Monday, December 1, 2008

Group Meeting

With the deadline of the Senior Design Document fast approaching, we met today to discuss the changes we needed to make in our working version of the document. Today also served as a status update for the group; we have gotten a stepper motor to test out our stepper motor drivers, and though the stepper motor is now able to turn, it is not correctly interfacing with the stepper motor driver. We believe that this problem is due to the fact that our stepper motor is unipolar, while the stepper motor drivers we bought are meant to interface with bipolar ones.

Also, in an attempt to fix our problems with interfacing a webcam to the NSLU, we have purchased a NEW webcam (our third for this project) with the hopes that drivers for this new webcam will be able to better interface with the NSLU than the Philips webcam we have been trying all this time.

Wednesday, November 26, 2008

Parts: All Arrived and Accounted For

Because we estimated that our ordered parts would arrive today, we took the liberty to go out to Frys and Action Electronics to buy all the necessary parts to be able to begin building the circuit board. Components bought today included a circuit board with mounting area, a bread board, a stepper motor (not one that we plan to use for the project--its step angle is too inaccurate for our purposes--but to use as a test case for us to experiment with), a DC voltage to attach to the circuit board, and soldering wire. The costs for today were about $60; added to the $40 for the components bought online, we have spent over $100 on this project already. And we have not yet ordered the stepper motors we need (out of stock until December 1st) so we are fast approaching our initial budget of $200.

We have also managed to hook up our NSLU to a router from one of our homes. With port forwarding active, we can now access the NSLU from any computer without dragging the entire NSLU with us wherever we went.

Monday, November 24, 2008

Update...

We have solved our problem of not having enough hard drive space by attaching a 15GB hard drive on a USB mount and attached it to the NSLU. We have successfully installed Debian onto it, though the problem of how to make the USB webcam correctly interface with the NSLU still remains a mystery. More hours of attempts have not proven fruitful. On a plus side, we have begun exploring other aspects of our project--aspects that are equally important. We have begun writing the Senior Design document, researched the pic and how to properly program it, drawn a first draft sketch of our circuit board schematic, and researched on what program to use to process our images. And with the difficulties with the NSLU, we have begun to think about contingency plans...

Friday, November 21, 2008

Parts Ordered

As the NSLU has continually shown itself to be a hindrance, we have decided to circumvent our current predicament and have turned our sights to other aspects of our project; namely, the parts we need to buy. After careful collaboration, we have decided to buy the pic and stepper motor controllers we needed, as well as a few EXTRA in case we damage some of them. The only parts left that have not yet been ordered are the stepper motors, and the board on which we plan to place all the components upon.

Thursday, November 20, 2008

Roadblock after Roadblock

Attempting to get a webcam viewing program on an NSLU is like a dog trying to catch its own tail: after going around and around in circles, it accomplishes nothing. Following tutorials on how to get the program up and running leads to errors that tell us that packages are missing; downloading those packages shows us that we are missing other files; trying to find those files are impossible. After going through three or four tutorials (and a vast number of downloads), we have stumbled upon a new roadblock: storage space. Our 1GB storage has finally capped with all the downloads we were using, and now we are forced to find an alternative storage module in which to work on. So progress will once again be halted until we buy more storage space.

Wednesday, November 19, 2008

Hard work does pay off

Success! Changing to Debian has proven to be the right move, and after more hours of toils, we have finally managed to install working Philips webcam drivers onto the NSLU! It now correctly detects the webcam when it is plugged into the USB port, and we are in the process of installing a webcam viewing application onto the NSLU for further testing. A problem that has been discovered with Debian is that it causes the NSLU to now take about an hour to boot up before we can use it. Meaning that everytime we wish to use the NSLU now, we are forced to leave the NSLU on for an hour in advance.

Tuesday, November 18, 2008

One Step Back...but Two Steps Forward?

After a cumulative 16 hours of research, installation problems, and frustration, it was decided by the group that the Unslung firmware was too lacking in functionality to provide the features that we desired. As such, we have turned our sights to a potentially better firmware called Debian. Though it took another 5-6 hours of installation problems, it was finally installed onto the NSLU. Potential improvements over the Unslung firmware include Devian's better functionality, its many useful packages (in respect to webcams), and its various resources on the Internet in the form of tutorials, FAQs, and online posts. Potential drawbacks include its noticeably larger memory usage and its complexity. Already we have discovered that we may need SQL to use some of its features, a language we are not too familiar with.

Saturday, November 15, 2008

Unexpected Roadblock

More research and hard work resulted in the installation of various packages to give your NSLU more functions; including a C compiler which we intend to use to install our webcam driver. We have found the drivers we wish to use and have added it onto the NSLU. However, we are now at a roadblock; how to see if the webcam is interfacing with the NSLU. The given NSLU GUI does not reveal any connection with the webcam when the webcam is plugged in, and finding a webcam viewer for the NSLU is proving difficult. Several programs have been researched--most promising of these being "Motion"--but installing this has lead to more complications.

Tuesday, November 11, 2008

Installing Linux to NSLU: Check

We have managed to upload Unslung (a firmware) to the NSLU and are able to access it through a router and desktop computer. However, attempts to download a Philips webcam driver onto the NSLU prove to be more challenging than meets the eye.

Monday, November 10, 2008

NSLU2 ready

We decided to use a NSLU2 to receive images from the infrared camera and perform calculations on them to find the locations of subjects. We were successful in loading unslung on it. The next step is to install necessary drivers on the NSLU2 so that it can communicate with our infrared webcam. We also narrowed down the design of the final product to two possibilities, each with its own required part list. A decision will be made soon regarding which option shall be pursued.

Monday, October 27, 2008

Infrared webcam ready

For this project, we decided that the most accurate way to track the targets that spotlights need to follow would be to place an infrared light emitter on top of each one of the targets. The next step would be to place a webcam modified that only picks up infrared light and place it on the roof (high above any possible prop obstacles) to pick up the infrared light and use it to locate the targets. We decided to test if this was a possible method.

We bought a webcam from frys to modify it into an infrared webcam ($15). However, the modification process proved to be a tad bit trickier than we initially thought, and the webcam was damaged in the process. Then we got another webcam from walmart and managed to successfully modify it ($20). The following is a small clip created using our modified webcam.

Background details about the video: This video was filmed using the modified webcam, in a mildly lit room, with light sources consisting of natural sunlight and a fluorescent lamp. The reader should not that neither one of these sources are visible in the video because of the modification of the webcam. Such light sources are censored out. The white dot that is seen is the infrared light emitted from a normal remote control.



So we have an infrared webcam ready.

Monday, October 20, 2008

Project information

Group members:
Vamshi Mannam (30387733)
Nihar Desai (66057656)
Melvin Asuncion (34916425)
Brian Solloway (86251105)

Mentor: Alexander Ihler

Project description: Our group intends to design a fully automated scale version of a spotlight, which will follow an object with no required person "behind the scenes" operating any machinery. We intend to cater to a number of reasonable assumptions to make this project practical for real life usage, including the accuracy of the projection of light from the spotlight and taking into account the various props that may be on the stage. Also we will have multiple spotlights each which its one target to follow.