First steps to autonomous outdoor flight...

-
Some of you might already suspect that I am trying to make my tricopter fly autonomously not only indoors, but also outdoors. I got some parts as a present from watterott.com (GPS & pressure sensor) and I bought an aditional compass sensor. First of all, I had a look at the GPS. It is a 10Hz, 65 channel SUP500F receiver (SUP500F). Results were a bit disappointing:

 I am standing still during logging (~1 minute), with a free view to the sky and no clouds. Next, I tested the pressure sensor (Bosch BMP-085). I was just using some very simple low-pass filters, and results were pretty ok already:

The sensor was moved several times from 0 to 2.5 meters. The compass (HMC6352) also works pretty ok. It delivers very precise heading information, even when the motors are running. But, when the sensor is tilted, there is an error in heading (due to the principle of measuring). There are much more expensive sensors available that can correct for tilt, but that is not really a problem I think: In position hold, the tilt angle will always be pretty low. When doing waypoint navigation, the heading can be read from GPS, hence heading data can be calculated from complementary filtering of gyro/ GPS/ compass if necessary. Here is a extremely boring video of my DLX with pressure sensor and compass active. From second 20 till the end of the video, I am not touching throttle or yaw anymore:


Finally, I had the opportunity to test a uBlox LEA4 GPS receiver from a friend of mine. On the first picture, I was standing still for ~1 minute, on the second picture I was walking around in several "8's". This GPS receiver seems to be pretty awesome, that's why I instantly ordered one. I am pretty confident that position-hold will work out with this sensor.

GPS, pressure sensor & compass are connected to an extra microcontroller which does all the processing. It sends a small byte array (4 bytes containing throttle, yaw, roll & pitch correction data) to the TriGUIDE via UART. The correction data is forwarded directly to the motors. Like this, modifications on the TriGUIDE firmware are minimal, and processing speed doesn't suffer.

Comments

  1. Peter14 June 2010 at 05:40

    Very nice, i will build my self a tri when i have time.. One thing i have thought about is if it would be possible to have two gps, one on the aircraft and one on the ground as reference to calculate out the errors to get cm accuracy. If you have access to two uBlox LEA4 GPS, you could do a test, and log them to see if they produce the same drift, if so it would be possible.. I am very curious, it might be that they do some fancy interpolation calculations inside, and do not produce the same position...

    ReplyDelete
  2. William Thielicke14 June 2010 at 08:21

    Hi, I think for d-GPS to work, both receivers need to have exactly the same conditions (satellites, time etc.), and that will never be the case unless special algorithms are used inside the GPS modules...

    ReplyDelete
  3. tensor14 June 2010 at 13:33

    Hey William. Your height control looks really stable. With the GPS I've experienced similar problems. Unfortunately it is not the quality of the receiver which makes it suitable for a control task but the settings of the internal filter. I compared a expensive FV-M8 and also the LEA4 by walking on a stone circle:
    http://tinyurl.com/GPS-FV-M8
    http://tinyurl.com/GPS-lea4
    The main advantage of the LEA4 is that you can configure the internal filter to suit our control task. I'm looking forward to your first position hold tests.
    ~Stefan

    ReplyDelete
  4. VANT tec blog24 June 2010 at 00:34

    Hello Willian ... you are the master !
    i making my own navi code too.
    i am using a ublox4 and atmega 168
    i started my project this week.
    can i talk with you about the calculation to auto pilot the uavp ?

    regards Eduardo ...

    ReplyDelete
  5. William Thielicke24 June 2010 at 20:10

    Hi Eduardo,
    yes we can discuss some things if you like. How about if you write me an email with the components you use and your ideas. My code is almost finished (but not working yet ;-D), I can also tell you my ideas.

