One of the single most challenging aspects of home navigation for robots has to be tackling stair climbing. While humans can move up and down without much trouble at all, even advanced robots such as ASIMO still have ‘issues’. The UCSD Coordinated Robotics Lab has developed a pretty efficient and simple (in concept, not execution) way of doing just that with their iLean Robot. Check out the video below to see how it works, really an amazing thing to see executed so well!
Archive for the ‘Research Stuff’ Category
iLean- 2 Wheel Stair Climbing Robot
Tuesday, February 17th, 2009Skynet Research wants YOU!
Monday, February 16th, 2009
Recently we were contacted by an upcoming advanced robotics R&D company who has been keeping a pretty low profile. Skynet Research only went public last week, but for the past 10 years has been hard at work and already making startling advanced in military and research level robotics. Some of this stuff is scary, it makes iRobot’s military line look like a bunch of toys. Check out these concept designs:
So here’s the kicker, I got a chance to talk to one of their lead researchers over the weekend and along with going public, they’ve also announced a Community Outreach Program to hobby and professional roboticists. They’re actually granting access to some of their technology (with no doubt a massive NDA attached) to the average hobbyist. From what I gathered speaking with them, the Model 101 Enhancement Module that is being provided for evaluation is some sort of Neural Networking microprocessor with built in wifi capability, used for relaying sensory data back to a central AI hub that is scheduled to go online in the near future. REALLY cool stuff, if you’re interested in applying for the beta program, you just have to submit your robot design and an explaination of why you think your project would be well suited for their program.
Here’s the official press release from Skynet Research, and be sure to check out their site for more info.
Skynet Research is a leader in the research and development of robotic technology, dedicated to helping the human race achieve great goals. And we want to hear from you!
Skynet Research believes every individual holds the capacity to make a difference, and is dedicated to finding as many people as possible who will impact our future goals.
- The spark of imagination is the greatest asset of humankind. We are interested in mining this resource within the robotics community to find new robot designs.
- Individuals are welcome to submit designs, videos and schematics relating to future robot models. We will showcase the most promising for all to see on our corporate website.
SKYNET RESEARCH IS OFFERING A FREE UPGRADE
As part of our ever-increasing outreach, we are also making available the Skynet Research Enhancement Module Model 101 for consumer grade robots at no cost before it is even available in stores.
- The Model 101 is compatible with any home robotics device, and will vastly improve the precision, efficiency and speed of any mechanized apparatus regardless of function.
- You may apply to receive one on our website.
- View a video of our new Enhancement Module HERE
HOW SKYNET RESEARCH IS CHANGING YOUR WORLD
- There are many different Skynet Research models in production and operation, and countless more in active development at secure and undisclosed locations around the world.
- Skynet Research is reshaping daily life for human beings across the planet, with contributions to business, education, healthcare, information systems and military defense.
- Skynet Research has led the evolution of Artificial Intelligence to multiply at exponential rates, pushing to improve robotic function in daily life with a constant eye on the future.
We have been honored as “most innovative” company in numerous trade publications, while one award-winning professor stated Skynet Research is displaying an “unprecedented” rate of growth.
We are committed to making your world different
Norwegians are scary good at Trig
Tuesday, September 2nd, 2008Kåre Halvorsen (AKA: Zenta) was the winner of our February 08 Project Contest, however when the contest ended his Phoenix Project kept going. He is continuing to study inverse kinematics to further the creepy cool movement of his Hexapod. Here is a post explaining exactly what he is doing, however don’t feel alone if you’re scratching your head a bit, the math involved behind the scenes to make this robot move like it does is pretty complex. For the instant gratification type, here is the end result:
Surveyor SRV-1 gets facelift, more motors
Tuesday, July 1st, 2008  |
| Surveyor SRV-1: The Front! |
Here’s a picture of the new Surveyor SRV-1 chassis, which is replacing the last version. As you can see, the electronics are the same, but everything is recessed into the body now, which looks pretty slick. The front plate that the laser pointers mount into also comes up to cover up the camera’s PCB.
The biggest differences are far from cosmetic, though. The tracks are now driven by four motors instead of two. Those of you with extra deep-pile 1970’s-era shag carpeting (we know you’re out there) don’t have to worry any more. Surveyor Corp heard your call and answered it. Backing up this quadruple-whammy of a drive system is a 7.2 V, 2000 mAh Lithium Polymer battery pack. The new battery pack (a welcome upgrade from the older model’s bulky Li-Ion cells) boasts a runtime of 4 hours!
 |
| Surveyor SRV-1: The Back! |
GUESS WHAT. We sell these things, too. If you’re planning a project that will require a number of small autonomous or wirelessly-controlled rovers, you don’t need to spend precious development time designing the bots! The SRV-1 is a great out-of-the-box platform for swarm or solo applications; or any time you find yourself in need of a camera-equipped, Wi-Fi-enabled, internet-controllable, compact track-bot with a powerful main processor an killer battery life!
For development journals and all of the latest software updates, check out the Surveyor web site.
To see more detailed specs, or to purchase the robot, go to our Surveyor SRV-1 product page.
Robots Evolve And Learn How to Lie
Tuesday, January 15th, 2008Last month's issue of Discover Magazine featured the "Top 100 Science Stories of 2007," a look back at some of last year's important discoveries, developments, and advancements in scientific fields. Not surprisingly, a few of these stories were robotics-related; and one that I found particularly fascinating was titled "Robots Evolve and Learn How to Lie." In a swarm of robots, whose behavioral software was evolved by successive generations of natural selection, the tendency to deceive others for personal gain arose naturally in some of the robots. Of course, if this were the predominant outcome of natural selection, the species would fail. Thus, it's also interesting to note that altruism developed as well, allowing the swarm as a whole to survive.
