Precomputed emergency trajectories let a robotic vehicle leave hazardous stop points after a malfunction and reach a nearby safe location.
A remote monitoring center detects emergencies, sends secure driving actions, and improves autonomous vehicle response reliability.
Classifies map route sections by landmark density and sensor recognizability to support accurate automated driving under changing conditions.
A dual-path vehicle controller pairs deep learning route planning with a rule-based safe-stop backup to reach ASIL-D emergency safety.
Non-visual EM alerts share predicted crash data between nearby vehicles so adjacent cars can react before onboard sensors detect danger.
A delimited test runtime lets vehicle controllers hot-swap program elements for real-time evaluation without disrupting safety-critical control software.
HRV-based fatigue learning and monitoring lets the vehicle trigger alarms, adjust spacing, and activate safety aids when driver fatigue rises.
Pre-evaluated stop candidates and risk factors help a remotely driven vehicle avoid intersection conflicts and reduce traffic disruption.
Sensor and trajectory cues help autonomous vehicles detect likely reversing cars earlier and adjust space or timing with lower processing load.
When automated driving meets blocked or rule-breaking road conditions, network-guided route sections help vehicles pass and reuse resolved paths.
Projects construction and implement data onto the ground so operators can keep eyes on the work area and avoid underground structure contact.
Eye-tracking updates a driver response profile to predict fatigue-linked abnormal driving events and trigger earlier vehicle control actions.