Australian roboticists from the Queensland University of Technology have developed a prototype agricultural robot that uses machine vision to identify both weed and crop plants before either uprooting or poisoning the weeds or applying fertiliser to the crop.
The machine is a wide platform designed to straddle a strip of the field upon which it is working, with electric wheel motors for propulsion. It is solar-powered, and it is envisaged that a farm could have several of them continuously at work.
At a superficial level there is nothing new in the robot, its propulsion, or even the plant husbandry and weeding equipment. The really clever technology lies in the identification and classification of the plants it will encounter. It is on the success or failure of this in real farm environments that the robot’s future will hinge. The university’s next step will be to take it on-farm, and the ABC report linked above has a wonderfully pithy quote from a farmer on the subject. You can see the machine in action in the video below the break.
Farming robots have a significant following among the hardware hacker community, but it is possible that the machine-vision and plant-identifying abilities of this one would be beyond most hackers. However it is still an interesting project to watch, marking as it does a determined attempt to take the robot out of the lab and into real farm settings.
It’s that time of year again. The nights are getting longer and the leaves are turning. The crisp fall air makes one’s thoughts turn to BattleBots: pumpkin-skinned BattleBots.
Kids these days can’t even draw without a computer
If you’re asking yourself, “could a laser-cut plywood bot, sheathed in a pumpkin, stand up against an all-metal monster”, you haven’t seen BattleBots before. Besides the hilarious footage (see video embedded below), a lot of the build is documented, from making a CAD model of a pumpkin to laser-cutting the frame, to “testing” the bot just minutes before the competition. (That has to be a good idea!)
The footage of the pumpkinbot’s rival, Chomp, is equally cool. We love that the hammer weapon is accelerated so quickly that Chomp actually lifts in the air, just as Newton would have predicted. We’re not sure if the fire weapon is good for anything but show, and facing plywood pumpkinbots, but we love the effect.
What’s going through the mind of those your autonomous vacuum cleaning robots as they traverse a room? There are different ways to find out such as covering the floor with dirt and seeing what remains afterwards (a less desirable approach) or mounting an LED to the top and taking a long exposure photo. [Saulius] decided to do it by videoing his robot with a fisheye lens from near the ceiling and then making a heatmap of the result. Not being satisfied with just a finished photo, he made a video showing the path taken as the room is being traversed, giving us a glimpse of the algorithm itself.
Looking down on the room and robot
The robot he used was the Vorwerk VR200 which he’d borrowed for testing. In preparation he cleared the room and strategically placed a few obstacles, some of which he knew the robot wouldn’t get between. He started the camera and let the robot do its thing. The resulting video file was then loaded into some quickly written Python code that uses the OpenCV library to do background subtraction, normalizing, grayscaling, and then heatmapping. The individual frames were then rendered into an animated gif and the video which you can see below.
Building a simple robot arm is a lot more straightforward than it used to be. If you have a laser cutter, or a bit of cash and don’t mind waiting for postage, there are inexpensive kits like the MeArm. If you have a 3D printer, there are any number of 3D-printed designs for you to tackle. What if you need to satisfy your urge to build a robot arm really quickly, and you don’t have a laser cutter or 3D printer? You’ve got a pile of servos from that remote-control project, how can you make the rest?
If you are [roboteurs], you raid the stationery cupboard, and create an arm using rubber bands, paper clips, and binder clips. The binder clips grip the servo arms and hold the whole thing together, the rubber bands provide extra attachment , and the paper clips are bent to form the jaws. It’s not the prettiest or perhaps the most capable of arms, but it undeniably is an arm, and we’d doubt it could be done any more cheaply.
In this particular case, the arm serves as a demonstration piece for [roboteurs]’ Printabots Maker Kit for people without a 3D printer. It uses their controller board, but there is no reason why it could not be used with any other board capable of driving servos.
We’ve covered innumerable robot arms over the years, This one may be the cheapest, but another contender might be this cardboard arm. None of them, however, are as cool as this steam-powered Armatron toy.
Children can do lots of things that robots and computers have trouble with. Climbing stairs, for example, is a tough thing for a robot. Recognizing objects is another area where humans are generally much better than robots. Kids can recognize blocks, shapes, colors, and extrapolate combinations and transformations.
Google’s open-source TensorFlow software can help. It is a machine learning system used in Google’s own speech recognition, search, and other products. It is also used in quite a few non-Google projects. [Lukas Biewald] recently built a robot around some stock pieces (including a Raspberry Pi) and enlisted TensorFlow to allow the robot to recognize objects. You can see a video of the device, below.
[Nurgak] shows how one can use some of the great robotic tools out there to simulate a robot before you even build it. To drive this point home he builds the tutorial off of the easily 3D printable and buildable Robopoly platform.
