You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit eddcf0f
Browse filesBrowse the repository at this point in the historyBrowse files
feat(telemetry): version the distance conversion, restore the odometer
Two consequences of the 3.6x speed error, both of which outlive the fix.
A trip distance is the integral of speed, so it inherits whatever conversion
was in force when it was recorded. Every trip already stored carries a
distance 3.6 times too long and nothing in the number says so: 8.46 km for a
2.4 km route reads exactly like 8.46 km for an 8.46 km route. A model trained
on that history would learn the error.
VehicleSpeedEvidence is BatteryPowerEvidence's counterpart on the distance
side, and EnergyTripSummary.modellableDistanceKm answers only for a distance
the current conversion produced. recordedDistanceKm is untouched — a driver's
own record does not silently rewrite itself, it is simply not something a
model may train on.
And the odometer reads again. PERF_ODOMETER is declared and never published
on this car, so total mileage showed unavailable while the head unit knew it;
the navigation adapter answers, and that reading is what gave the speed error
its independent reference in the first place.
Refs: CP-003, CP-012
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
0 commit comments