April 2, 2008
Progress with IR-sensor Robotic Flower
Parts and Tools Acquisition: As of now I have received the vast majority of ordered parts including the Qwerk robotic processor (“the brain”), the aluminum tubing and connection pieces used to make up the main body (stem), three sharp IR sensors, 6 micro servos (HS-81) and one larger servo (HS-422). I have purchased a basket for the “flower pot”, poster board and assorted craft accessories (glue, construction paper, double-sided tape etc.).
Still Missing: The servo extension cables (x10) needed to connect to the Qwerk for power and testing and the to initialize and coordinate the flower's motions, were out stock when the rest of the parts were ordered from Lynxmotion. The recipe recommends 24” cables, but I have decided to order a set of 12” cables to see if those will work. I called Lynxmotion several times and they told me that the 24” cable would be in stock soon. Yesterday they called back to say that the “servo extension cables” were in stock but neglected to define the cable length of the available stock. I'm in the process of finding out right now which ones will be available and by when.
The AC adapter for the Qwerk has to be made from a female coax cable, and a male/female kyosho battery connector. Both of those parts should are being ordered this afternoon by my friend in town who is willing to use their credit card in exchange for the cash value. I spent a few hours over the weekend attempting to make due without them but fell just a little short. Without these parts I can't yet power on the Qwerk to obtain a Qwerk ID, test the servos, or establish the ports for each servo and IR sensor.
Finally, I need two screws, two lock-washers, and two nuts all sized 2-56x3/8”. This has been the most annoying thing to happen all week. Although most of the needed screws etc came with their accompanying parts, these just didn't. I went through all the random screws I could find in my lab room and at home, and came SOOO close to a perfect match, but in the end I had to just order them online (Walmart, and Home Depot didn't have them, Brown and Roberts said they thought they had them but couldn't find them in their inventory). These screws are just needed to attach the rest of the robot to the Qwerk, so for now it's no real biggie.
Construction Progress
The stem and body have been assembled and properly jointed. All 7 servos have been connected to the frame and positioned to allow for proper rotation. There is center of the flower without the optional speaker amp, but aside from aesthetic value it's not necessary.
I created a template for the flower petals, cut them out, and tested their support of the added weights. Although I only used poster board, they appear to hold up, but if there is any structural problem with the petals after bending, the issue can be easily remedied with a new template, or a reinforcement of the board. The flower is essentially assembled except for its connection to the Qwerk board (I might just go ahead and used whatever I can find to attach it, and then reattach later), and the connection of each petal+L bracket to the individual servos. I'm hesitant to do this until the Qwerk can be powered on and each servo tested and initialized. It seems like they would just get in the way if attached now, but I think I'll try it out just to see how they move together when the servos are manipulated manually.
Attached IR wiring to Sensors, and attempted securing the sensors to the petals. That's when I decided that the poster board is not strong enough to support the L brackets and the IR sensors. I'm going to switch to either cardboard (I have plenty of it) or buy some “conservation board”... at this point I strongly prefer using what I can find versus buying anything else. The dimensions of my template need to e adjusted slightly to give enough leeway for the IR sensors.
Programming: I haven' looked over enough of the Java source code, but I did read through several of the forums on the Terk website to get a feel for what I should I concentrate the limited remaining time on to get this project running by the end of the term. There are two newer versions of Java code for the Terk robotics project: one is for the newest robot recipe called Create, and the other is the “My first Robot” programming tool (I think I've already mentioned those in a previous post). Anyway, the actually programming will not be overly difficult once the Qwerk ID is obtain (after the AC adapter is made), and the Flower Power program is turned on. With the FP interface running I can plot out the movements I want the flower to execute, and with the API I can begin programming it. Easier said than done, I know, but think now would be a good time to call in all those volunteers who wanted to participate this semester but couldn't for some reason or another.
