Older blog entries for c6jones720 (starting at number 40)

Just in case any robotics guys in the Avon & Somerset region are interested, theres something going on at the HP labs.

HP SCIENCE LECTURE ANNOUNCEMENT

The next HP Science Lecture takes place as follows:

Date: Wednesday 24th January 2007 Time: 5pm UK Title: Walking with Robots: a new kind of engagement between humans and robots Abstract: http://www.hpl.hp.com/news/lectures/Abstracts/Winfield.html Speaker: Alan FT Winfield, Associate Dean (Research) UWE, Bristol Venue: Cabot Auditorium, Building 3, HP Laboratories, Bristol and broadcast Live to Palo Alto

HP Science Lectures are open to the public. Please forward this message to any friends or colleagues who might be interested. A drinks party is held following the lecture to give the audience an opportunity to meet the speaker in person.

Registration for the lecture IS A REQUIREMENT and places will be booked on a first come first served basis. Please either reply via email to debra-marie.duffety@hp.com, or contact me on the number below.

Debbie-Marie Duffety Executive Assistant Media Technologies Lab Hewlett Packard Laboratories Tel: 0117 3128759 Fax: 0117 3129937

Ive decided to have all of my Android Electronics assembled for me. I started building the circuits shown on my blog http://c6jones720cybernetix.blogspot.com/ by hand using Veroboard but I soon realised that with one SSC and up to 8 servo motors per limb, using Vero was just too much hassle. I dont mind paying a few hundred quid for the assembly costs. Im going to revisit robot vision in my spare time. I recompiled some of Motter's examples. He really is a true genius but I think I'll do things my way instead..

Lost all the work on my Hard disk! - How annoying thats about 4 or 5 years of stuff.

Did I mention Ive got a new blog? You can see the last android I built here: http://c6jones720cybernetix.blogspot.com/

Im working on a new one because the old one was too flexible.

Well the other day I built another dual servo controller so now I have three controllers.

Ive got a full sized bipedal robot with six degrees of freedom. I have only to calibrate it before I write software to make it walk

I decided to take a break from vision processing and finish off the legs for my biped. (I hadnt looked at it for a while). I spent ages trying to figure out why it wasnt doing anything. I reburned the program into the micros, stripped the thing down and put it back together and was pulling my hair out until I realised the servo connector was in the wrong socket! - Duh.

Put it in the right one and what do you know it worked, must be a miracle.

I finally figured out how to get video streams and still pictures from webcams and analog video cameras in C/C++. (It was a lot easier than I'd been led to believe)

I did some simple vision stuff on my older android project but that was simple statistical motion detection, and that was using somebody else's program to grab time lapsed frames for me.

Now I know how to do that myself, Im going to have a go at the hard stuff.

Whilst Im waiting for the circuit boards for my bipedal robot to arrive, Im going to brush up on all the stuff I learned about boundary, region and edge detection and have a little play.

- Motters was it difficult to use two cameras simultaneously?

Ive added another degree of freedom to my bipedal robot. Now it can balance on one leg! Wont be long until it walks.

Well Ive been closely following James Brutons robot designs and have come up with my own full sized biped mechanism based on similar principles.

I tested it last night under manual control and I think with a bit of tweaking, pretty soon I may have something that walks.

I had a look at James Brutons design for bipedal locomotion and Ive had a go at making a similar project for my Android. Now I think it is possible to make a simple human sized biped using a minimum number of components.

31 older entries...

X
Share this page