I'm on the Senior Design Ongoing 03 team. We're building a computer controlled helicopter to fly waypoints, find an opening in a building, and take pictures of the inside.
The current goal is to simply get the helicopter to fly under computer control. The first step is getting it to hover.
My contributions to the team include working on the IMU, GPS, and handling the Linux administration of the PC104 board we are using for the controller.
The IMU is our acceleration sensor. It provides data in a serial format, which is then converted on the computer to usable information for the flight control software. The code was mostly completed, and my work involved devising a structure to pass data to the control program and modifying the existing code to work with this new data structure.
The GPS was an old unit that had not been tested. The unit communicates with both NMEA standard communications and a proprietary binary stream. Our first task was sending the commands to convert the unit to NMEA. Once this was accomplished, we ran the unit's self-test and discovered that it repeatedly failed the self-test. The company who created the GPS is unwilling to repair the unit without information that Lockheed-Martin has lost over the years. Until we get a replacement GPS unit, there is nothing else that can be done with this task.
The PC-104 board we are using has a Disk-On-Chip to boot from, and is running LinuxMZ. Unfortunately, the kernel used does not support the network card built into the board. I'm working on compiling a kernel for the system that will handle both the Disk-On-Chip boot device as well as the network card, and any other devices we need to support. The existing method to move files onto the board has been to use a floppy - unfortunately this does not allow large files to be moved, nor does it allow SSH connections (or telnet connections) to be made to the system for the purposes of controlling it once it's on the helicopter.
Once these tasks are completed, I will be helping handle any other Linux issues that may arise, as well as getting the wireless networking online once parts arrive for it. The plan is to have a wireless card on the helicopter, and a base station that we use to record telemetry data and send commands (load waypoints, tell it to GO!, etc) to the helicopter.
Senior Design Poster (Powerpoint, I appologize)