’m no mining driver, but I could imagine that driving back and forth all day could get tedious and even sleep-inducing. Yes, the Australian mining site depicted in the video now uses six fewer human drivers, but they get a lot more work done, freeing up humans to perform other tasks that aren’t yet automated.
In construction, autonomy has been slowly creeping into the industry. On jobsites I’ve done drone work for, I’ve been starting to see a few little gadgets working without a driver. They still have a human controller, but small dirt compacting rollers are becoming more common.
Tasks that require precision are already seeing a lot of assist features. Graders (aka “blades”) used to be manually set by the driver, and required a lot of adjustment and fix-up to get things precise. Now, the blades have an RTK GPS unit that’s accurate down to under a centimeter. The blade dynamically adjusts itself, and can even work slopes without losing accuracy. Caterpillar points out that the same technology can help buckets and dump trucks all stay stable, safe, and accurate.
Getting the job done near-perfectly on the first try isn’t the only benefit to these assist technologies. By adding cameras, collision warnings, and systems to keep drivers from tipping expensive equipment over, even new and inexperienced operators can operate large machines with far lower risk. Safety is a big benefit, but it also helps the job get done at all when experienced operators can’t be found.
Obviously this is a stepping stone toward full autonomy, but it gives companies, the public, and workers key benefits today.
Like all industries, construction has a lot of “we’ve always done it this way” in it. Resistance to change is a factor, but sometimes people just want to keep doing what they’ve always done because they know it works and don’t have extra time for a learning curve getting autonomous vehicles going on the job. As Holly Welles at Constructconnect points out, many construction companies rent machinery for short jobs or to evaluate something before putting it into play.
Instead of having to invest big bucks in a new and seemingly unproven bit of equipment, they can use it on a trial basis. This will give many construction companies the opportunity to use them for better cost savings and get lower bids in on key jobs. Nobody cares about tradition when it’s bid day, and whoever gets the lowest price without cutting corners gets the job.
Because the economy values efficiency, we will probably see the construction industry, mining, and farming adopt autonomous vehicles faster than the personally-owned vehicle industry.
Don't want to miss a cleantech story? Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!
Have a tip for CleanTechnica, want to advertise, or want to suggest a guest for our CleanTech Talk podcast? Contact us here.