Vehicle gap metric (#1971)

CHANGELOG.md

@@ -26,7 +26,8 @@ Copy and pasting the git commit messages is __NOT__ enough.
 - Added engine `observation_workers` configuration which can be used to configure the number of parallel sensor workers: 0 runs the sensors on the local thread, >=1 runs using the multiprocessing backing.
 - Added engine `sensor_parallelization` configuration of sensor parallelization backing, options ("mp"|"ray"): "mp" python multiprocessing, "ray" ray worker backing.
 - Added engine `reset_retries` configuration engine retries before the simulator will raise an error on reset.
-- Introduced new comfort metric in metric cost functions.
+- Introduced new comfort cost function in metric module.
+- Introduced new gap-between-vehicles cost function in metric module.
 ### Changed
 - The trap manager, `TrapManager`, is now a subclass of `ActorCaptureManager`.
 - Considering lane-change time ranges between 3s and 6s, assuming a speed of 13.89m/s, the via sensor lane acquisition range was increased from 40m to 80m, for better driving ability.
@@ -48,6 +49,7 @@ Copy and pasting the git commit messages is __NOT__ enough.
 - In the metrics module, the records which is the raw metrics data and the scoring which is the formula to compute the final results are now separated to provided greater flexibility for applying metrics to different environments.
 - Benchmark listing may specify specialised metric formula for each benchmark.
 - Changed `benchmark_runner_v0.py` to only average records across scenarios that share the same environment. Records are not averaged across different environments, because the scoring formula may differ in different environments.
+- Renamed GapBetweenVehicles cost to VehicleGap cost in metric module.
 ### Deprecated
 ### Fixed
 - Fixed issues related to waypoints in junctions on Argoverse maps. Waypoints will now be generated for all paths leading through the lane(s) the vehicle is on.

docs/benchmarks/driving_smarts_2023_3.rst

@@ -196,7 +196,7 @@ Train
     $ cd <path>/SMARTS/examples/rl/platoon
     $ python3.8 -m venv ./.venv
     $ source ./.venv/bin/activate
-    $ pip install --upgrade pip wheel
+    $ pip install --upgrade pip
     $ pip install -e ./../../../.[camera_obs,argoverse]
     $ pip install -e ./inference/
 
@@ -246,7 +246,7 @@ Evaluate
     $ cd <path>/SMARTS
     $ python3.8 -m venv ./.venv
     $ source ./.venv/bin/activate
-    $ pip install --upgrade pip wheel
+    $ pip install --upgrade pip
     $ pip install -e .[camera_obs,argoverse]
     $ scl zoo install examples/rl/platoon/inference
     $ scl benchmark run driving_smarts_2023_3 examples.rl.platoon.inference:contrib-agent-v0 --auto-install

examples/rl/platoon/train/reward.py

@@ -45,8 +45,6 @@ def step(self, action):
                     agent_obs["events"]["collisions"] | agent_obs["events"]["off_road"]
                 ):
                     pass
-                elif agent_obs["events"]["agents_alive_done"]:
-                    print(f"{agent_id}: Agents alive done triggered.")
                 elif agent_obs["events"]["actors_alive_done"]:
                     print(f"{agent_id}: Actors alive done triggered.")
                 else:

smarts/benchmark/driving_smarts/v2023/metric_formula_platoon.py

@@ -30,9 +30,9 @@
 from smarts.env.gymnasium.wrappers.metric.params import (
     Comfort,
     DistToObstacles,
-    GapBetweenVehicles,
     Params,
     Steps,
+    VehicleGap,
 )
 from smarts.env.gymnasium.wrappers.metric.types import Record
 from smarts.env.gymnasium.wrappers.metric.utils import (
@@ -63,10 +63,10 @@ def params(self) -> Params:
             dist_to_obstacles=DistToObstacles(
                 active=False,
             ),
-            gap_between_vehicles=GapBetweenVehicles(
-                active=False,
-                interest="Leader-007",
-            ),  # TODO: Activate after implementing gap_between_vehicles cost function.
+            vehicle_gap=VehicleGap(
+                active=True,
+                actor="Leader-007",
+            ),
             steps=Steps(
                 active=False,
             ),
@@ -85,15 +85,15 @@ def score(self, records_sum: Dict[str, Dict[str, Record]]) -> Score:
         +-------------------+--------+-----------------------------------------------------------+
         | DistToDestination | [0, 1] | Remaining distance to destination. The lower, the better. |
         +-------------------+--------+-----------------------------------------------------------+
-        | GapBetweenVehicles| [0, 1] | Gap between vehicles in a platoon. The higher, the better.|
+        | VehicleGap        | [0, 1] | Gap between vehicles in a convoy. The lower, the better.  |
         +-------------------+--------+-----------------------------------------------------------+
         | Humanness         | [0, 1] | Humanness indicator. The higher, the better.              |
         +-------------------+--------+-----------------------------------------------------------+
         | Rules             | [0, 1] | Traffic rules compliance. The higher, the better.         |
         +-------------------+--------+-----------------------------------------------------------+
 
         Returns:
-            Score: Contains "Overall", "DistToDestination", "GapBetweenVehicles",
+            Score: Contains "Overall", "DistToDestination", "VehicleGap",
             "Humanness", and "Rules" scores.
         """
 
@@ -123,10 +123,10 @@ def score(self, records_sum: Dict[str, Dict[str, Record]]) -> Score:
         dist_to_destination = costs_final.dist_to_destination
         humanness = _humanness(costs=costs_final)
         rules = _rules(costs=costs_final)
-        gap_between_vehicles = costs_final.gap_between_vehicles
+        vehicle_gap = costs_final.vehicle_gap
         overall = (
             0.50 * (1 - dist_to_destination)
-            + 0.25 * gap_between_vehicles
+            + 0.25 * (1 - vehicle_gap)
             + 0.20 * humanness
             + 0.05 * rules
         )
@@ -135,7 +135,7 @@ def score(self, records_sum: Dict[str, Dict[str, Record]]) -> Score:
             {
                 "overall": overall,
                 "dist_to_destination": dist_to_destination,
-                "gap_between_vehicles": gap_between_vehicles,
+                "vehicle_gap": vehicle_gap,
                 "humanness": humanness,
                 "rules": rules,
             }

smarts/env/gymnasium/driving_smarts_2023_env.py

@@ -162,7 +162,7 @@ def resolve_agent_interface(agent_interface: AgentInterface):
     )
     max_episode_steps = 1000
     waypoints_lookahead = 80
-    neighborhood_radius = 50
+    neighborhood_radius = 80
     return AgentInterface(
         accelerometer=True,
         action=agent_interface.action,

smarts/env/gymnasium/platoon_env.py

@@ -169,7 +169,7 @@ def resolve_agent_interface(agent_interface: AgentInterface):
     )
     max_episode_steps = 1000
     waypoints_lookahead = 80
-    neighborhood_radius = 50
+    neighborhood_radius = 80
     return AgentInterface(
         accelerometer=True,
         action=agent_interface.action,