Surrounding-object-based permission levels arbitrate competing vehicle control requests from multiple apps to reduce interference and protect safe operation.
Combining road type, nearby vehicles, and sustained low speed helps prevent false congestion detection and erroneous automated driving mode changes.
When lane markings disappear, the controller follows a lead vehicle only after checking for parallel traffic to avoid adjacent-lane collisions.
A remote management device preselects a leading vehicle by route and timing, enabling safe automatic following where LDM coverage is limited.
Roadside displays relay sensor-detected road events to approaching vehicles beyond direct sensing range, enabling earlier speed or route adjustment.
V2V and road-to-vehicle data help verify forward obstacles at intersections, suppress unsafe speed-hold prompts, and trigger deceleration.
Classifying crossing objects as asserting or yielding makes AV intent clearer, reducing unnecessary manual overrides and improving trust.
Abnormality-aware stop planning switches between emergency and controlled stops to avoid blocking intersections and other traffic participants.
Emergency detection and automatic module disconnection isolate faults in modular autonomous vehicles, aiding rescue and reducing downtime.
Accuracy-weighted external environment data helps determine lane travel direction more reliably when sensor-derived lane and object information is imprecise.