Older blog entries for cschur (starting at number 46)

Hi All,

I am in the process of designing several optical shaft encoder projects for work for a new robotic motor application, and bought some very promising encoder disk photo reflector chips I thought Id pass on to you. The tiny Sharp GP2S series of IR photo reflectors have a range of only a few mm but put out a clean square wave which can be tidied up a bit with a schmidt trigger or compartor with hysteresis if needed. They are inexpensive at 84 cents each and for that price I got a dozen of them! The plan for now is to see how well they read an shaft encoder disk printed out on a laser printer for positional information. Very cool devices!

My vacbot project at home is comming along. Ive got the wheels attached now, and at full voltage its way too fast. But I intend to run it on PWM and so will shoot for a slow half a foot per second for the task of keeping our carpets clean. Im working on a new bumper design that will consist of those 2x3 small circuit boards from radio shack with four small membrane type push switches from Jameco in between on the corners forming a sandwhich. Should be sensitive, yet take up very little space.


I got around to starting a web page on the vacbot Im working on, based on the "Dirt Devil" motorized sweeper as a platform. This page shows the starting point, before modifications. Ive been busy turning parts on the lathe for the wheel attachements and now the huge 6 inch lawnmower wheels are on it. They have a polyethelene core, and are very light and durable. And for $3 from walmart, they are perfect for the job of carpet roaming. Here is the start page for the new robot:


I bought the box and circuit board, connections etc last night, and will be wiring a 16F628 and L298 to control the twin black and decker battery operated drill motors Ive adapted to run it.

Im all ready thinking on how I want the FSM to control the vacuuming action. Put in the center of the room, then it start an overalpping spiral increasing in size until it hits something with its bumpers. then it goes to the random wander mode. THEN every 5 minutes or so, when a timer in the Pic times out, it goes back into the spiral mode. This way huge areas of the rooms will be cleaned quickly. Corners and agains the wall will be the last to be done.


Ive got some great projects going on to report. First, as an industrial robotics engineer, I am exposed to many components that have good potential in the home robotics arena. One part, Ive been using a lot of lately is the L298 motor H bridge driver IC. If you havent looked at this one, you should - its very cheap, has a high current capacity especially when used with a heat sink, and is easy to use. There are a few issues with this part, that make it a bit difficult to implement. First, the wierd pin layout can be easily over come by bending the outer row of pins to line up on the standard .1 centers to make two neat rows. Also, there are 8 external clamping diodes to consider. These keep the inductive kick from a stopping motor from getting back into your L298 and hence into your uP and poping it. While you can use a 1N4000 series diode for low current motors, you really need to use a shottkey fast recovery diode for the larger motors. THese are cheap too, and you can order both L298 and Shottkey diodes at Digikey or Jameco for like 4 bucks:

http://rocky.digikey.com/scripts/ProductInfo.dll? Site=US&V=497&M=L298N

Another neat thing to get you going on your next robotic project is the new Dirt Devil electric sweeper. This is one of those push sweepers with the rolling brush inside for medium duty cleanups in the house and industry. But this one, is VERY cool, in that it is electric/rechargeable and should be relatively easy to make robotic. I couldnt resist so bought one from Walmart for under $30. It is actually an amazing floor and carpet vac! Super cleans. It comes with a charger, and a 7.2v internal battery. I ran mine on a ful charge to see how long it would go - 3.5 hours! Now that will scrub those floors to a polish, and keep that shag rug spotless. I am going to add a dual wheel drive, and some bumpers to get it going. Youve got to get one of these things, they beg to be made into a robot. This is an add for the more expensive model, they dont have one yet for the new under $30 one yet:


Finally, last night I got our new robotic digital sky camera in operation. Youve seen my robotic film camera and barn door type mount. This new one operates the hand held remote by pushing the button with a mechanical linkage and timing the bulb exposures, operating the star tracker for all night unattended sky shooting. Remember, a robot doesnt have to have wheels to be a robot! This one takes the hard labor out of haveing to sit there all night long and manually click the cameras shutter, rewind the tracker and time exposures. More on that and some great images I got last night with this setup are comming...



Well, Ive concluded that the RIS software with the mindstorms kit is truly wimpy! Can hardly do anything with it. But if I keep in mind Im only trying to prove the validity of new mechanical designs, then its ok. Its hard to get serious on doing months of hard work on a robot that you will have to tear apart to get the kit back again when its over with. Enough said on Lego.

