That looks like a pretty standard type pan/tilt turret to me from that picture. Do you have a better picture of it? I had to find a way to get a sensor closer to the floor because WALTER has always had a problem with low lying obstacles. I just hope the PING has a wide enough field of view. If not, I can put an IR Ranger in its place.
I am working on a design for a new arm thing to put on the top of WALTER. It will hold 5 to 8 sensors, depending on how I build it, and possibly a gripper. I just have to figure out how long the first segmet has to be to get over WALTER’s top deck on three sides (front, left, and right side) so it can reach down towards the floor. I’ll post pictures of a couple potential designs when I have them finished.
Its actually a modified version of the standard Pan/Tilt Lynxmotion offers. Bassically, the panning servo is mounted sideways instead of verticle. Heres a pic from my Rover:
Exactly, didnt know how to word it at first… I can tell you though, that this setup is great for seeing obstacles at the foot of the robot as well as on 45 degree angles at each side. It actually gives about 135 degrees of viewing area in front of the bot (if im not mistaken) depending on what kind of servo you use. Also, idt you would have a problem with the ping bouncing off the floor with this setup as much as you would with a regular pan and tilt becuause it is relatively straight and dosent point directly to the ground when looking down, but rather on an angle. anyways, its really good to have, its one of the best features on my robot!
I get a total of 90 degrees (+/- 45 degrees from center) with my pan/tilt arrangement. I have IR Ranging sensors pointing across the wheels, so they handle the rest of the viewing range.
I can tilt the PING up a full 90 degrees (good for checking overhead clearance) and down about 10 degrees to adjust it if there are echoes from the floor. I would have more downward tilt range if I mount the PING right side up, and I still might be able to get all the low obstacles. I have not actually used the PING yet, so I don’t know how its field of view is yet. I won’t know what configuration works best until I get WALTER working again - and I have to convert all his code (13Kbytes total) to the Atom PRO before that will be possible.
Sounds interesting. Your the first one I know to put a ping on a pan tilt setup at the base of the robot. Deffinately will be a challenge but Im sure it will come back with great results.
It makes sense to pair an IR sensor with an Ultrasonic sensor. What one doesn’t see, the other one probably will. There is much less chance of something being missed if both types of sensors are there.
I would love to see an SES double sensor bracket that could handle an ultrasonic and IR sensor pair. If you saw the pictures of the last turret I made for WALTER, you can see what a pain it was to get two sensors stacked like that (IR above PING). Having a nice dual sensor SES bracket would be soooooo nice - it would not be hard to make from the model of the current multipurpose sensor mounting bracket. It would also be much lighter than the method I had to use. My method required a heavier duty servo, since the HS-422 was not strong enough to hold position with my dual sensor bracket kludge. I did find an almost perfect dual sensor bracket at SuperDroids, but it doesn’t have quite a long enough mount to be SES compatible and will not work with all SES designs. I’d have it made myself if I had the ability and could afford the cost of prototyping.
I have finished my design for a possible arm to attach to WALTER’s top deck. It may not work with the gripper because there may be too much weight for the last arm servo to hold, even if I use an HS-645 there.
I can reduce the weight of the senso/gripper mount a little by not having the wrist rotate servo and just mounting the gripper to a C bracket. I could also use the shorter C brackets instead of the long C brackets.
Sensors attach to the “wings” on the U channel as well as to the top of the U channel. I’l put dual sensors on each “wing” consisting of a Sharp IR Ranger and a PING or other ultrasonic sensor such as the SRF05 (standard) or SRF08/SRF10 (i2c).
My idea for this arm is to have it be able to reach a full 90 degrees either side of center, and also be able to reach to the ground within that range. I would very much like to mount a small camera on this arm also on the top middle of the U channel with an TPA81 (i2c) thermal array sensor. A compass module (i2c) will be mounted on the back “wing” of pan servo mount.
looks nice! BTW what is up with superdroidrobots.com. I desperately need to buy a 2.4GHZ camera from them for this week’s science fair but everytime I try to go to the site, it says I have to log in. What is the deal? Do i need a login key?
Im in desperate need of a 2.4GHZ camera! Ive been working on this project since July and Im only 75% done. Unfortunately a small science fair is March 2nd and I already signed up. If I dont get most of the testing and writing/backboard done by This week Im doomed
Their site is apparently broken, or perhaps has been hacked. We will just have to wait for it to be fixed. There are a couple small items I want from there also, like some dual sensor brackets.
I have been working on a redesign and upgrade for W.A.L.T.E.R. lately.
The first thing I want to do is fix the ability to detect those pesky low (from 0 - 1.5" off the floor) obstacles WALTER always has had problems with. I removed the sensor turret from the top deck, redesigned it, and mounted at the very front and center of the bottom deck. The new from sensor platform can pan +/- 45 degrees of center and tilt the sensor(s) on it up a full 90 degrees and back (approximately 15 degrees) far enough to see anything on the floor. I currently have a PING sensor mounted here and want to add a Sharp IR Ranger.
I have come up with a different purpose and design for the new arm I will add to WALTER’s top deck. This new arm (preliminary design, untested) will be able to cover +/- 90 degrees from center and reach all the way to the floor. There is a sensor platform mounted at the front which will have a Sharp IR ranger, Ultrasonic sensor, camera of some sort, and a TPA01 thermal pile (for detection of warm bodies and other heat sources). There will also be a gripper at the very end of this new arm, which will have a full 3DOF (up/down, side to side, wrist rotate) of movement.
I just ordered the motors, mounts, and quadrature encoders to convert W.A.L.T.E.R. to using motors instead of servos for locomotion. I am also ordering the materials to make W.A.L.T.E.R.'s new 10" size body.
This conversion will be preparation for building Walk 'N Roll after The BiPod is completed and working. It will also be part of my SES motor mount project to create SES compatible motor mounts for gearhead and R/C style motors with an Open Servo based motor controller.
Today, I got four panels of the same material WALTER’s current body is made of. There are four 12"x12" panels and they are all Lynxmotion Yellow.
Now I have to get the decks cut out. I am planning to make one standard deck, one wheeled deck that will acccomodate up to 5 1/2" diameter wheels (with 1/2" clearance on each side) that are up to 2 7/8" wide (with a bit of clearance), and two decks with tabs to attach legs too. The wheeled deck will accomodate the wheels entirely within the body diameter.
I can tell you though, that this setup is great for seeing obstacles at the foot of the robot as well as on 45 degree angles at each side. It actually gives about 135 degrees of viewing area in front of the bot (if im not mistaken) depending on what kind of servo you use. Also, idt you would have a problem with the ping bouncing off the floor with this setup as much as you would with a regular pan and tilt becuause it is relatively straight and dosent point directly to the ground when looking down, but rather on an angle. anyways, its really good to have, its one of the best features on my robot!
