Misc. IGVC Archives

From GT RoboJackets

Jump to: navigation, search

Here is an archive for dumping old information which will be kept for historical purposes.

Contents

[edit]

IGVC 2006 Orphaned Pages

[edit]

IGVC 2006 Current Events

[edit]

Sat, Apr 8, 2006

  • Andy will finish electrical system prototype
    • Make CAN cables
    • Wire Motor drivers for power
    • Box GPS unit (Will it work in a metal box? Do we have a plastic one?)
  • If time permits, I will also work on migrating our wiki to a robojackets wiki
[edit]

Sat, Apr 1, 2006

  • Due to vandalism on the site over spring break, the following configuration changes were made by Andy:
    • editing by anonymous users disabled
    • self-registration by anononymous users disabled; only SysOps can register users now, using the Special:Userlogin page
  • As a sideeffect, Google should no longer be able to index our user pages, which contain the personal contact information of a number of IGVC members.
  • The following users now have both SysOp and Bureaucrat privileges
  • I've successfully upgraded the wiki's version of MediaWiki from 1.4.10 to 1.5.7! --DavidF 20:20, 1 April 2006 (EST)
[edit]

Sat, Mar 11, 2006

I will not be at this meeting, but I [gave] Andy the key and our new Digikey parts so that work can be done on all of the cables. I am going to see what I can do with the sonar from home. We might want to meet during a weekday next week since spring break is coming soon and we should try and have hardware done by then. --Logan

I have the key and parts from Logan, and hopefully this weekend, I will

  • Build the power portion of the the CAN breakout box
    • mill holes for connectors
    • attach 12V fuse block But I mounted the fuse panel upside down according to convention... whoops! (Still works though, and maybe better...
  • and some of the electrical distribution system (we got a gold fuse block - thanks Steven)
    • Maybe box which holds the poser distribution system (No box, no idea)

--Andy 18:31, 10 Mar 2006 (EST)

I am planning on doing a complete rewrite of Ben's vision code. It will probably take a while. --DavidF 14:26, 11 Mar 2006 (EST)

  • The majority of Ben's legacy vision code is now removed or completely rewritten.
  • The new vision code still needs to be implemented. Unfortunately I can't do that now effectively because Logan has the IGVC laptop, which is the only computer on which the IGVC code can fully compile and run.
  • Today I also introduced a number of flags that control whether the code uses (and thus depends on) certain external libraries. By deactivating these flags I can almost get the entire IGVC codebase to compile on my own laptop (which lacks all of the external libraries except QT). Unfortunately there are a few linker errors remaining which I do not know how to fix. They appear to originate within QT's auto-generated .cpp files, but I could be wrong.
[edit]

Sat, Mar 4, 2006

  • David: Debug the module code for the sonar (this may involve manually implementing the I2C protocol).
    • So after you left David, we found the mysterious why we can't communicate with the sonar module via CAN problem.
      • Short answer, it was electrical and a very obvious mistake.
      • Long answer, something screwy with the programming cable was causing the processor to freak out when it was plugged in.
      • Lesson learned: unplug programming cable unless you are presently programming the device.
      • --Andy 11:30, 5 Mar 2006 (EST)
[edit]

Sat, Feb 25, 2006

  • David: Testing CAN hardware
    • Joysticks (wired, wireless)
    • Motor Driver Interface
    • GPS
    • Sonar
  • David: Write/test code for new CAN hardware
    • Sonar (still working out the bugs though)
  • David: Plan/architect/design the navigation system for the production codebase.
  • With any luck, Paul should be dropping in to show us his GPS unit.
    • We still plan on ordering our own GPS unit.
[edit]

Sat, Feb 18, 2006

  • ALL : So far no one has paid Robojackets dues. Please bring $35.00 in cash to this meeting! I (Logan) will be taking up dues there.
  • David: Label all power cords & network cords that the server directly/indirectly depends on
  • David: Review all code that controls the CAN bus. If the CAN-related hardware is in a sufficiently complete state (which mostly depends on Andy and Logan), I will also begin testing the actual CAN hardware.
  • David: Plan/architect/design the navigation system for the production codebase.
    • Difficult as this will probably involve:
      • Providing a switch (either in software, hardware, or both) to select between two autonomous behaviors, one for the Navigation Challenge (which has already been written, but not tested) and the other for the Obstacle Course (which needs to be designed, written, and tested).
      • Designing the autonomous behavior for the Obstacle Course virtually from scratch.