Im working on two interesting robots now. The first is a precision navigation robot, to experiment with pinpoint navigaion to a given point and back. Ill need this experience to start the next huge robot project, the GeoBot, a geologic explorer robot that will autonomously go to a specific location (my yard first!) and collect interesting rocks - THEN bring them back to me all by itself. It will be awsome. But first, I am using my PICbot to do the navigation proceedures and get them down.

The other robotic device is to make my Canon 10D automatically take long exposures of the night sky on a tracking platform, much like my Aurora Cam robot. But I will have a plunger push the remote button to start, and then time all the exposures all night long unattended. This will be also a cool project.

Well, thats what Im working on for now, write me if you have any great ideas to include in my upcomming geobot - the rocks are waiting to be discovered!



Hi all,

I am working now on two separate robotic projects. I have to temper this work with my astrophotography (www.schursastrophotography.com). On one front, Ive spent the $200 bux and got one of those mindstorms kits. I always wanted to know why there was such a cult-maddness about that system. I built my first robot with it, a simple wall avoiding light sensing rolling bot, I called XBot. You can see it here (dont laugh):


Now that your done laughing, well see where I head on this. I expect to be able to test new mechanical designs primarily, because the RCX code is so, well, wimpy.

NExt, Im starting a now multiprocessor PIC robot project, NAVbot. My plan is to get a small table top robot to very accuately navigate to precise coordinates and experiment with the CMU cam. The goal is to get the routines and math down to incorporate into the much larger tracked GeoBot which will be an outdoor bot that uses the Stuart M5 tank base that has been stripped to the bone.


HI all,

Well, we finally did it. We finished our PAAMI project, and the last goal was realized this weekend, to locate, collect and deposit an empty soda can in the charger bin. Ive put together some movies of this final project, you can find them here:


Thanks for looking! Chris

Made more progress on the final challenge for PAAMI. The top most processor in the hierarchy is the processor that will search, identify and grab a soda can, and then deposit into a bin on the top of the charger. Ive programmed all the pieces separately, and they all work, now its a matter of assembling it into a complex Finite State Machine. None the less, last night when the "can search" switch is thrown, the robot goes into a can search routine. The robot goes forward or avoids obstacles then every 5 seconds the new processor subsumes, stops, rotates 360 in 55 steps while scanning for a can with the new can sensor. If a can is seen, it stops, makes the "happy sound" and for now stops. Eventually it will approach the can to put it in the grab zone, and pick it up and go off to the charger.

One issue is the fact that if I dont leave at least 4 to 6 seconds between can find processor subsumptions, the robot has some difficulty escaping objects since it takes a few seconds to go away from impacts. This leaves gaps in the coverage on the zone around the robot to find a can. Right now it can reliably spot a can up to 16 inches away. Pretty cool, ay?


Hi All,

Havent posted a message in a while, been busy getting my act together for the upcomming Riverside Telescope Makers conference in California. Im giving a talk there.

On the bot front, Ive been struggling with getting the a/d converters working right with Picbasic with the 16f73 device. It works fine for one A/d, any channel, but If I tried to take measurements from two channels consecutively, it would muck up. Finally, I had to use the ADCIN command, and it works perfectly. I now have a can sensor that on the bench gives a go/no go lamp when a soda can is within its beams. Im adding the sensor array to the front of Paami, and have now wired it in to the last level 8 processor. Now onto programming the FSM's for this level!

Write me!

Chris Schur comets133@yahoo.com

Hi all,

Well, here it is - our first operational images and movies of our new robot arm/gripper on PAAMI. Right now, Im just showing you how it works, and how it will be used. The arm motor is slow right now to keep the arm from jerking when it lifts, and will be replaced with a standard servo which is on order. This is a great example of priority arbitration architechture. The highest level now is the Grab module, and subsumes the other levels when triggered.



Chris comets133@yahoo.com

HI all,

Another update on PAAMI. We are on the last processor module we intend to implement, the can seek and grab module. Im using a PIC16F73, which has 22 inputs and a nice4 A/d inputs at 10 bits. Its also great because it will hold 8k of program memory. This will be the 12th parallel processor working in her. This weekend, I wired in the servos on the arm including the lift and grab motors, and got the new processor up and running. I was able to FOR THE FIRST TIME to have the robot reach out, grab a soda can and lift it and store it on its back. Pretty cool, ay?

Now heres the freaky part. I removed one of the two boards in the bot, and had it on the bench to add the new processor. But the other parallel processors were also powered up on that board, and were alive and thinking! While I tested power and voltages on the socket for the new processor, the dozen lights or so on the board were flashing, showing status and trying to drive the bot around! Oh man, thats kindof freaky working on a living AI....

Chris comets133@yahoo.com

37 older entries...

Share this page