Magnetic Clevis

Since 2 years, i’m playing around with natural permanent magnet. I’m really surprise that nobody is using this natural force as it should. There is a lot of new way to explore.

Here is a little concept of of magnetic articulation driven by 3 or 4 little electric engines.  The main goal is to let natural magnets hold the biggest part of the weight and use a very little energy to drive it.  It could be use as a stepper engine except that the angle could be setup very precisely. The strength of the joint could be modify without adding extra electric energy. Also the velocity factor could be very interesting. There is no mechanical contact between the 2 heaviest parts so no sound, no heat and no usury. Only the driven part: the electric motors, the driving screws and the spherical bearing will!

As i saw in many industrial manufacturer since few years, the electromagnetic bearing are now very in use. (ex: SKF)

Here is a very simple concept of natural magnet bearing:

This is an aluminium shaft with box magnet all around at 45 degrees. On each sides, i put a ring magnet in the repulsive side.  Of course, it’s a schema, so i let the rings floating in space to keep the pictures readable. So more the ring are closer from each other, more there is pressure over the whole shaft.

After that, my question was how to made this turn.  As i was studying chopper’s swashplate few years ago, i fund a way to adapt this mechanical method to my concept. Simply by creating a disharmony between the rotor’s box magnets strength like this:

So by tilting a little the rings in symmetry over it, this will give a direction to the shaft. Now the question is how to make the rings turn in symmetry without using a lot of energy? For this i setup a set of 3 driving screws to hold the rings magnet with spherical bearing. The 2 shaft’s strokes are set in an opposite stepping, that way, when i turn the shaft there are getting closer each other, or farest in opposite rotation.

Magnetic Clevis from Lempereur Bertrand on Vimeo.

Here a little schema of the Final concept:

This is with 3 shaft but it could work also with 4.

If you want to experiment some stuff with magnets:

http://www.supermagnete.fr/eng/

http://www.hkcm.de/

For industrial stuff:

http://shop.hpceurope.com/an/home_catalogue.asp

Constant Velocity Joint

NLP and Ontologies

For a while now I’ve been thinking about using knowledge representation – Ontologies as a base for creating a modular Natural Language Processing system focused on extracting structured data from unstructured. For example we can create/use Ontologies (models) that describe “simple” concepts like: Address, Time, Task, Expense, Transaction etc… and use them to “match” information from a text stream. The reason i’m writing this is because i think that there is a common ground for collaboration… I know Stefan is interested in RDF/OWL,  the company that Neven is involved is in a very near domain and finally i was playing with Google Wave which i think is a good platform for creating intelligent bots that will be very easy to distribute if they turn out to be useful :)

Here are some references:
http://wordnet.princeton.edu/
http://protege.stanford.edu/
http://jena.sourceforge.net/
http://www.openrdf.org/
http://code.google.com/apis/wave/guide.html

Swarm Robotics

Here is an article plus a lot of videos about Swarm Robotics

About robots and what other people are doing

I have list of websites about robots and robotics that are very interesting. I did this research some years ago when I had more free time to surf the net, but I will probably get back to the subject of robots and robotics soon.

I am personally most interested in micro-robots which are considered those that are in size less that 1 cubic inch. The Ants from MIT (http://groups.csail.mit.edu/lbr/ants/) is very good example.

The Swiss EPFL institute had done great deal of research and work in that area. Not all the pages are currently active, but they have archives of most of their projects. Here are some links:

• http://diwww.epfl.ch/lami/ – The LAMI (Microprocessor Systems Lab) – the founder is now running this business: http://www.didel.com/.
• http://asl.epfl.ch/epfl/ – ASL/LSA Autonomous Systems Lab.

Here: http://nanolab.me.cmu.edu/projects/waterrunner/ – a project from Carnegie Mellon University – Water Runner Robot.

Of course Linux has its share in the business too: http://www.linuxfordevices.com/c/a/News/Open-source-robotic-simulation-packages/ – “Open source robot simulation software simplifies the creation of physical …”. IBM are in the business too – http://www.ibm.com/developerworks/linux/library/l-robotools/.

For  “iFanatics” there is that silly walking  iPhone robot (http://robots.net/article/2670.html) create by Japanese eningeer Kazu Terasaki.

If you want to start practice here you can find some gear:

• http://www.acroname.com/
• http://www.robot-electronics.co.uk/
• http://www.active-robots.com/
• http://www.allelectronics.com/
• http://www.newmicros.com/

… and some tutorials:

• http://www.cs.uiowa.edu/~jones/step/index.html

Other links that you may fond interesting:

• http://robots.net/
• http://www.solarbotics.com/

Open source robot simulation software simplifies the creation of physical Linux-based robots, according to an overview article at IBM’s DeveloperWorks site. Simulators such as ODE, Simbad, TeamBots, KControl, Gazebo, and Carmen can help test ideas before putting them into hardware, writes author M.

And now, the most ridiculous robotics project I’ve ever seen: Deep Green, featured today on TechCrunch.

Here’s an interesting robotics project: iCub, the Toddler Robot. Go check it out!

Lets make robots

I’ve recently stumbled on a site about making robots http://letsmakerobots.com/. There some good videos and a lot of info about parts. I like the PING))) Ultrasonic Sensor for detecting obstacles and distance. For me the ultimate goal would be to make a robot that knows its position and orientation in my apartment and can create a map in its memory that can be used to navigate itself without bumping in any obstacles. To do this i may have to use a more sophisticated board than http://arduino.cc/.  My first step is to position 3 sound sources with different frequencies in my apartment and record the combined signal,  calculate the position of the mic based on the speed of sound using Matlab etc… Easy setup difficult calculations