search

parallaxinc / Flight-Controller

Quadcopter Flight Controller
Other
24 stars 20 forks source link

Slow flight loop #36

Closed MatzElectronics closed 7 years ago

MatzElectronics commented 7 years ago

The current flight loop is 4ms - should it be increased to 5ms to accommodate the additional math required to GPS, and a few additional features?

JasonDorie commented 7 years ago

Done. Not submitted, not flight tested, but IMU appears to be working properly.

I've also reduced a bunch of the data being sent to GroundStation, including the radio channels, the yaw / pitch / roll power values, counter, and one or two others. Since the update rate is now 20% slower, it means we have 20% more time to send serial data. With the reduction in packet size, we should have more overhead for serial retries, and generally lower packet loss. It also doesn't hit the FTDI buffer thing as much either.

I figure since this version won't be back-compatible anyway, I might as well break everything at once. I bumped this to version 3.0.0.

J

On Sat, Sep 17, 2016 at 1:52 AM, Matthew Matz notifications@github.com wrote:

The current flight loop is 4ms - should it be increased to 5ms to accommodate the additional math required to GPS, and a few additional features?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/parallaxinc/Flight-Controller/issues/36, or mute the thread https://github.com/notifications/unsubscribe-auth/ANak_rEzAtgWRJj8Xw-jlT7chJcbUINuks5qq6oxgaJpZM4J_jEW .

MatzElectronics commented 7 years ago

Thanks - I figured that would be the case :)

Matthew Matz | STEM/Robotics Educator Parallax Inc. | Direct: 916-625-3019 | www.parallax.com | @M atzElectronics http://twitter.com/MatzElectronics

On Sat, Sep 17, 2016 at 5:41 PM, Jason Dorie notifications@github.com wrote:

Done. Not submitted, not flight tested, but IMU appears to be working properly.

I've also reduced a bunch of the data being sent to GroundStation, including the radio channels, the yaw / pitch / roll power values, counter, and one or two others. Since the update rate is now 20% slower, it means we have 20% more time to send serial data. With the reduction in packet size, we should have more overhead for serial retries, and generally lower packet loss. It also doesn't hit the FTDI buffer thing as much either.

I figure since this version won't be back-compatible anyway, I might as well break everything at once. I bumped this to version 3.0.0.

J

On Sat, Sep 17, 2016 at 1:52 AM, Matthew Matz notifications@github.com wrote:

The current flight loop is 4ms - should it be increased to 5ms to accommodate the additional math required to GPS, and a few additional features?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/parallaxinc/Flight-Controller/issues/36, or mute the thread https://github.com/notifications/unsubscribe-auth/ANak_rEzAtgWRJj8Xw- jlT7chJcbUINuks5qq6oxgaJpZM4J_jEW .

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/parallaxinc/Flight-Controller/issues/36#issuecomment-247816670, or mute the thread https://github.com/notifications/unsubscribe-auth/AS0quNNNwE8cUXH-44EtCLLBWmAE8U48ks5qrIi2gaJpZM4J_jEW .

JasonDorie commented 7 years ago

I've tested this on the small quad and it's working, so the big quad will be fine.