JC's Posts (3)

Sort by

Project PropFIT

Propeller Flying Inertial TestbedGoal:To develop and refine software that will read sensor values and perform the necessary calculations to produce a stable and accurate IMU (inertial measurement unit).Performance:The quality of the code will be evaluated on how well a simulated horizon aligns with the actual horizon. Using the Propellers video drivers, an artificial horizon will be displayed over a real world view to see how well they align.Features:The Parallax Propeller’s architecture allows it to run multiple IMU routines simultaneously (true parallel processing) so software refinements can be compared to a baseline routine. For example a floating point kalman filter can be directly compared to an integer kalman filter to see which performs better. Or the same filter can be run at higher frequencies to see if performance improves.Hardware:Airframe – HobbyLobby Wingo Had one too many crashesAirftame - Easy StarMicroprocessor – Parallax PropellerIMU – Sparkfun 5DoFCamera – Inteligent Flight KX171Here is a partially assembled board used during testing. There were a few layout problems that have been fixed.

Board_Pics_3-09%20010_web.JPG

The underside of the board with the 5DoF IMU. The camrea is only mounted to the board for testing.

Board_Pics_3-09%20011_web.JPG

The project home is herehttp://code.google.com/p/propfit/

here is a picture of the boardPropFIT1.1.brd and the eagle file.
Read more…

Another Parallax Propeller Autopilot

Introducing the "OughtToPilot"

Default.aspx

OughtToPilot-Project-L.jpg

It seems the parallel processing ability of the prop makes it a top candidate for autopilot brains. This project was entered in the Parallax Design Contest and won second place!! The designer opted for a fuzzy-logic algorithm over a kalman filter for the IMU.The OughtToPilot employs a non-linear fuzzy-logic algorithm to blend two different types of inertial sensors (accelerometers and gyros) in a computationally efficient manner. This model free blending and estimation method was chosen over Kalman filtering to reduce the design-side system modeling effort and the real-time computational burden. All computations are conducted upon integer based pseudo float numbers with most values corresponding to a fractional number with three digits after the decimal point, multiplied by a thousand.The source code and material list is available from the link above.
Read more…

iHUD

iHUD-main-motion-blur.jpg

iHUD-in-Isuzu-Rodeo-wit-arrows.jpg

iHUD-in-Larr%27s-cockpit-cropped-II.jpg

ihudFrom the website:iHUD is an application (app) that turns the iPhone and - with limited features the iPod touch *) - into an aerospace-inspired mobile Glass Cockpit. iHUD derives its name from Head-Up Display, which depicts motion and flight-related pertinent guidance information and data for optimal situational awareness.iHUD depicts an extraordinary graphic interface with a simulated horizon and a vehicle reference symbol, dynamic speed, altitude, and vertical velocity ribbons and digital display window, rotating compass card with user selectable heading bug, slip/skid ball, and an accelerometer (G-meter).WOW that alone is worth getting an iphone
Read more…