* pre-enable on brake at standstill
* test by getting rid of noTarget
* Update events.py
* rename
* stash what we got so far
* cmt
* not sure if best way to fix, but enabled with current states defs is fine
* no noEntry in preEnabled
* bumpcereal
* uncomment
* Update ref_commit
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
* increase lag on prius
* better offline values computation
* split prius ecu more, set deadzone to 0
* add deadzone back
* replace deadzone with slack
* reduce deadzone to 0.2
* update refs
* add first draft enable button timeout
* use allow_enable
* it doesn't count if brakePressed
* enable on rising edge of resume (matches stock)
* not today
* add comment describing the fault this avoids
* cleaner
* handle incrementing speed
* rename
* add test for not changing speed if enabled changes mid-press
* spacey
* ugh
* bumpo
* need this to fix a fault (draft)
* already have
* this should be cancel
* fine to do for all
* fine to do for all
* bump
* bumpo
* bump to master
* Update selfdrive/car/gm/interface.py
* Update selfdrive/car/gm/interface.py
* Update ref_commit
* add regen braking field that's just used to add a pedalPressed
* bump
* bump
* Update ref_commit
* we want the standstill check
we want the standstill check
* see what diff is now
* Update ref_commit
* GM camera: use ECM brake pressed bit
* bump panda and use more reliable bit
* bump panda
* back to ECMEngineStatus
* bump
* Update selfdrive/car/gm/carstate.py
* bump panda
* Update ref_commit
* note for dsu cars
* don't hardcode dsu footnote (1/2)
* test
* revert the experimental long toggle
* another day:(
* DSU doesn't have to be toyota specific, it's a generic carParam!
* this is safe, just make sure it's unknown
* these cars do not have OP long for some reason
* UNSUPPORTED_DSU_CAR
* forgot this
* fix autoResume
* fix autoResume 2
* put gm camera voacc behind disable radar toggle
* bump panda
* bump panda
* bump panda
* experimental long
* fixes
* car control notes and fixes for Bolt EUV
* might enable stop and go
* consistent name
* min enable speed seems to be around 5 kph
* camera acc can engage under 5 kph if stopped
* comment
* comment
* remove this for now
* only real brake
* comments
* update max brake
* bump
* clean up/fix
* same if
* simplify
* fix
* old comment
* no brake_pressed
* temporary fault fix
* tune longitudinal
* update docs
* bump panda
* GM camera ACC cars have no regen in ACC
* cleaner
* cleaner
* fix
* set max gas
* fixes
* fix LKAS unavailable warning from camera
* only camera
* bump panda
* bump panda
* bump panda
* bump panda
* clean up gmcan
* clean up CC
* flip
* rm
* rm comment
* clean up
* custom starting/stopping probably not needed
* Update selfdrive/car/gm/carcontroller.py
* fix crash
* long tuning
* we need long control state to resume
* CAMERA_ACC_CAR not needed
* no interp on accel
* tuning
* formatting
* formatting
* formatting
* formatting
* formatting
* no need to init ccp
* makes more sense now
* can I kick it?
* bumpity bump bump
* bump panda
* everybody's gotta be special
* attempt at improving stopping/starting
* reduce lateral feedforward
* tweak gradient and comfort bands
* oops
* accel from the dept of redundancy dept
* bump opendbc
* jerk limiter not needed and causing problems
* don't leave stop/start flags hung when not enabled
* reduce comfort band with lead car
* borrow decel comp from HKG
* align stopping state thresholds
* (re)try letting PI clean up some of this
* bump panda
* tweaks and comments
* elide superfluous import
* VW MQB: Longitudinal prereqs
* gate this too
* bump CI
* cleanup
* more cleanup; require gateway integration
* don't run long state/actuators unless sending message
* setpoint visibility in instrument cluster
* show fixed lead car if lead visible
* hold my beer?
* tuning
* the code is darkest before the refactor
* a little more
* set freewheel and stop distance correctly
* rounding issue maybe?
* ACC_04 Charisma profiles and HUD alerts
* fix ACC_04 counter, fix gradient and accel bugs
* revert tuning change in upstream focused branch
* need 0.5m/s to leave
* reduced braking when not absolutely needed
* oops
* bump panda
* match opendbc to master
* filter ACC_13, disable secondary accel
* bump panda
* bump panda
* bump opendbc and panda
* startAccel is deprecated
* testing manual trans min engage speeds
* pass stock ACC type to TSK
* bump panda
* bump opendbc
* bump opendbc
* test hax for Brake alert
* remove ACC_13 with matching panda bump
* fix submodule refs
* update long control safety param name
* bump panda
* actuator delay corrected by Mk1 eyeball
* a little more
* tweak stop/start thresholds
* Revert "tweak stop/start thresholds"
This reverts commit 60abd57d20.
* more lag reduction
* bump opendbc
* support both analog and digital clusters
* tuning adjustments
* bump opendbc and panda
* accept two button types for resume
* support separate set and resume
* bump panda
* #24706
* partial revert 6b93ac27
* fixes
* one more
* refactor all the things
* follow experimental long refactor
* fill in ACC hud status
* temp force E2E availability
* start cleaning up
* comment touchup
* gimme some comfort baby
* experiment
* SnG me maybe?
* a little more rollout
* is this why I'm not leaving?
* turns out you need startAccel to leave
* try without ACC_04
* cleanup unused variables
* temp hack to standstill handling
* temp hack to button handling
* trim rollout distance with slower vEgoStopping
* ultra fat comfort band due to noisy E2E actuator
* I like big changes and I cannot lie
* remove lead car display for now
* cleanup, drives somewhat better
* partial FtS/basic support
* diff reduction
* not needed
* more cleanup
* oops
* more cleanup
* restore lead car visible for PQ
* bump panda
* better enable button solution
* redundant
* OP long on *all* the things
* OP long on all the things *sometimes*
* docs generator a little TOO clever
* solve starting-state a different way
* try that again
* bump panda
* oops
* follow DBC refactor
* retry CI
* signal name changed
* sync with custom enable button branch
* oops
* more follow refactor
* bump panda
* sync opendbc to master
* bump panda
* updated comments
* light P, no I
* bump panda
* bump panda again
* add openpilot long test_models route
* bump panda with replay route
* temporarily disable experimental long for merge
* bump panda to master
* resolve conflict?
* now can we go to head?
* explicit length check
* update refs
Co-authored-by: Comma Device <device@comma.ai>
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
* send at 10hz when inactive
* try to make it fine to switch rates
* fix rate
* todododo
* fine if we skip, we usually send too early
* clean up
* this may be required, 50Hz is not really needed to sync/initialize
* preserves previous behavior (not sure if this makes sense)
* Revert "preserves previous behavior (not sure if this makes sense)"
This reverts commit 3b297bca72.
* Revert "this may be required, 50Hz is not really needed to sync/initialize"
This reverts commit a6b4693814.
* rm com
* Update ref_commit
* gate behind GM Cam
* common logic
* bet
* update refs
* Use calibrator to stabilize wide transform
* Need array
* Needs init
* Needs to be arr
* publish mshg
* Check size
* Update ref
* fix calibration param read when no wide calib
* need self!
Adeeb Shihadeh
Shane Smiskol
Vivek Aithal
Jason Young
YassineYousfi
Harald Schäfer
ZwX1616
Kurt Nistelberger