docs: expose MovementTest time horizon

This commit is contained in:
梁薄云
2026-08-05 13:31:36 +08:00
parent 3ba9304338
commit 09753143f7
@@ -6,21 +6,19 @@ Reduce the observation-test longitudinal QP size without relaxing the solver acc
## Decision ## Decision
Add a configurable `OutputTimeStepSeconds` to the trajectory-observation MovementTest with a default of `0.10d` seconds. Add configurable `TimeHorizonSeconds` and `OutputTimeStepSeconds` values to the trajectory-observation MovementTest, with defaults of `6d` and `0.10d` seconds.
The EM time horizon remains `6d` seconds. At the parking-test speed limit of `0.20m/s`, a 0.10-second knot interval corresponds to at most 0.02m of travel between knots. `TimeHorizonSeconds` controls how far into the future the ST plan covers. `OutputTimeStepSeconds` controls how many knots divide that horizon and must not exceed it. At the parking-test speed limit of `0.20m/s`, a 0.10-second knot interval corresponds to at most 0.02m of travel between knots.
## Scope ## Scope
- Add, snapshot, and validate `OutputTimeStepSeconds` in `TrajectoryObservationSettings`. - Add, snapshot, and validate `TimeHorizonSeconds` and `OutputTimeStepSeconds` in `TrajectoryObservationSettings`.
- Expose the setting in the MovementTest as a public field with a default of `0.10d`. - Expose both settings in the MovementTest as public fields with defaults of `6d` and `0.10d`.
- Copy it to the session-local `EmPlannerConfiguration.Scheduling.OutputTimeStepSeconds`. - Copy both values to the session-local `EmPlannerConfiguration.Scheduling` configuration.
- Verify custom settings reach the planning request and non-positive values are rejected. - Verify custom settings reach the planning request and non-positive values are rejected.
## Non-goals ## Non-goals
- Keep `TimeHorizonSeconds = 6d`.
- Keep `ObserverPeriodSeconds = 0.05d`. - Keep `ObserverPeriodSeconds = 0.05d`.
- Keep `SolverTimeoutSeconds = 0.50d`, `MaximumOsqpIterations = 12000`, and all `1e-5d` solver tolerances. - Keep `SolverTimeoutSeconds = 0.50d`, `MaximumOsqpIterations = 12000`, and all `1e-5d` solver tolerances.
- Do not change production EM defaults, maps, vehicle constraints, or command output behavior. - Do not change production EM defaults, maps, vehicle constraints, or command output behavior.