* mse err from 0.028070712 -> 5.8073703e-05
* build with weights fixup
* need thneed lib also
* don't break for binaries
* static analysis says i need init
* check the bias
* load_dlc_weights
* nicer scons
* tested scons
* fix static
* pylint issue
* new ref
* a few more asserts
Co-authored-by: Harald Schafer <harald.the.engineer@gmail.com>
old-commit-hash: 8d6f49aecf
* update filters only when all messages are alivbe and valid
* update message valid and fix unit test
* update refs
* move check outside loop
* modify fake message fn in test
* deprecate inputsOK and resolve PR comments
* avoid double looped list comprehension
* follow import conventions
* modify paramsd valid to only be invalid in case of commIssue
* update refs
old-commit-hash: 7e6903b58f
* Fix bigmodel on tici: cam intrinsics arent static
* Give TICI route own name for import
* Make function useable in offline test
* typo
* change tici ecam focal to trained focal
* No shadow variable
* Fixed ref
* Force update this time
* Update model replay ref commit
old-commit-hash: 48154703a7
* disable change cost completely on standstill and gas press
* cleanup
* set accel to zero
* clean up logic around standstill
* update ref
old-commit-hash: 428d412c1f
* Added wide cam vipc client and bigmodel transform logic
* Added wide_frame to ModelState, should still work normally
* Refactored image input into addImage method, should still work normally
* Updated thneed/compile.cc
* Bigmodel, untested: 44f83118-b375-4d4c-ae12-2017124f0cf4/200
* Have to initialize extra buffer in SNPEModel
* Default paramater value in the wrong place I think
* Move USE_EXTRA to SConscript
* New model: 6c34d59a-acc3-4877-84bd-904c10745ba6/250
* move use extra check to runtime, not on C2
* this is always true
* more C2 checks
* log if frames are out of sync
* more logging on no frame
* store in pointer
* print sof
* add sync logic
* log based on sof difference as well
* keep both models
* less assumptions
* define above thneed
* typo
* simplify
* no need for second client is main is already wide
* more comments update
* no optional reference
* more logging to debug lags
* add to release files
* both defines
* New model: 6831a77f-2574-4bfb-8077-79b0972a2771/950
* Path offset no longer relevant
* Remove duplicate execute
* Moved bigmodel back to big_supercombo.dlc
* add wide vipc stream
* Tici must be tici
* Needs state too
* add wide cam support to model replay
* handle syncing better
* ugh, c2
* print that
* handle ecam lag
* skip first one
* so close
* update refs
Co-authored-by: mitchellgoffpc <mitchellgoffpc@gmail.com>
Co-authored-by: Harald Schafer <harald.the.engineer@gmail.com>
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
Co-authored-by: Comma Device <device@comma.ai>
old-commit-hash: 85efde269d
* add thneed optimizer
* local work group opt
* kernels and final mods
* release files
* build system touchups
* fix kernel path, rand inputs for self test
* broken since extra is gone
* update model replay ref
Co-authored-by: Comma Device <device@comma.ai>
old-commit-hash: 90beaebefb
* or brake_switch
* remove car exceptions
* Revert "remove car exceptions"
This reverts commit cad6a552aa4a17c69616014d3e9333d30c1fadd3.
* does have ts noise, use updated
* bump panda for tests
* bump
* messsssy draft
* clean up
* clean up
* bump
* CS: process all messages at once, like we do during real openpilot usage
* handle multiple messages updated in one cycle
* bump
* bump
* use less confusing, but equivalent cp.vl
* Update selfdrive/test/test_models.py
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
* no msg
* bump opendbc
* Update selfdrive/car/honda/carstate.py
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
* same test logic
* add brake_switch_prev to class
* need to set previous brake switch when not lagging (most of the time)
* need to compare
* only update previous if updated this cycle
rm
* need to maintain previous active state
* move brake_switch
* bump opendbc
* bump
* bump
* bump
* bump
* bump
* bump
* bump opendbc
* rename
* oops
* bump opendbc
* bump
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
old-commit-hash: 209bfe536b
* base LatControl class, move sat check out of pid.py
clean up
clean up
* fix
* global variable for min control speed
* nicer name
* unify latcontrol class init arguments
* add to release files
* saturated if close to limit
* move angle mode saturation checks into class
* check_saturation function takes in current saturated status
undo
* apply latcontrol_angle's active checking to all controllers
* clean up
* move those back
* make abstract baseclass
* add test for saturation
* keep clip
* update ref
* fix static analysis
Co-authored-by: Willem Melching <willem.melching@gmail.com>
old-commit-hash: 9de8f8cd8c
* hacks to make process replay engage again
* dont change that
* enable engagement check again
* chmod +x
* first working regen
* proper logMonoTime
* fix video framerate consistency
* mpc is valid again
* proper alive checks
* revert loggerd change
* ensure engaged in regen, fix managerState, peripheralState
* ubloxRaw is unused
* add submaster config for gm
* regen all services we can run without HW
* fix loggerd
* loggerd: matroska without the extension
* update ref
old-commit-hash: d3d10e014a
* lat_mpc: make v_ego, rotation_radius parameters instead of states
* lat_mpc: remove rotation_radius argument, since it is part of the parameters
* lat_mpc: use qp_solver_cond_N = 1
slightly faster and in line with case study in Fig. 2/ 3 in Frison2016 - https://cdn.syscop.de/publications/Frison2016.pdf
An Efficient Implementation of Partial Condensing for Nonlinear Model Predictive Control
* adapt test_lateral_mpc to formulation with parameters
* lat_mpc: set parameters in reset() and copy values
* acados_ocp_solver_pyx: make options_set useable
* update ref
Co-authored-by: Willem Melching <willem.melching@gmail.com>
old-commit-hash: 0681474840
