Date: Mon, 16 Jan 2006 02:23:18 -0600 (CST) Subject: carry handle and mission x X-UID: 122 Content-Type: IMAGE/JPEG; NAME="img2547.jpg" Content-Type: IMAGE/JPEG; NAME="img2549.jpg" Last update, I was so tired that I wrote "will be ready on Sunday" when I was thinking "will not be ready on Sunday". So things were not ready Sunday. Friday evening, I was so tired that I started to make stupid mistakes. A good indicator is when something breaks. I pushed the end of an antenna into a 1/4" rubber hose. (Why would someone do this, you may ask? It seems logical when you are exhausted.) Then while pulling the antenna out, the end cap came off. This is what happens when you are tired and lack judgement. Fortunately, there was no meaningful damage that I can perceive. A little electronics silicone and the cap goes back on. Lacking an EE background, I had guessed that the radiation pattern would be out the sides of the antenna. Now that I know for a fact that it is a helical geometry, this is pretty much confirmed (according to the English Wiki on antennas). And end-fire antenna would be pretty much useless for mobile terrestial application so it must be side-fire. Various sundry but necessary structural parts are being added to the vehicle. The angled lexan plate is modeled after what I've seen on 4 wheel ATVs. It wasn't easy to make even this with a heatgun. Anything much larger and an oven or heat strip would be necessary. The plate protects the front in the event of collisions as well as making the robot easier to handle. I don't have as much fear of smashing the delicate front steering potentiometer assembly. There's a lexan plate for the electronics box to sit on. I'm going to velcro the aluminum box down. Believe it or not but I had not considered velcro before today when I happened to see it in Home Depot. I kept thinking in terms of silicone adhesives and straps. The white rod down the length of the vehicle is a carry handle and also provides some protection in event of a rollover. The rod is actually a 7/8" wooden dowel covered with heatshrink tubing. I feel a little guilty about using a natural material like wood. But actually, it's probably the best material next to a fiber reinforced composite such as glass or carbon fiber in epoxy. At the microstructural level, wood is itself very much like a composite material. That's why it is relatively strong for it's weight and stiffness. A plastic pipe would be heavier, not as stiff, and arguably more prone to catastrophic failure from shattering. And the wood is "warmer" to the hand than most plastics. I didn't want to use aluminum as it might interfere with the radios and GPS receiver. Most of all, I had this dowel and extra heatshrink tube laying around that I would otherwise never use. During the Joint Strike Fighter fly-off between Boeing and Lockheed-Martin, Lockmart won the contract with an aggressive demonstration they referred to as "Mission X". The X-35B had a short take off, went supersonic, refueled in mid-air, went supersonic back to base, and then landed vertically. I'm thinking that I need a goal like this for the robot. On the way back from Home Depot, I remembered that I had done all of my speed estimates based on the low gear of the Panasonic drill motors. I'm worried about low speed torque. But in high gear, the speed may be doubled. So my guess of 15 to 20 mph maximum may actually be 30 to 40 mph. This is not unreasonable as model RC cars can often go 40 or 50 mph. At 30 mph, this is fast enough for the robot to keep up with traffic speed on many city streets. Of course, I have no idea what the actual performance of the robot will be until I start testing it. I'm in the habit of driving with the left hand only. That leaves the right free to operate controls. I'm thinking that "Mission X" for the robot is: 1. Power up all systems in the apartment complex parking lot 2. I drive the car up to the front gate to open it (requires weight on pressure sensor) 3. The robot follows me and waits 4. I drive out through the gate and pull over to the side 5. The robot follows me out and positions itself ahead of me and waits 6. After I send the all clear, it makes a left turn down the road into the park 7. I follow the robot in the car, acting as spotter with the option of manual override 8. The robot makes a right turn and drives down the road and over a small bridge 9. Shortly after the bridge, it makes another right to exit the park 10. Another right and it waits at the protected intersection to Buckner road 11. After I give the all clear, it makes a right and accelerates up to top speed down the road. It is a long straight away with a posted limit of 40 mph. I'll shadow behind the robot with hazard lights and protect it. 12. At the end of the quarter mile or so, it makes another right to Poppy drive and returns back to the apartment complex where it waits for me to drive up and open the gate. 13. After the gate opens, I drive through and it again follows. 14. I park near the entrance and the robot finds and halts near the leasing office for safety. 15. All systems are powered down. This is probably difficult but yet seems possible. Continuous right turns minimizes risk. I'll have to get more cameras to record everything. I may need help. I'm thinking optimal would be a driver, a computer/comms operator, a camera operator, and a grip to grab the robot in case anything goes wrong. We'd all be in the car. And a time with very light traffic is preferable, perhaps a Sunday when many are going to church. I guess this sounds crazy but I can't tell. Some of the Grand Challenge teams were doing things like this. So it can't be so strange.