    ReplyDelete

Post a Comment

Popular posts from this blog

My 15 inch long-range setup for 2020 (with CAD files)

-
Image
This is my latest 15 inch ultra-long range setup with 50 minutes flight time: Propellers: 15x5.5" Tarot with custom hub  Motors: T-Motor MN3110 470 kV  Frame: 240g custom carbon structure  Lipo: 2x 4S 4000 mAh SLS 20C (parallel)  Camera: Micro Runcam micro sparrow  Firmware: INAV  FC: Holybro Kakute F7  ESC: Tekko32 F3  GPS: BN-180  Compass: QMC5883L  Receiver: FRSky R9 slim+  Hovering flight time is 50 minutes at 8.5 Amps and 40% throttle. Current goes up to 12 A when flying 50-60 km/h. Flight characteristics are really good when tuned properly (you'll nee a HUGE D-term). Here is the CAD file of the frame. INAV 2.4.0 PID settings:  set mc_p_pitch = 85 set mc_i_pitch = 60 set mc_d_pitch = 150 set mc_p_roll = 85 set mc_i_roll = 60 set mc_d_roll = 150 set mc_p_yaw = 200 set mc_i_yaw = 70 set mc_p_level = 40 set mc_i_level = 10 set max_angle_inclination_rll = 500 set max_angle_inclination_pit = 500 set dterm_...
Read more

Finished I²C to PWM converter and TriGUIDE mini

-
Image
The last nights were a bit shorter for me, since the PCBs I ordered at bilex-lp.com arrived. They look really good and seem to be of high quality (I ordered a different "solder resist" colour, well, I like bright colors, but anyway... ). So I had the opportunity to do my very first SMD solderings. This really seems to be comfortable... I will completely switch to SMD I think. The I²C to PWM converters work! Here are the features: Weight under 1 g Works with (most likely many) standard ESCs (currently tested: HK SS 18-20A, ...) I²C address selectable via solder jumpers (4 standard mikrokopter addresses dec 82, 84, 86, 88) Refresh rate selectable via solder jumpers (417 Hz, 292 Hz, 155 Hz, 49 Hz) Pulse width (I²C 0-255): 990µS - 2010µS Pulse width at startup (3 s): 920µS Motor off when no I²C connection longer than 256ms Motor off when microcontroller crashes (veeeery unlikely) Powered by the BEC of the ESC, or by an external 5V source (prepared as well) Very eas...
Read more

Aerodynamics in racing multirotors!

-
Image
Why future racing copters really should look different. by Dr. William Thielicke aka Willa aka Shrediquette ABSTRACT In this article I try to demonstrate why FPV racing multirotors need to look different. Some small modifications to the frame would (in theory…!) result in 70 % higher top speed! All that needs to be done is to align the arms parallel to the propeller flow, and to tilt the main body of the copter by about 40 degrees. I am presenting a very simple and robust racing copter design that incorporates these ideas. Furthermore, I am calculating the aerodynamic drag of different copter concepts using basic equations. The aim of this article is to make you realize the importance of aerodynamics and to stimulate people to design more innovative racing frames. INTRODUCTION Until recently, multirotors were mainly used as a “hovering device” and the top speed of these copters did hardly matter. Now, multirotor racing has become popular and all competitors are seeking ...
Read more

More propeller tests

-
Image
Recently, I discovered some small counter-rotating propellers that I didn't know before. I ordered some and did some tests. The red 4x4.5" props look really good, nice thin airfoils and a good precision. During the tests, they made the least amount of noise, and are currently my preferred ones. The performance of the 4x2.5" props is even better, but they have a pretty terrible sound and seem to be less balanced. Both propellers outperform the AirAce propellers, but this is not really a surprise because the AirAce have a smaller diameter and more blades. What is also interesting is the weight of the propellers. A propeller with a lower weight potentially has a lower moment of inertia which is beneficial for the control loops: AirAce: 2.1 g 4x2.5": 1.5 g 4x4.5" : 1.3 g So this is an advantage for the 4x4.5" propellers, especially because the weight of this propeller is more concentrated to the center, which should further lower the momentof inertia. ...
Read more

Part 1/3: Side Force Generators (SFGs) in FPV Racing

-
Image
Summary: FPV racing multirotors with side force generators (SFGs), will drift less during turns, making them potentially faster on race tracks  (watch the video comparison here ) . But the SFGs need the be carefully placed and dimensioned, so that aerodynamic centre and centre of mass are at the same position. Otherwise it will not work.  Introduction Recently, I was thinking about how to improve the handling of racing copters. I noticed since quite a long time, that different copters behave quite differently in fast turns: Some drift quite a lot (especially when they carry a larger battery or an action cam), others drift less. I was always preferring copters that drift as little as  possible. That is also why I am trying to reduce the weight of my copters as much as feasible. The amount of drift depends on the weight of the copter and the lift and drag of the fuselage for sideways air streams (side-slip). The larger the lift and drag, the more side force the fusel...
Read more