Note: I was still not able to get a P-card from SGA. So this meeting, as David recommended, will mostly be about understanding software and possibly testing whatever can be tested. --Logan

[edit]

Future

  • It seems that my schedule has become quite busy this semester, so I probably won't be updating this page for a while. Email me, Spencer, or Logan for information about our upcoming meetings. DavidF 19:26, 10 Feb 2006 (EST)
[edit]

Sat, Jan 14, 2006

[edit]

Sat, Jan. 7, 2006

US FIRST Kickoff at S. Polytechnic. IGVC meeting??? I guess not --AndyB 19:57, 7 Jan 2006 (EST)

[edit]

Sat, Nov. 26, 2005 - No Meeting!

Since Nov. 26 is during the Thanksgiving Break, no meeting will be held at this time.

[edit]

Sat, Nov. 19, 2005

What's the deal? Nobody showed up! --David 15:24, 20 Nov 2005 (EST)

[edit]

Before the meeting

[edit]

Items to bring to the meeting

  • Ben: Bring footage from last year's IGVC competition - (on DVD?)
[edit]

During the meeting

  • Issue: Not enough work is getting done - take action!
    • Consider making future meetings biweekly instead of weekly
    • Role allocation for IGVC members
  • David: Demo Hough transform for detecting lines
  • Michael: Finish setting up the permanent IGVC server
  • Review/post footage from Ben (assuming he brings the DVD this time)
[edit]

Other Items

  • Andy: Plans to bring 1) a router and 2) CAT-5 cable on Mon, Nov. 21, 2005.
[edit]

Sat, Nov. 12, 2005

This agenda was put together by David. Feel free to make modifications as necessary.

[edit]

Setup the permanent IGVC webserver

[edit]

To Do

  • Install software and copy data files from the temporary webserver
    1. Install Linux
    2. Install MySQL
    3. Install MediaWiki
    4. Flush the SQL tables that the MediaWiki setup generated
    5. Dump the SQL tables from the temporary IGVC server to file and import it into the SQL database on the permanent server
    6. Copy the MediaWiki configuration files from the temporary IGVC server to the permanent server
  • Test that everything works - both locally and remotely
[edit]

Outstanding Issues

  • Issue: Not enough work is getting done - take action!
  • Make sure the permanent server is configured properly
    • Setup a firewall to block connections to unnecessary TCP/UDP ports
    • Enable BIOS setting for automatic restart upon power failure
    • Make sure all configuration settings for the server (like the MySQL admin account credentials and the OS admin account credentials) are all recorded in a safe place that is not likely to be forgotten
    • Label the server, the cords connected to the server (especially its power source and network cabling), the router/switch connected to the server: DO NOT UNPLUG! THIS IS THE SERVER.
  • Making the permanent server accessible from the web
    • We have a couple of options:
      1. Request a new static IP for the server from OIT
      2. Share the static IP that has been allocated to the adjacent computer
        • At the moment, the closest network routing device is a switch, which is not capable of orchestrating the sharing of the existing static IP. Therefore in order to share the IP, we would have to buy a router that is capable of NAT (but not capable of wireless) to replace the switch.
      3. Just hijack the IP for the adjacent computer - probably not an acceptable choice
  • Implementing a backup strategy for the permanent server
    • Write an automated shell script that occasionally backs up the MediaWiki configuration files as well as the contents of the MySQL database to another location, such as:
      1. a separate hard drive
        • This would of course require us to purchase and install another hard drive into the server.
      2. an off-site location
        • This location would need to be available 24/7 in order for the backups to be automated reliably. Perhaps a commercial webserver would be a good location?
[edit]

Review footage from past IGVC years

Ben should be bringing a DVD containing IGVC footage to the meeting. In particular it would be a good idea to review footage containing lines on grass, as we are currently don't have any footage of that.

It might be a good idea to select some of the footage to upload to the IGVC wiki so that us image-filter monkeys have something to work with.

[edit]

Find the camera

The last I checked, we had somehow misplaced our camera. Naturally we need to find it.

The usual locations of the camera should be checked and then suspected areas assessed.

The camera (and laptop) have been found.

[edit]

Review Ben's code

If all goes well, I (David) should have a summary of Ben's code from last year posted on the Wiki by the time of the meeting. We need to review the code and start writing our own codebase, perhaps using Ben's code as a model.

[edit]

Miscellaneous

We will need implementations (not just designs) of algorithms for detecting lines on grass. In order to facilitate the development of such algorithms, it would be a good idea to upload some footage of white/yellow lines on grass so that the algorithm implementers have something to work with.

Retrieved from "http://wiki.robojackets.org/w/Misc._IGVC_Archives"
Personal tools