Automotive

ADAS

ADAS — Advanced Driver-Assistance Systems — is the umbrella term for features that use radar, LiDAR, and camera data to warn or briefly intervene, without taking over driving. Every individual feature is built on the same handful of sensors — the real engineering is in fusing their data and deciding what to do with it, not in exotic new hardware per feature.

Quick reference
What it is
Driver-assistance features (ACC, AEB, LDW/LKA, blind-spot monitoring, ...) built on fused radar/LiDAR/camera data.
Mental model
A set of co-pilot features sitting on top of the same three sensors — the sensors don't change per feature, the fusion and control logic does.
Key numbers
Most ADAS control loops run at 10–50 Hz; safety-rating programs (NCAP and similar) increasingly require specific ADAS features as standard equipment.
Reach for it when
Understanding or building a feature that watches the environment and warns or briefly intervenes — short of actually driving the car (that's levels of driving automation).

Sensor fusion: one perception, many sensors

No single sensor is good at everything (see each sensor's own strengths/weaknesses table on the Radar page), so ADAS systems combine radar, LiDAR, and camera data into one shared understanding of the surroundings before any individual feature acts on it:

Radar, camera, and LiDAR data feed into a central sensor fusion module, which produces a unified object list that multiple ADAS features draw on Radar Camera LiDAR Sensor Fusion unified object list ACC AEB LDW / LKA
Each feature below the fusion layer reads the same unified object list — a pedestrian detected by camera and confirmed by radar range is one fused object, not three disagreeing reports.

Adaptive Cruise Control

ACC is a textbook closed control loop: the driver sets a target speed and following gap, and the system continuously adjusts throttle and brake to hold them against a lead vehicle whose speed and distance keep changing.

A closed control loop: the controller commands the actuator, which changes the vehicle's speed, which the sensor measures and reports back to the controller Controller driver-set target vs. measured gap Actuator throttle / brake Vehicle actual speed changes Radar / camera measures gap & closing speed command accel / decel changes gap measured gap
This is a feedback loop, not a one-shot calculation — it runs continuously, every control cycle, for as long as ACC is engaged.

Lane Departure Warning vs. Lane Keep Assist

Both features start from the same camera measurement — the vehicle's lateral position relative to detected lane markings — and differ only in what they do with it:

A vehicle drifting left of lane center triggers either a Lane Departure Warning or, on a Lane Keep Assist system, an automatic corrective steering input lane center vehicle lateral offset (camera-measured) LDW: warn the driver LKA: apply steering torque
LDW stops at telling the driver; LKA takes the same measurement one step further and briefly steers the car back toward center itself.
LDW (Lane Departure Warning)
Alerts the driver (sound, vibration, visual) when drifting out of the lane without signaling — it never touches the steering.
LKA (Lane Keep Assist)
Goes further: applies a corrective steering torque to nudge the vehicle back toward lane center, still expecting the driver to remain in control.

A catalog of common ADAS features

FeaturePrimary sensorsWhat it does
ACCRadar, cameraHolds a set speed and following gap automatically.
AEBRadar, cameraBrakes automatically for an imminent collision the driver hasn't reacted to.
LDW / LKACameraWarns about, or corrects, unintentional lane drift.
Blind Spot MonitoringCorner radarFlags vehicles in the zones mirrors don't fully cover.
Rear Cross-Traffic AlertRear corner radarWarns of approaching traffic while reversing out of a parking spot.
Traffic Sign RecognitionCameraReads speed-limit and regulatory signs, surfaces them to the driver.
Parking assistUltrasonic sensors, cameraDetects nearby obstacles and/or steers automatically into a parking space.
Driver monitoringInward-facing cameraWatches for drowsiness or distraction.

Real-world applications

Gotchas

Field notes