85375c2201
Rename the FMU->IO output controls to reflect the fact that they are controls, not servo values.
12 years ago
b8250de1e6
Assorted compile fixes.
12 years ago
d81edb09cf
whitespace/formatting
12 years ago
f9520ee39d
Factory method for a simple mixer that converts PWM/PPM values to the standard internal format.
12 years ago
0ae5997bd0
Fix some scaling errors in the PWM <-> mixer-internal conversions.
12 years ago
6b9d5dac4d
Rough in the new mixer path for PX4IO.
12 years ago
6ede0e2f18
Add the ability to reset a mixer group. Report the remaining buffer size from load_from_buf.
12 years ago
d5da457e29
Fix PX4IO to run C++ static ctors
12 years ago
35c82ff2fc
Make mixer ioctls load from a memory buffer rather than a file. This is prep for uploading the memory buffer to IO to be processed there.
12 years ago
f0da789626
Remove the unused complex-multirotor setup ioctl, since it's not implemented anywhere.
12 years ago
a8451a2d18
Update apps/drivers/mpu6000/mpu6000.cpp
...
Don't call ::close on ORB publication handles.
12 years ago
a2aa9dd8fd
Made MPU-6000 gyro optional
12 years ago
76895af6eb
Fix several aspects of the PWM output driver; enable auto-reload, use named constants for various control bits, and use a more polite mechanism at disarm time to avoid runt pulses.
...
This may address an issue we've seen where we get occasional malformed PWM output pulses, possibly due to a race between compare updates and the timer.
12 years ago
8d716dea45
Teach 'fake' to set the arming state as well.
...
Whitespace.
12 years ago
b8044d9786
use <err.h> more consistently in the fmu driver.
12 years ago
6d0bea0298
Fix the PWM servo ioctl base so it's not overlapping the GPIOs
12 years ago
03b51c69e0
Added more LED state logic and improve code.
...
The LED will now also indicate when the FMU is ARMED. Switched to using
a 16-bit value where each bit indicates what state the LED should be
in.
12 years ago
197e573885
Add an additional safety switch LED blink sequence when FMU and IO are armed
...
If both the FMU and the IO board are armed then the secure switch will
blink two times quickly then a pause followed by two quick blinks and
so on.
12 years ago
82cbac70ee
Fixed calibration check
12 years ago
fd771f67f2
Adjust the control mapping from DSM receivers to correspond to the standard PPM control mapping for channels 0-4.
12 years ago
7c3b28d503
Lock out the PPM decoder if the DSM or S.bus decoders have seen a frame recently.
12 years ago
1485a4ec1a
Fix breakage to the DSM parser introduced with the input prioritisation logic. Back out to a "any input wins" strategy; connecting multiple receivers to I/O at the same time is currently not supported (read: strange things will happen).
12 years ago
6e328b4d7a
Add a 'monitor' verb to the px4io command so we can watch inputs to IO (it could get smarter).
12 years ago
451ecc1bf4
Remove a few cut-and-paste author attributions.
12 years ago
269bd9f403
Force the fade speed to something sensible. Deal better with failed probes.
12 years ago
2a8ef50df4
A driver and shell command for the BlinkM I2C LED controller.
12 years ago
de88732e8e
Prevented unhealthy RC input from propagating through the system
12 years ago
efd3b9dea6
Clean up the FMU communications init.
12 years ago
7d9d307ab0
We don't need non-blocking I/O for this context anymore; it's OK for it to block.
12 years ago
8c4e9de70a
Use the right constraint for the output mixer; we might end up wanting more virtual control channels.
12 years ago
1e6e06595a
Avoid processing S.bus channels that cannot be communicated to FMU
12 years ago
2ac0cac11f
Build fix - need <termios.h>
12 years ago
6f572637de
Fixed missing heading for mtkcustom and nmea
12 years ago
ef4a54666d
Harmonized PPM, S.BUS and DSM input (order: first preference S.Bus, then DSM, then PPM, first available and valid source is chosen), tested with FMU, valid channel inputs
12 years ago
e0df7e6a76
save ~200 bytes of RAM by correctly positioning the S.bus decoder table in flash.
12 years ago
d16d66f990
Enabled UART3, added JTAG make target for IO, removed potentially problematic usleep
12 years ago
c961dd8bab
Just for fun, add a (completely untested) S.bus decoder.
12 years ago
9fa794a8fa
Rework the PX4IO software architecture:
...
- Use a separate thread for handing R/C inputs and outputs.
- Remove all PX4IO R/C receiver configuration; it's all automatic now.
- Rework the main loop, dedicate it to PX4FMU communications after startup.
- Fix several issues in the px4io driver that would cause a crash if PX4IO was not responding.
12 years ago
9c8101d4f1
Add some more information to comments.
12 years ago
d0efd1a419
Fix the DSM (spektrum) protocol decoder, and add some format auto-detection to it.
12 years ago
3321ca0888
Don't print the status line so much; we seem to drop a lot of receive characters this way.
12 years ago
03a82e0a03
Fix includes for debug output.
12 years ago
92e1d5eb78
Possible fix for #78 - increase the wait timeout for discard when flashing PX4IO. It's not clear this solves the issue, but I can't reproduce it with this added.
12 years ago
15236d1ff2
Hotfix: calibration status returns sanity checks with better granularity
12 years ago
caf0fefa32
Cleaned up different uorb topics, cleaned up excessive stack sizes
12 years ago
4d6fb3a037
Added delay test to measure comm delays with a led / scope
12 years ago
98657b8ae4
Added rates setpoints as system outputs
12 years ago
11b0242f55
Not arming FMU in HIL mode
12 years ago
b757ebbda8
Not arming FMU in HIL mode
12 years ago
4298f68fcd
IO does not arm in HIL mode
12 years ago
px4dev
Lorenz Meier
Simon Wilks
Julian Oes