Long control: Double delay is super complicated (commaai#32694)

* Double delay is super complicated

* No more upper bound

* DEAD

* Update ref

cereal/car.capnp

@@ -490,8 +490,7 @@ struct CarParams {
   startingState @70 :Bool; # Does this car make use of special starting state
 
   steerActuatorDelay @36 :Float32; # Steering wheel actuator delay in seconds
-  longitudinalActuatorDelayLowerBound @61 :Float32; # Gas/Brake actuator delay in seconds, lower bound
-  longitudinalActuatorDelayUpperBound @58 :Float32; # Gas/Brake actuator delay in seconds, upper bound
+  longitudinalActuatorDelay @58 :Float32; # Gas/Brake actuator delay in seconds
   openpilotLongitudinalControl @37 :Bool; # is openpilot doing the longitudinal control?
   carVin @38 :Text; # VIN number queried during fingerprinting
   dashcamOnly @41: Bool;
@@ -703,4 +702,5 @@ struct CarParams {
   brakeMaxVDEPRECATED @16 :List(Float32);
   directAccelControlDEPRECATED @30 :Bool;
   maxSteeringAngleDegDEPRECATED @54 :Float32;
+  longitudinalActuatorDelayLowerBoundDEPRECATEDDEPRECATED @61 :Float32;
 }

selfdrive/car/gm/interface.py

@@ -148,7 +148,7 @@ def _get_params(ret, candidate, fingerprint, car_fw, experimental_long, docs):
 
     ret.steerLimitTimer = 0.4
     ret.radarTimeStep = 0.0667  # GM radar runs at 15Hz instead of standard 20Hz
-    ret.longitudinalActuatorDelayUpperBound = 0.5  # large delay to initially start braking
+    ret.longitudinalActuatorDelay = 0.5  # large delay to initially start braking
 
     if candidate == CAR.CHEVROLET_VOLT:
       ret.lateralTuning.pid.kpBP = [0., 40.]

selfdrive/car/honda/interface.py

@@ -74,7 +74,7 @@ def _get_params(ret, candidate, fingerprint, car_fw, experimental_long, docs):
     if candidate in HONDA_BOSCH:
       ret.longitudinalTuning.kpV = [0.25]
       ret.longitudinalTuning.kiV = [0.05]
-      ret.longitudinalActuatorDelayUpperBound = 0.5 # s
+      ret.longitudinalActuatorDelay = 0.5 # s
       if candidate in HONDA_BOSCH_RADARLESS:
         ret.stopAccel = CarControllerParams.BOSCH_ACCEL_MIN  # stock uses -4.0 m/s^2 once stopped but limited by safety model
     else:

selfdrive/car/hyundai/interface.py

@@ -94,8 +94,7 @@ def _get_params(ret, candidate, fingerprint, car_fw, experimental_long, docs):
     ret.startingState = True
     ret.vEgoStarting = 0.1
     ret.startAccel = 1.0
-    ret.longitudinalActuatorDelayLowerBound = 0.5
-    ret.longitudinalActuatorDelayUpperBound = 0.5
+    ret.longitudinalActuatorDelay = 0.5
 
     # *** feature detection ***
     if candidate in CANFD_CAR:

selfdrive/car/interfaces.py

@@ -210,8 +210,7 @@ def get_std_params(candidate):
     ret.longitudinalTuning.kiBP = [0.]
     ret.longitudinalTuning.kiV = [1.]
     # TODO estimate car specific lag, use .15s for now
-    ret.longitudinalActuatorDelayLowerBound = 0.15
-    ret.longitudinalActuatorDelayUpperBound = 0.15
+    ret.longitudinalActuatorDelay = 0.15
     ret.steerLimitTimer = 1.0
     return ret
 

selfdrive/car/tesla/interface.py

@@ -23,7 +23,7 @@ def _get_params(ret, candidate, fingerprint, car_fw, experimental_long, docs):
     ret.longitudinalTuning.kpV = [0]
     ret.longitudinalTuning.kiBP = [0]
     ret.longitudinalTuning.kiV = [0]
-    ret.longitudinalActuatorDelayUpperBound = 0.5 # s
+    ret.longitudinalActuatorDelay = 0.5 # s
     ret.radarTimeStep = (1.0 / 8) # 8Hz
 
     # Check if we have messages on an auxiliary panda, and that 0x2bf (DAS_control) is present on the AP powertrain bus

selfdrive/controls/lib/longcontrol.py

@@ -73,16 +73,10 @@ def update(self, active, CS, long_plan, accel_limits, t_since_plan):
       v_target_now = interp(t_since_plan, CONTROL_N_T_IDX, speeds)
       a_target_now = interp(t_since_plan, CONTROL_N_T_IDX, long_plan.accels)
 
-      v_target_lower = interp(self.CP.longitudinalActuatorDelayLowerBound + t_since_plan, CONTROL_N_T_IDX, speeds)
-      a_target_lower = 2 * (v_target_lower - v_target_now) / self.CP.longitudinalActuatorDelayLowerBound - a_target_now
+      v_target = interp(self.CP.longitudinalActuatorDelay + t_since_plan, CONTROL_N_T_IDX, speeds)
+      a_target = 2 * (v_target - v_target_now) / self.CP.longitudinalActuatorDelay - a_target_now
 
-      v_target_upper = interp(self.CP.longitudinalActuatorDelayUpperBound + t_since_plan, CONTROL_N_T_IDX, speeds)
-      a_target_upper = 2 * (v_target_upper - v_target_now) / self.CP.longitudinalActuatorDelayUpperBound - a_target_now
-
-      v_target = min(v_target_lower, v_target_upper)
-      a_target = min(a_target_lower, a_target_upper)
-
-      v_target_1sec = interp(self.CP.longitudinalActuatorDelayUpperBound + t_since_plan + 1.0, CONTROL_N_T_IDX, speeds)
+      v_target_1sec = interp(self.CP.longitudinalActuatorDelay + t_since_plan + 1.0, CONTROL_N_T_IDX, speeds)
     else:
       v_target = 0.0
       v_target_now = 0.0

selfdrive/test/process_replay/ref_commit

@@ -1 +1 @@
-e536df5586a71b22baa789dc584d7eab67f1fbbb
+0ba779ec9f624872b1d038acb15095b726ff8983