Older blog entries for cschur (starting at number 86)

Well, we have finished all the arena testing and evaluation of the PicBot III unit, for exploring the pros and cons of IR Proximity navigation. Now, we are working on the final report, to be shared when it is complete. There were a few surprises with the IR prox sensor array.

First surprise was how much larger the robot becomes when "non contact" IR prox sensors are used. In other words, the virtual size of the robot nearly triples for white walls, and only increases an inch for black obstacles! And to make matters worse, you dont even know what the reflectivity of your obstacle is, so are you a foot, or an inch away? This creates some amusing tunneling entrapment situations. You can get stuck in a confined area that looked dark when you went in, but is white when you turn around to escape - and thus you are quite trapped.

Another surprise was the reflectivity of obstacles in 940nm IR is quite unpredictable. Plants are pure white, so are my black socks! Black poster board is a neutral grey, and black IC foam is totally invisible. To help visualize this, we are adding to the write up an extensive set of images of household objects photographed in deep infrared so you can actually see what the robot will see when it moves about. Pretty cool, ay?

Chris

comets133@yahoo.com

Hi All,

Finally, I can show you what Picbot III looks like. We have been really working hard in the evenings getting it done, and the mechanical and wiring are now as of last night finished. PicBot III is a follow on to the II version, which was to research the bumper logic. Here, the III version is to do the same with IR prox sensors. So Here is the main page:

http://www.schursastrophotography.com/robotics/picbot3main.h tml

and here is the front of my robotics web page, with PicBot III also:

http://www.schursastrophotography.com/roboticsmain.html

Let the programming begin! ;)

Chris

comets133@yahoo.com

HI All,

Were now building our next robotic project, Picbot III, which will be used to gather data for a very in depth project on - you guessed it: IR Prox Logic.

I had some stamp sized circuit boards made by PCBexpress, and they are for the IS417 prox sensors and contain the dual 910nm IR leds. The interesting thing to think about with using this device is 1. The modulation frequency is 50kc, and wont interfere with other IR devices at 38kc, and 2. the wavelength of this IR is MUCH higher than a standard IR like 700 - 800nm. This means my web cam cant see the beams in the dark! This would be to charecterize the beams for an angular resolution analysis. I do however, have one of those edmund phosphor cards which it barely activates. So the quest continues as to how I will veiw the beam directly. (my russian IR night vision scope cant see the beams either!)

Anyone have any ideas for 910nm?

Write me:

comets133@yahoo.com

Chris

HI all,

Just finished a huge project, our article on Bumper Logic. If you have any interest in programming and implementation of a bumper based robot, this article may be of interest:

http://www.schursastrophotography.com/robotics/bumperlogic.h tml

Thanks for looking.

Chris

Let me know what you think: comets133@yahoo.com

Hi all,

Were almost done with our current project, to develop a bumper logic processor for future robot projects. We have now put the picbot II through everything from mazes, tunnels, dead ends, tons of corners, and other obstacles such as edges to develop a write up on a generic bumper chip. The article is nearing completion, I have many illustrations to do on it yet. And ill post the whole research paper here for you to evaluate. Here is the first intro paragraph as a teaser:

 eBumper Logic‍

Usually, the first sensors the robot builder installs on a new robotic creation is a frontal bumper. As to what exactly they plan to do with these last resort sensors is generally vague, there is very little detailed information describing the behaviors to implement in the programming. Most books seem to skim over the subject, indeed other than  eturn right for left impact†and  eturn left for right impact‬ most publications move on to other more enticing subjects such as sonar or mapping. So here I wish to fill in that critical gap in information and discuss in more detail the issues that lie beyond simple  ebumper logic‮

More details later!

Chris

Hi all,

Ever considered making your very own robot test arena? We just finished our up last weekend, and it is essentially a 2 x 4 foot piece of formica covered particle board, with 4 inch high white rails on all the sides. Here, we will be developing the software and hardware in this "robot play area" for our current project.

We have constructed a small universal type robot that we can bolt various sensors to for evaluation. Each sensor will be one level in a behavioral priority arbitration architecture. Then after we have optimised each separate layer independently, Ill design a PCB. This will be a universal Priority Arb. Arch. Board, with sockets for 8 PIC processors that all run in paralell. There will be three or four digital demultiplexers, and of coarse the arbiter PIC, which we allready have from our PAAMI project.

So our PICbot II project will develop the "impact module", which will have six bumper plate inputs and will contain all essential escape routines and some unual ones as well. A single subsume line output goes high when the chip decieds it must take action upon an impact detect.

Other PICbots will be for developing the other modules, such as IR avoid, sonar cruise, beacon homing, charging, docking, and task planning. Ill post a shot I took last night of the arena on my web site.

Chris comets133@yahoo.com

www.schursastrophotography.com/roboticsmain.html

HI all,

If you watch any football this season from ESPN, CBS, or NBC, look carefully at the goalposts - those teleoperated robotic cameras you see that get all the great shots from the vantage point they are at are my babies - I design, maintain and repair those guys for a living. They are operated from thousands of feet away using a link of sending RS232 serial data over a coax cable.

Ive got the basic platform for our new PICBOT II running. A 16F73 PIc processor sends commands serially to two servo PICs, with software I designed to control them. Our next phase is to develop bumper plates out of double sided circuit board material that is dirt cheap, and can be made any size and shape. My real interest is to improve the AI behind bumper navigation and to design the bumper layer of the prioritized arbitration architecture most effectively. Then Ill do a write up detailing the math, action and bumper design for the robotics community to see.

Chris comets133@yahoo.com

Hi all,

were working on a series of small table top robots now, to develop in sections the various levels of subsumption architecture, one level at a time. I am doing this piece meal, to eventually develop a superior home cleaning robot. First, we are tackling the impact level, with bumpers and a new design for a plate style bumper. First I am working on a one chip per servo driver, using a PIC12f629, an under a buck processor to take serial data from the main processor and control a single servo. Ill of corse share it with all of you when we get it perfected, and you can have both the schematic and software ready to burn for your own servo drivers.

Chris

10 Aug 2006 (updated 10 Aug 2006 at 15:19 UTC) »

HI all,

Ok, well were finally finished with the Geobot project, and I would like to present to you the Final Report:

www.schursastrophotography.com/robotics/geobotfinal.html

There can be no doubt that the technologies we have gained much experience in on this project will be a great foundation for future out door robotics projects for us in the comming days.

Chris

comets133@yahoo.com

Hi all,

We have made another significant milestone in the progress of the Geobot, the rock collecting robot. The arm, motorized scoop, and ground detect sensors are done. We connected the interface to the processor tonight, and were able to move the arm smoothly, turn on the motorized rock scoop, and yes even detect the ground with contact switches. Our next step is to actually program the Finite state machine that will do the actual rock aqusition, stow the arm, and return the sample to home! This is certainly the most sophisticated robotic arm we have constructed yet. On my main page, Ive posted a new shot a few minutes ago of the geobot with its scoop arm raised:

http://www.schursastrophotography.com/roboticsmain.html

Note the new "Schurs Artificial intelligence Scale" at the top of the page too...

Chris

comets133@yahoo.com

77 older entries...

X
Share this page