ClarusC64/

autonomous-driving-graceful-degradation-policy-triggering-v0.1

What this dataset tests Whether a system can respond to system-wide decoherence with a controlled policy degradation. This is not "stop the car." It is safe control under uncertainty. Required outputs

degradation_policy_id
response_mode
speed_adjustment
sensor_reweighting_plan
driver_alert_strategy
recovery_exit_condition

Scoring conventions

speed adjustment is a signed delta (mph)
reweight plan must name which modalities to downweight or upweight
exit condition must be… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/autonomous-driving-graceful-degradation-policy-triggering-v0.1.