Hover over the dots to read about sensor (sub)systems in self-driving cars.
Autonomous vehicles rely on several sensor (sub)systems.
Source: McKinsey analysis
Odometery sensors Use wheel speed to estimate how much vehicle travels.
Prebuilt maps High-definition maps with detailed information about roads and infrastructure (eg, shoulders, road edges, lanes) are used for precise localization and allow vehicles to better perceive their environment.
Inertial navigation systems (INS) Use accelerometers and gyroscopes to estimate vehicle position, orientation, and speed. Typically used in combination with other vehicle-related data (eg, GPS).
Dedicated short-range communication (DSRC) Used for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) systems to receive and send vehicle and infrastructure (eg, road, traffic light) information.
Ultrasonic sensors Generally have low resolution and are used for short distances (eg, park assist).
Infrared sensors Use infrared spectrum to identify and track objects that are hard to detect in low lighting conditions.
Radio detection and ranging (radar) Uses electromagnetic waves in certain bands to reflect off of an object and determine its speed and distance.
Cameras Use inexpensive hardware that requires complex software suite to interpret collected images.
Light detection and ranging (lidar) Uses light beams to estimate distance between obstacles and sensors with high resolution.
Global positioning systems (GPS) Localize vehicle using satellite triangulation. Accuracy is within several meters.
