March 2007 Archives

Today Holger coupled his Microcopter with a GPS Sensor. He did this in a quite simple way and the used closed-loop control will swing up. That doesn't mean it's not impressive! It already shows how it would work. Please note how often Holger is not controlling the Microcopter! Long times it flys autonomiously!

May I present Holger's show:

Get the Flash Plugin to see this movie.

Direct link: Download the FLV Movie

Hammer's power UAVP

| | Comments (0) | TrackBacks (0)

Hammer, another UAVP enthusiast, released a new movie today! It shows his new UAVP setup using Hacker A20-L20 motors. It seems that Hammer's UAVP has won quite some agility and power! The sound of the motors is quite impressive too.

The movie shows quite good how powerful these motors are.

Get the Flash Plugin to see this movie.

Direct link: Download the FLV Movie

I already talked about different implementations of UAVPs. One of them, the Wolferl-Board, is a UAVP built by Wolfang Mahringer, one of the pioneers of QuadCopters. Wolfgang succeeded in building a successful flying UAVP closed-loop control already back in January 2006.

The following picture shows Hammer's QuadCopter using a Wolferl-Board:

Wolfgang continuously extended and improved his board design troughout the last year and it's now a very reliable base platform for building an UAVP. Many successful flying UAVP are built usinging a Wolferl-Board as base.

The day before yesterday I wrote a mail to Wolfgang asking if I could receive a copy of the software which runs on a PIC microcontroller in the center of the board. I could not belive it, but already some hours later I received the source code (version 3.05) from Wolfgang!

Thank you Wolfgang! For the great work you did, for the source code, for making it Open Source and available to everyone!

The big black chip in the center is the 8bit microcontroller PIC 16F876 built by the company Microchip. Note that it has a 8bit data bus but 12bit code and address bus which is a bit special.

The PIC16F876A features 256 bytes of EEPROM data memory, self programming, an ICD, 2 Comparators, 5 channels of 10-bit Analog-to-Digital (A/D) converter, 2 capture/compare/PWM functions, the synchronous serial port can be configured as either 3-wire Serial Peripheral Interface (SPI) or the 2-wire Inter-Integrated Circuit (I2C) bus and a Universal Asynchronous Receiver Transmitter (USART). It can be programmed in C and all needed development tools are free of charge which is something which suits me very well.

Dear interested reader, this is the thingy which runs the C source code I've received from Wolfgang. The code gets compiled into a .HEX file which then can be programmed into the PIC.

Looking into the code I realized that I got quite a nice bit of software here. Wolfgang's code is good documented and has a README explaining how to setup the free IDE MPLAB and the CC5X compiler needed to generate code for the PIC. The compiler integrates nicly with the IDE.

Wolfgang also wrote a very nice PDF documentation how to build a Wolferl-Board. This document contains plans, layouts and everything needed to build a Wolferl-Board by yourself.

Digging the code I realized that the code contained a command interface implemented using the serial console on the board which I understud quite easily. Using that serial console you can send commands to the board and control certain aspects and parameters of the closed-circuit loop. Soon I realized that it would be quite easy to extend this command interface with new commands.

I decided to implement new commands to continously monitor all sensor and actor data using the serial port. Furthermore I wanted to implement another command using which the 4 motors of the QuadCopter can be remotly controlled. This is exactly what David, who is an UAVP enhusiast also, chatted about on IRC last week.

I started last night and in the morning around 0500 I could compile the code on my own box and understud how to add new commands. It's now work in progress, but I think it won't take too long to finish...

Expect more information about this project soon...

In my last article I talked about Holger's Microcopter which uses a barometric sensor to measure air pressure to stabilize it's altitute. You can imagine that air pressure only differs minimal between a place 1m higher and another place 1m lower. So the sensibility of the air pressure sensor has to be very high to be able to recognize small altitude changes.

Such high sensibility has it's pitfalls and it seems to be fun to test them out. Holger found out that opening a door while flying indoors already results in such a big air pressure change that the Microcopter would race through the floor or roof if the linear acceleration sensors would not compensate!

Holger released a new video showing his indoor experiments...

Get the Flash Plugin to see this movie.

Direct link: Download the FLV Movie

In my last article I wrote about the inherent stability problem of QuadCopters and how closed loop control of modern electronics is able to compensate and stabilize it to make it possible to fly and steer such a construct.

Flying a QuadCopter while there is a storm at first sounded impossible to me.

Holger, an inventive guy from the QuadCopter Forum - inventor of the Microcopter - was crazy enough to try and test his Microcopter in a storm. Holger's Microcopter uses barometric sensors to measure air pressure which are used to stabilize the Microcopter's altitiude.

Looking at the movie one realizes, that Holger's Microcopter is up to the challenge and Holger is able to fly it without problems:

Get the Flash Plugin to see this movie.

Direct link: Download the FLV Movie

Did you ever see a QuadCopter fly? It rocks!

A QuadCopter or UAVP (Universal Aerial Video Platform) looks like this:

uavp-ralph-8.jpg

One nearly can't belive that such a construct is able to lift into the air! Modern electronic is able to implement impressing reaction times which are needed for such an enterprise!

Such a QuadCopter needs to update the motor power 100 to 300 times per second (100 to 300Hz) to be able to fly stable. This means the electronic has to do a measurement on all sensors all 10ms to 3ms!

Imagine... such a QuadCopter construct is inherently instable!

It's hanging in the air on 4 propellers each driven by it's own motor. The construct, the motors, the propellers - neighter is exactly the same but still the electronics has to be able to compensate and to drive the motors in such a way that the whole construct stands still in the air!

To make my point, take a look at the following small movie, showing a X3D QuadCopter starting and flying some rounds above a field...

Get the Flash Plugin to see this movie.

Direct link: Download the FLV Movie

I like the idea so much, that i decided to build my own QuadCopter, preferable using a real embedded Linux system on the QuadCopter.

That would allow for a lot of new features, like monitoring the fly parameters via a shell connection using bluetooth as well as logging those parameters. Furthermore it would allow to write the control in software and exchange it with friends easily.

Expect more information and an update to my own construction plans soon!

Today Hammer released another small movie showing the agility of his UAVP. He's playing indoor with fast attitude changes which can play bad games with stability...

Get the Flash Plugin to see this movie.

Direct link: Download the FLV Movie

Archives

Communication

Syndication

Atom 1.0
RSS 2.0 GeoURL

W3C XHTML 1.0 W3C CSS Creative Commons License

get hCard

Add to Technorati Favorites
OpenID enabled
Powered by Movable Type 4.1

Del.icio.us Tags

About this Archive

This page is an archive of entries from March 2007 listed from newest to oldest.

February 2007 is the previous archive.

April 2007 is the next archive.

Find recent content on the main index or look in the archives to find all content.