Older blog entries for Kiryu (starting at number 19)

Well, I am here at the University of Southern California Robotic Interaction Lab. I am building an "inexpensive" motion capture suite using inertial measurement units. Each unit is coming in at about 200 dollars, but has 6 DOF. I have designed a radio link between the suit and a host computer that will basically stream the data in. All the sensor parts are on order and the boards are being made. Each IMU is only 1.75x1.75x1" and is interfaced through a minidin connector, which also functions as the programming port. I think the design is solid and we expect that you can capture all movements of the arm, minus the fingers, using just 3 IMUs. So, that makes 6 for the arms, 6 for the legs, 1 for the head, and 1 for the upper torso = 14 in total. That is about $2800 plus another $300 for the link, puts us at about $3100, to capture the entire human body (less fingers & toes). That is a LOT better than $20,000, which is the base price of most of the motion capture systems out there. I am also sketching out the specs for what could turn into a humanoid robot interaction testbed, we will see.

The Mininaut project is now complete (as far as I am concerned :-) It got quite a bit of attention locally by appearing on several news stations and in a couple of newspapers. More importantly, I have secured a postition this summer at the University of Southern California to prototype an inexpensive motion capture suit; that should be a fun project. I think I am changing my research focus, I should be moving to enabling technologies for molecular robots. I have a proposal ready to be submitted for funding in mid May and will be working with several other people from a range of fields, including mathematics, physics, and chemistry. We plan on developing a predictive algorithm for molecular construction of robot parts. Everyone around here seems to think it is going to be a good project, I hope we finish :-) Anyway, if you are interested to know about anything that I have been or am involved with, let me know, I like to discuss these things :-) Oh, and thanks for reading this far!

3 Mar 2003 (updated 30 Apr 2003 at 22:45 UTC) »

An almost complete version of the poster I did for the mininaut project is available in powerpoint format on my webpage here. I ended up with first place :-) on to the research competition for the school and then will compete soon for all of the cal state universities. Busy times, hope all is well for everyone else.

Just to let everyone know, I have updated my website with the "Mininaut" project that I am working on. You can view it here

My robot is coming along well (the research one for the university.) I will finish the physical/electronic design this month, then I will start the programming. Not too much to report as far as progress because most of it has been done on the schematic. I did pick up a nice Sony VAIO laptop that is only 4 lbs - NICE! Check out the website that no longer serves as a home for the company I was developing, but rather a portal for robot activity.

24 Aug 2002 (updated 24 Aug 2002 at 00:36 UTC) »

If you stopped by LivingCreatures before, this is just to let you know that we have a new chat available to our members.

23 Aug 2002 (updated 23 Aug 2002 at 01:02 UTC) »

My companies website, www.LivingCreatures.com , has gone through a complete redesign. Not only does it serve as an information center for our upcoming products, but it is also a complete web portal where you can post to different forums and participate in the community base. Please check it out.

20 Aug 2002 (updated 20 Aug 2002 at 23:49 UTC) »

The humanoid robot is coming along well. The structure is carbon fiber, with a little aluminum and brass throughout. I have modified all of the servos for a 4-wire interface and have mounted them. Some are permanently fixed with epoxy, as this was the best way due to the size constraints.

In other news, my company website is up, although it is being redesigned. The tough thing with a start-up is getting everything that needs to be done - done. You can check it out and send me an email on what you think. I should have some photos of the humanoid project available soon.

Wow, it has been a while! My scaled down humanoid project is now funded by the University. Although not designed to perform its original function, it is nice to finally have someone else buying my parts for me :-) I should have a copy of pro/engineer arriving at my door in the next week, which I will use to model the parts. I picked the Axiom PB555 processor to run the robot- this is a kick butt processor, but it is not for the lightweight. I should have more regular updates happening since this project is going to rule my life for the next year.

Just letting everyone know that I am still around. The quarter is almost over and that means more time to be working on robots. I have an idea for a micro actuator based on muscle wire that would be fully proportional and act like a servo. What would be the advantage of this? A fully functional robotic human torso measured in tens of grams :-) The interesting part will be trying to make these little things. There is a company trying to sell "nanomuscles", but they want $21 each in volume, which is way more expensive than the $1 a motor would cost. The design is good and I started thinking about it when I realized that the current design for my robot is way too expensive for it to be repeatable and something everyone could enjoy making. If I was able to design the robot with a couple of atmel controllers and leave out the rabbit 2000 and use these new muscles to power it rather than the servos that I have, it would be smaller, lighter and most importantly- cheaper. Just a thought to let everyone know that I am still around :-)

10 older entries...

Share this page