Update gas_interceptor_threshold due to experiments wth RDX and Odyssey - #117
Update gas_interceptor_threshold due to experiments wth RDX and Odyssey#117joeljacobs wants to merge 1 commit into
Conversation
|
This PR goes along with My Pedal OP PR |
|
|
||
| // these are set in the Honda safety hooks...this is the wrong place | ||
| const int gas_interceptor_threshold = 328; | ||
| const int gas_interceptor_threshold = 800; |
There was a problem hiding this comment.
As we were discussing on the other PR, this should be 120/0.2539 (+1 to be slightly conservative and not false trigger panda cancellations before openpilot does) in order to be consistent with opendbc and openpilot. If this threshold does not work, we should figure out why.
There was a problem hiding this comment.
I know that 600 did not work, so 472 would surely not work. If I set it higher, occasional momentary gas signal spikes (either a-to-d or physical pedal anomaly) will cause OP to throw a "controls mismatch" error.
However, leaving the dbc to -120 doesn't seem to cause OP disengagements with the same spikes. If I went lower than 120, it would completely disengage (which was especially dangerous when braking is involved.)
I don't know why it seems the dbc is more forgiving.
There was a problem hiding this comment.
I should note that it does a good job knowing it should disengage when I press on the gas with these settings.
There was a problem hiding this comment.
"I don't know why it seems the dbc is more forgiving." <-- yeah... this needs to be figured out before we can merge changes.
There was a problem hiding this comment.
Could it have anything to do with the GAS2 value? If you divide 120 by 0.126992032, you get a much closer value of 944.94.
3 participants
This new threshold value was arrived at through experimentation with the RDX and Odyssey. Values much lower than this would cause OP to suddenly disengage due to anomolies with the Gas Pedal analog readings.