* wip torqued
* add basic logic
* setup in manager
* check sanity and publish msg
* add first order filter to outputs
* wire up controlsd, and update gains
* rename intercept to offset
* add cloudlog, live values are not updated
* fix bugs, do not reset points for now
* fix crashes
* rename to main
* fix bugs, works offline
* fix float in cereal bug
* add latacc filter
* randomly choose points, approx for iid
* add variable decay
* local param to capnp instead of dict
* verify works in replay
* use torqued output in controlsd
* use in controlsd; use points from past routes
* controlsd bugfix
* filter before updating gains, needs to be replaced
* save all points to ensure smooth transition across routes, revert friction factor to 1.5
* add filters to prevent noisy low-speed data points; improve fit sanity
* add engaged buffer
* revert lat_acc thresh
* use paramsd realtime process config
* make latacc-to-torque generic, and overrideable
* move freq to 4Hz, avoid storing in np.array, don't publish points in the message
* float instead of np
* remove constant while storing pts
* rename slope, offset to lat_accet_factor, offset
* resolve issues
* use camelcase in all capnp params
* use camelcase everywhere
* reduce latacc threshold or sanity, add car_sane todo, save points properly
* add and check tag
* write param to disk at end of route
* remove args
* rebase op, cereal
* save on exit
* restore default handler
* cpu usage check
* add to process replay
* handle reset better, reduce unnecessary computation
* always publish raw values - useful for debug
* regen routes
* update refs
* checks on cache restore
* check tuning vals too
* clean that up
* reduce cpu usage
* reduce cpu usage by 75%
* cleanup
* optimize further
* handle reset condition better, don't put points in init, use only in corolla
* bump cereal after rebasing
* update refs
* Update common/params.cc
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
* remove unnecessary checks
* Update RELEASES.md
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
* lateral overriding is overriding
* Update test
* remove
* also could do something like this and only have one OVERRIDE ET
* Revert "also could do something like this and only have one OVERRIDE ET"
This reverts commit 5c381641c0.
* full names
* bump cereal
* test every event type
* update refs
* single longitudinal toggle for e2e and disable radar
* write disable radar param
* rename
* better param name
* clean that up
* update refs
* update translations
* not live for experimental
* write it out
* vanish
* see if this works at all
* can revert this
* think adding a no entry conditionally is nicer
* then we can revert this
* 0 makes more sense
* Revert "0 makes more sense"
This reverts commit efc89e8a23.
* gm CC uses > 70
* bump cereal
* comment
* test on Honda
* whoops
* works
* add exception with todo
* moved button enable events to controlsd
* get rid of that
get rid of that
* different values for now
* car interfaces add enable event, controlsd can block it
* Regen and update refs
* delete if not set
* One place
one place
* regen routes are uninitialized first few frames
* Trim start of segment so it's like original segment
* stash
* regen
* not working
* clean up
* more cleanup
* revert
* bump ceral
* actually check resume button
* whoops
* pcmCruise cars don't use setSpeed, so we're good
* engage correctly in sim
* Update ref_commit
* Update refs
* push
* do process replay
* commit
* update refs
* clean up
* clean up controlsd
* clean up controlsd
* Add assert
* debubuggier param name
* can be peristent
* Revert "can be peristent"
This reverts commit 6e6d3f6423.
* Update selfdrive/test/process_replay/process_replay.py
* sensor flag in paramsd
* increase lateral acc err alert threshold
* add invalid sensor alert from paramsd
* update ref, bool before adding to msg
* account for sign difference in locationd and paramsd
* revert ref
* always log leads, we hide them in ui
* only spam resume when future is > vEgoStarting
* do rest but vw
* vw
* remove comments
* rename to resume
* maintain original button msg rate
* mazda: ensure no resume if cancelling
* same for non-HDA2
* Always run planner if not opLong
* try 0.2
* 0.1 should be pretty safe
* add test for resuming
* fix test
* stricter test, speeds[-1] is 0.14 when starting here
* no walrus
* fixup mazda cc
* remove extra import
* set while gas pressed to resume
* better order
* what was I thinking?
* Update selfdrive/controls/lib/drive_helpers.py
Co-authored-by: Gijs Koning <gijs-koning@live.nl>
* Update selfdrive/controls/lib/drive_helpers.py
Co-authored-by: Gijs Koning <gijs-koning@live.nl>
* only if under set speed
* some clean up and ciel the v_cruise_kph
* clean up
* actually floor
* Should be cruise min
* On phone, will clean up
* Think no floor is fine
* clean up and catch setCruise
* update comment and just always do rounding and clipping
update comment
* flip order
Co-authored-by: Gijs Koning <gijs-koning@live.nl>
* Initial commit
* Fix bugs
* Need more torque rate
* Cleanup cray cray control
* Write nicely
* Chiiil
* Not relevant for cray cray control
* Do some logging
* Seems like it has more torque than I thought
* Bit more feedforward
* Tune change
* Retune
* Retune
* Little more chill
* Add coroll
* Add corolla
* Give craycray a good name
* Update to proper logging
* D to the PI
* Should be in radians
* Add d
* Start oscillations
* Add D term
* Only change torque rate limits for new tune
* Add d logging
* Should be enough
* Wrong sign in D
* Downtune a little
* Needed to prevent faults
* Add lqr rav4 to tune
* Try derivative again
* Data based retune
* Data based retune
* add friction compensation
* Doesnt need too much P with friction comp
* remove lqr
* Remove kd
* Fix tests
* fix tests
* Too much error
* Get roll induced error under 1cm/deg
* Too much jitter
* Do roll comp
* Add ki
* Final update
* Update refs
* Cleanup latcontrol_torque a little more
* control with joystick
* slow it down
* always joystick for not car
* clean ups
* not stateful
* move submaster
* only if we aren't in test mode
* update refs
* double ki speed, update ref
* this ref
Co-authored-by: Comma Device <device@comma.ai>
Lee Jong Mun
Adeeb Shihadeh
Vivek Aithal
Shane Smiskol
HaraldSchafer
Robbe Derks
Willem Melching
grekiki
George Hotz
George Hotz
Lukas Petersson