The Seattle Robotics Society welcomed back one of its original members at the June meeting. Anton Staaf, who attended his first SRS meeting back in 1981 at the old United Products store, just graduated from Carnegie Mellon University. He is winding up a month at home in Seattle before moving on to his new career at a high-tech software firm in Santa Barbara, California. Congratulations, Anton! You done good, and the SRS is proud of you.
Anton brought along a sample of the work he was doing at CMU. (I say "work," but it looks suspiciously like play to me.) His four-legged walking robot stands about one foot high and sports a pack of C-cell NiCd batteries, a home-brew 68hc11 controller board, and three R/C servos in each leg. He added custom electronics so the four 68hc11 output-compare lines, normally used to drive a single servo motor each, now drive three motors each using a multiplexing technique.
The walker did its thing, stepping smartly across the front table, looking for all the world like a pint-sized Battle-Mech from Star Wars. Even after the batteries ran down, the 'bot still drew a crowd of SRS members and a flock of questions. I especially liked the way Anton built the legs. He made each leg from a set of custom sheet-metal parts, designed both to hold the servos in place and to act as linkages for the motion. He replaced the back of each main leg servo with the sheet-metal leg material, then used the original servo screws to fasten the servo back together. This made a leg no wider than the original servo and used no additional fasteners.
During the ritual pizza-feed at Godfather's just down the road, Anton mentioned his plan to add some metal-working equipment to wherever he lives in Santa Barbara. That way, he won't have to drill and file every single bit of metalwork by hand. Sounds like a good plan, Anton. I'm sorry Dan Mauch picked this meeting to miss; I'm sure he could have given you some good hints.
I brought along the beginnings of my Mars explorer robot, which I've dubbed Tackle-bot for good and sufficient reasons. I made the frame from a top-of-the-line fishing tackle box I picked up recently. I pulled all of the hardware off of BYRD, my old beer-cooler robot, and have been installing it on Tackle-bot. I already have the motors on, thanks to Dan and his handy metal saw. The motors are kinda beefy, so I used a 1/4-inch aluminum plate on the underside of the box to stiffen the plastic and provide a mounting surface.
Next, I opened up the box and mounted a BotBoard-2 into the center of the top tray; it fits perfectly using only four spacers and some hardware. Ultimately, I think I'll replace the BB-2 with a 68hc12 board, but the 68hc11 board was handy and I was in a drilling mood.
What's so cool about using a tackle-box as a robot base lies in how easily you can reach everything inside. The trays all lift up and out of the way so all parts of the inside, even down in the bottom, are available. You can even fill some of the left-over tray compartments with small tools, so the robot can carry its own service equipment around with it. But I need to get cracking on Tackle-bot if I want to be ready for the SRS Mars exploration contest at the July meeting.
Speaking of July, the Mars explorer mission is due to land 4 July and hopefully everything goes well. You can read all about the Mars mission at any of the many NASA mirror sites around the world. I use http://mars.catlin.edu/mpf since it gets relatively few hits each day. Some of the instrument packages are quite ingenious, and the designers find many different ways to use the same sensor to do extra duty. For example, the explorer will test the Martian wind velocity at three different levels by sticking up a rod with three wind socks on it. But to sense the state of the three wind socks, the ground controller will simply point the stereo video camera at the wind socks and analyze the picture that comes back.
The camera also figures into the package that analyzes Martian airborne dust. To test for particles of magnetic material, one experiment will tip a plate containing a small permanent magnet out into the Martian breeze for a while, then tip the plate back inside the explorer. The video camera will then send back images of the magnet so the scientists on Earth can analyze the picture for any collected dust.
I use these examples to show how you can take one or two flexible sensors, such as a vision system, and bootstrap in several more complex sensors to make your explorer robot more versatile. I'm looking forward to seeing the other entries in the SRS Mars exploration contest. I hope all goes well, both for the SRS robots and the NASA 'bot on the real Red Planet.
Keep on keeping on...