This week was the second and last week for presentations for the first wave of presentations. This week we went, and though I didn't think I was too bad Deocampo said our presentation was disappointing, and that Dr. Furman was angry with it. I think he might have been talking about the Small Scale Solar team which I was not in charge of. For some reason the solar team disbanded and small scale solar is joining us again. Anyways, not very happy to receive the response we did but I guess it just means we need to improve on a few things.
Anyways since most of the research has been done for the small scale project our goal will be to turn the parts the previous years have created into a working model. After our presentation and Dr. Furman criticizing us for using a new communication component, and saying "what do you mean there's no controls from the Korean team?" made me thing. First of all Nasrat meant location's system, and second so far I have only ever seen the pod run the entire circumference of the track and not use the "shortcut" diagonal track. Which lead me to think perhaps the current systems relies on counting magnet sensors to find its way. The problem with this design now is that though hall sensors are fast they are essentially only transmitting one bit of information, and for the small scale to work the way we want to its going to have to be more sophisticated which means when an interrupt like that happens we need to get more out of it.
From this point forward we will be doing more development work on the cart, Nasrat will be researching his location system, which if he does not do in time, my bar code system will also be in development which can cover for locations. I think a better idea now would be to think of the bar code system as a replacement for the hall sensors. The track team is doing well in their research since Ernest left a lot of information behind for us to work with, and the cart team will be first making the bogey more reliable. Whether or not they decide to change to a swinging arm or simply change the link between the servo and arms. The controls team will also start messing with the sensors that need to see which ones they want to use. I personally purchased a CC3000 wifi shield for the Arduino and I will be figuring out how to program the main controls in Java since that would be the easiest for us to create a GUI with and since it runs on a virtual machine there will not be limitations on what operating system it will be used on.
Another thing that was not mentioned on the presentation was that I wanted to use a WiFi shield because its will be easier for us to work with, since its a more powerful communications protocol. I have nothing against the Xbee but why would we use one of those if there is something more powerful which later developments can be made on. If for any reason a lot more data has to be transferred, or if the model expands beyond 10 carts then they will need to change it again anyways so why not now? It's definitely not a cost issue since both chips cost around the same depending on which one you buy. The one I was proposing also had a breadboard area which we can put our other components on to stream line the design. The way they made it last year or over the summer has three levels of shields, one which is the xbee and the other only has a few components on it. Why not put it all on one board? This point wasn't addressed during the presentation I guess because there was some miscommunication, however I hope people under stand my decision now.
No comments:
Post a Comment