[Ladvien] has figured an inexpensive way to control a robot from a remote PC with a static webcam. Inspired by swarming robot videos such as those from the UPENN Grasp lab, [Ladvien] wanted to build his own static camera based system. He’s also managed to create one of the more eclectic Instructables we’ve seen. You don’t often find pseudo code for robot suicide mixed in with the project instructions.
Fixed cameras are used in many motion capture systems, such as the Vicon system used by numerous film, game, and animation studios. Vicon and similar systems cost tens of thousands of dollars. This was a bit outside [Ladvien’s] budget. He set about building his own system from scratch. The first step was the hardest – obtaining permission from his wife to screw a webcam into the ceiling. With that problem overcome, [Ladvien] brought openCV and python to bear. He created Overlord, his webcam vision and control system. A vision system with nothing to control would be rather boring, so [Ladvien] created DotMuncher, Overlord’s radio controlled robot slave.
The basic processing system is rather simple. DotMuncher carries a magnetometer on board, which it uses to send heading information to Overlord. Overlord is pre-calibrated with an offset from magnetic north to “video game north” (toward the top of the screen). Overlord then uses openCV’s color detection to find DotMuncher in the current scene.
Overlord finally generates a virtual “Dot” on screen, and directs DotMuncher to drive over to it. When the robot gets to the dot, it is considered munched, and a new dot is generated.
Continue reading “Inexpensive Robot Tracking System is Swarm Ready”
The screen capture above shows a group of swarm robots working together to move the blue box from the left side of the frame over to the right. It’s just one of many demonstrations shown in the video clip after the break. The clip is a quick sampling of the many swarm robotics research projects going on at the University of Sheffield’s Natural Robotics Lab.
The main focus for all of the research is to see what can be accomplished by getting a large group of relatively simple machines to work together. Each device has a microcontroller brain, camera, accelerometer, proximity sensors, and a microphone. By mixing and matching the use of available components they can test different concepts which will be useful in creating utility robot swarms for real-world tasks. The video shows off the robots grouping themselves by like characteristic, a test called segregation (the purpose of this didn’t resonate with us), and group tasks like moving that box. The nice thing is that a series of white papers is available at the post linked above (click on the PDF icon) so that you may dig deeper if these projects are of interest to you.
Continue reading “Treasure trove of swarm robotics research”
Over the years bicycle design has changed. Materials were upgraded as technology advanced, and accumulated knowledge helped bicycle builders make improvements along the way. But deep analysis with the intent to make meaningful improvements has not been widely embraced. Reasearchers at UC Davis are looking to expand into this frontier by letting the bicycle tell us how it can be improved. This is one of the test bikes they’ve been working on, which is mainly aimed at data harvesting. They’re hoping to find some real improvements based mostly on how the machine can get out of the rider’s way as much as possible. The thought here is that the rider’s body makes up 80-90% of the volume of the vehicle and should be accommodated in every way possible.
Sure, this could be a case of trying to build a better mouse trap. But listening to the discussion in the video after the break really drives home the complex issues of stability and locomotion that go into these seemingly simple vehicles. We’re going to guess the final recommendations will not involve making the bike five times taller.
Continue reading “The bicycle can tell us how to make it better”
NC state’s constructed facilities laboratory is a place where things get broken for science. We’ve shared several videos lately of things being sliced, diced, sheared, exploded, and smashed, purely for the fun of it, and now we feel like we should compensate a little bit. No, we’re not going to undergo physical punishment, instead, we’ll share some educational destruction.
In the video after the break, you can see a few things pushed to their absolute limits, then a bit further. The Constructed Facilities Laboratory is a research lab that tests the limits of some of the infrastructure that we rely on daily. Bridges, roads, walls, support beams. Someone needs to figure out what they can really handle. Even more interesting than the short video below, are all the different videos in the tour that explain how the facility is constructed an how they operate. Take a few minutes and enjoy the tour.
Continue reading “Destroying stuff for the good of all mankind”
Modern society owes so much to medical research, though what happens behind the scenes in a laboratory is usually far less than glamorous. A group of scientists at the University of Cambridge are working to develop synthetic bone tissue, but the process to create the samples used in the study is incredibly tedious.
To make the bones, a substructure must be dipped in a mixture of calcium and protein, rinsed, then dipped in a mixture of phosphate and protein…hundreds and hundreds of times. Equipment that can automate the process is available but very cost prohibitive, so the scientists did what they do best and built a set of robots to do the work for them.
Their new bone manufacturing setup was constructed using Lego Mindstorm kits, which were a perfect solution to their problem in several ways. The kits are relatively cheap, easy to construct, easy to program, and able to perform the same function precisely for days on end.
Now instead of burning time manually creating synthetic bone samples, the group can focus on the more important facets of their research.
Continue reading to see a video presented at the 2012 Google Science Fair, showing how everything came together for the crew at Cambridge.
Continue reading “Lego Mindstorms used to automate tedious laboratory tasks”
We’re taking a field trip from the backyard, garage, and basement hacking in order to look in on what research scientists are up to these days. A group from the Johns Hopkins Institute for NanoBioTechnology has been manufacturing quantum dots for use in the medical field. Made up of Cadmium Selenide, this is a nanomaterial that you can think of as individual crystals of the smallest size possible. Quantum dots have many uses. Here, [Charli Dvoracek] takes the recently manufactured dots and activates them with antibodies capable of targeting cancer cells. Once mixed with a biological sample, the dots embed themselves in the walls of the cancer, allowing the researchers to find those cells thanks to the phosphorescent properties of the dots.
The video after the breaks walks us through the various steps involved in growing these dots. [Charli] has the benefit of a fully outfitted lab, using tools like an argon-filled glove box to protect her from harmful off-gases. You’re not likely have this sort of thing in your home laboratory, but as we’ve seen before, you can make some of your own equipment, and produce interesting chemicals with simple processes. If you’re someone who already tinkers with chemistry experiments we want to hear about your exploits so please drop us a tip about what you’re up to.
Continue reading “Brewing up some quantum dots”
We know that you can transform the mechanical motions of your body into electrical energy, like when you turn the crank or shake a mechanically-powered flashlight. These types of mechanical motions are quite large compared to many of the day-to-day (and minute-to-minute) actions you perform–for example walking, breathing, and thumb wrestling.
What if we could harvest energy from these tiny movements? Researchers at the Korea Advanced Institute of Science and Technology are seeking the answer to this question with piezoelectric barium titanate. The electrical output of their devices is very small (in the nanoAmps) but over a long period and over many repetitions it would be possible to run a small electric device–even a biologically-embedded one. An alternative to blood power?
There is clearly a lot of potential in this technology, and we’ll be interested to see if and when we can start messing around with this stuff. Heck, it’s already been used to power a small LED and you all know just how much everyone would jump at the chance to cover themselves in self-powered LEDs…