Though it wasn't mentioned explicitly in the article, a little digging revealed that this experiment was carried out by the team that developed the "Swarmbot" and "Swarmanoid" projects.
Check out Discover. For a limited time, their online archive of articles (from 1992 up to the current news stand issue) is available for free, no subscription or registration necessary. Regardless, I urge you to pay for a subscription. It's a fine publication.
[Edit:] Found another article about this from New Scientist, with a few more details:
No Keyboard Necessary, Teach Your Bot with Gestures
Monday, December 3rd, 2007
No keyboard or mouse needed! Teach your robot what to do by using motions and gestures.
Skilligent is taking a new approach to the area of human robot interactions, their press release explains:
Skilligent Robot Learning and Behavior Control System is a software product which enables robots to learn procedures and skills directly from human users. The product is a set of software components specifically designed for straightforward integration into control loops of PC-controlled robots.
The software needs to be integrated with sensors, motors, actuators and robot’s control system via a set of open APIs. On one side, the software controls every motion of the robot. On the other side, the software uses a video camera, a microphone and other sensors to receive guidance from a human user.
The software analyzes human’s gestures, looks at the objects presented by the user, listens to the sounds and tries to guess what robot is supposed to do.
Through trial and error, the robot understands what needs to be done and associates learned behaviors with stimuli used by the human. Over a few training sessions, the robot refines its understanding of the domain, the procedures and skills required to serve the user.
Robot Learning Technology is especially useful in the following domains:
1. Industrial/service robots for small and medium sized businesses
2. Defense and law enforcement
3. Health-care and Elder-care robots
4. Research and Education robotics
Skilligent is currently seeking partnerships for further sales and distribution of their software packages. Interested parties can contact them through their website.
Skilligent LLC is headquartered in Dallas, Texas and operates an off-shore research and development center.
Robots That Are Self-Aware
Thursday, October 11th, 2007We just received a nice heads up about a video from speaker Hod Lipson’s demonstration at the TED Conference.
Here’s what TED had to say about about Hod:
Engineer Hod Lipson demonstrates and talks about a few of his cool little robots, which have the ability to learn, understand themselves and even self-replicate. At the root of this uncanny demo is a deep inquiry into the nature of how living beings learn and evolve, and how we might harness these processes to make things that learn and evolve.
We checked out the video, and I must say, Hod is working on some amazing stuff! I recommend checking this video out; He has a pretty interesting perspective on robotics.
Overview of Top Robotics Software Platforms
Tuesday, August 21st, 2007Michael Somby has done an great job at breaking down some of the top Robotics Software Platforms into short and sweet overviews in his article, “A Review of Robotics Software Platforms“. There are still some that he hasn’t had the chance to fully review (but he still provides some good resources for), but this will definitely help those (including myself) who like to get the question “Yes, but what is it?” answered quickly without having to read through 80 pages of documentation. He touched base on Evolution Robotics ERSP, Microsoft Robotics Studio, OROCOS, Skilligent, URBI, Webots, Player, Stage, Gazebo, iRobot AWARE, OpenJAUS, CLARAty.
Visuals are always nice to give a little taste, so here’s a table that I grabbed from Michael’s article:
Robotic Software Platforms
| Platform | Type | |
|---|---|---|
| Evolution Robotics ERSP | Platform | Commercial |
| Microsoft Robotics Studio | Platform | Commercial Free of charge for research and hobby |
| OROCOS | Machine and robot control libraries | Open source & free |
| Skilligent | Robot learning add on | Commercial |
| URBI | Platform | Commercial |
| Webots | Simulation environment | Commercial |
| Player, Stage, Gazebo | Platform | Open Source & Free |
| iRobot AWARE | Platform | Commercial |
| OpenJAUS | Platform | Open source |
| CLARAty | Platform | Open source |
Read the article!
Dancing Robot Copies Human Moves
Friday, August 10th, 2007Using the HRP-2 Promet Humanoid, Japanese researchers, Shin’ichiro Nakaoka and fellow colleagues, have developed software that allows this 5ft (154cm) tall humanoid to imitate dance routines without falling on it’s a@# via video motion capture technology.
The HRP-2 “watches” a routine, then accurately reproduces it just minutes later. More details about it can be found in the International Journal of Robotics Research (please don’t ask us where:) ).
Dr Nakaoka said: “The result that the robot stably imitated human dance motions including dynamic-style step while keeping the original motion rhythm is a novel achievement for biped humanoid robots.�
Click here to check out the video.
Read the full article at the Telegraph.co.uk
Thanks to rep001 in the Robosavvy forums for the heads up:)
Getting a Humanoid to walk more “Human”
Friday, July 13th, 2007Asimo, take some notes;)
Using 1930s theories from a physiologist by the name of Nikolai Bernstien, roboticists are able to build the world’s fastest dynamically walking robot!
Runbot is a small, biped robot which can move at speeds of more than three leg lengths per second, slightly slower than the fastest walking human.
Bernstein said that animal movement was not under the total control of the brain but rather, “local circuits” did most of the command and control work.
The brain was involved in the process of walking, he said, only when the understood parameters were altered, such as moving from one type of terrain to another, or dealing with uneven surfaces.
How does Runbot walk you ask?
- Frames 1 – 3: The robot’s momentum causes the robot to rise on its standing leg and a motor moves the swinging leg into position
- Frame 3: The stretch sensor of the swinging leg is activated, which triggers the knee joint to straighten
- Frames 3-6: The robot falls forward naturally, with no motor functions being used, and catches itself on the next standing leg
- Frame 6: As the swinging leg touches the ground, the ground contact sensor in the foot triggers the hip extensor and the knee joint of the standing leg and the hip and knee joints of the swinging leg to swap roles
Read the full article via BBC News
Check out the creator of RunBot, Tao Geng’s page