The robot runs on Robot Operating System at its core. ROS is interesting because of its decentralized and input/output agnostic messaging system. For example, if you leave everything alone but swap out the motor output from actual motors to a simulator, you can see how the robot would respond to any arbitrary input.
[Nurgak] uses another piece of software called V-REP to demonstrate this. V-REP is a simulation suite for robotics and has a few ROS nodes built in. So in order to make a simulated line-following robot, [Nurgak] tells V-REP to send a simulated camera image to the decision making node of the robot in ROS. It then sends the movement messages back to V-REP which drives the pretend robot around.
He runs through a few more examples, proving that it’s entirely possible to become if not a roboticist, at least a really good AI programmer without ever dropping the big money on parts to build a robot.
At my university, we were all forced to take a class called Engineering 101. Weirdly, we could take it at any point in our careers at the school. So I put it off for more interesting classes until I was forced to take it in one of my final years. It was a mess of a class and never quite seemed to build up to a theme or a message. However, every third class or so they’d dredge up a veritable fossil from their ranks of graduates. These greybeards would sit at the front of the class and tell us about incredible things. It was worth the other two days of nondescript rambling by whichever engineering professor drew the short straw for one of their TAs.
The patent drawing.
One greybeard in particular had a long career in America’s unending string of, “Build cool stuff to help us make bad guys more deader,” projects. He worked on stealth boats, airplanes with wings that flex, and all sorts of incredibly cool stuff. I forgot about the details of those, but the one that stuck with me was the Cyclocrane. It had a ton of issues, and as the final verdict from a DARPA higher-up with a military rank was that it, “looked dumb as shit” (or so the greybeard informed us).
A Cyclo-What?
The Cyclocrane was a hybrid airship. Part aerodynamic and part aerostatic, or more simply put, a big balloon with an airplane glued on. Airships are great because they have a constant static lift, in nearly all cases this is buoyancy from a gas that is lighter than air. The ship doesn’t “weigh” anything, so the only energy that needs to be expended is the energy needed to move it through the air to wherever it needs to go. Airplanes are also great, but need to spend fuel to lift themselves off the ground as well as point in the right direction. Helicopters are cool because they make so much noise that the earth can’t stand to be near them, providing lift. Now, there’s a huge list of pros and cons for each and there’s certainly a reason we use airplanes and not dirigibles for most tasks. The Cyclocrane was designed to fit an interesting use case somewhere in the middle.
In the logging industry they often use helicopters to lift machinery in and out of remote areas. However, lifting two tons with a helicopter is not the most efficient way to go about it. Airplanes are way more efficient but there’s an obvious problem with that. They only reach their peak efficiency at the speed and direction for which their various aerodynamic surfaces have been tuned. Also worth noting that they’re fairly bad at hovering. It’s really hard to lift a basket of chainsaws out of the woods safely when the vehicle doing it is moving at 120mph.
The cyclocrane wanted all the efficiency of a dirigible with the maneuverability of a helicopter. It wanted to be able to use the effective lifting design of an airplane wing too. It wanted to have and eat three cakes. It nearly did.
A Spinning Balloon with Wings
Four wings stick out of a rotating balloon. The balloon provides half of the aerostatic lift needed to hold the plane and the cargo up in the air. The weight is tied to the static ends of the balloon and hang via cables below the construction. The clever part is the four equidistant wings sticking out at right angles from the center of the ship. At the tip of each wing is a construction made up of a propellor and a second wing. Using this array of aerofoils and engines it was possible for the cyclocrane to spin its core at 13 revolutions per minute. This produced an airspeed of 60 mph for the wings. Which resulted in a ton of lift when the wings were angled back and forth in a cyclical pattern. All the while, the ship remaining perfectly stationary.
Now the ship had lots of problems. It was too heavy. It needed bigger engines. It was slow. It looked goofy. It didn’t like strong winds. The biggest problem was a lack of funding. It’s possible that the cyclocrane could have changed a few industries if its designers had been able to keep testing it. In the end it had a mere seven hours of flying time logged with its only commercial contract before the money was gone.
However! There may be some opportunity for hackers here. If you want to make the quadcopter nerds feel a slight sting of jealousy, a cyclocrane is the project for you. A heavy lift robot that’s potentially more efficient than a balloon with fans on it is pretty neat. T2here’s a bit of reverse engineering to be done before a true performance statement can be made. If nothing else. It’s just a cool piece of aerospace history that reminds us of the comforting fact that we haven’t even come close to inventing it all yet.
If you’d like to learn more there’s a ton of information and pictures on one of the engineer’s website. Naturally wikipedia has a bit to say. There’s also decent documentary on youtube, viewable below.
