DataTerminology.comNEW
Also known as: Autonomous vehicle, Self-driving
A vehicle capable of perceiving its environment and navigating without human input using AI-driven sensing, prediction, and control.
DESCRIPTION
A self-driving car integrates perception (cameras, LiDAR, radar), localization and mapping, prediction of agents, planning, and control to operate autonomously. It relies on real-time computing, robust sensor fusion, and fail-safe architectures. Operational design domains constrain where autonomy is guaranteed. Safety engineering, redundancy, and extensive testing underpin deployment. Regulatory frameworks and standards define automation levels and safety requirements.
ALTERNATIVE DEFINITIONS
NHTSA
Automated vehicles use technology to assist with driving tasks or to drive entirely without human input, depending on the level of automation.
Frames autonomy within levels and safety context from a regulator.
SAE International (J3016)
J3016 provides a taxonomy and definitions for terms related to driving automation systems for on-road motor vehicles.
Authoritative standard defining levels of driving automation.
Encyclopaedia Britannica
An autonomous vehicle is a motor vehicle capable of navigating without human input by sensing its environment and operating driving functions.
General-purpose definition for public understanding.
RELATED TERMS
Sensor Fusion Computer Vision LiDAR SLAM Path Planning Control System V2X Communication Safety Case AI Governance
COMPETING TERMS
Advanced Driver-Assistance Systems Human-Driven Vehicle
KNOWLEDGE AREAS
AI IoT & Edge Data Real-time & Streaming Data AI Governance
TARGET AUDIENCE
AI Engineer MLOps Engineer Software Developer Data Scientist Chief Technology Officer
EXTERNAL SOURCES
- Wikipedia - Self-driving car
- NHTSA - Regulatory safety overview and definitions
- SAE International - Standard for automation taxonomy and levels
- Waymo - Vendor safety approach and methodology
TOP QUESTIONS
What are the SAE levels relevant to a Self-Driving Car?
SAE J3016 defines Levels 0–5, where Levels 4–5 denote high/full automation without human fallback within an operational design domain.
Category: Overview
What subsystems make a Self-Driving Car possible?
Perception, localization, prediction, planning, and control, supported by sensor fusion and robust compute, enable autonomous operation.
Category: Requirements
How is safety validated for a Self-Driving Car?
Through simulation, on-road testing, scenario coverage, redundancy, and compliance with safety standards and regulatory guidance.
Category: Testing
Why is mapping important for a Self-Driving Car?
HD maps provide prior context to complement sensor data, improving localization and planning reliability in complex environments.
Category: Application
What is a common misconception about Self-Driving Cars?
That Level 2 driver assistance equals autonomy; in fact, the human remains responsible at lower levels.
Category: Misunderstanding
What are major risks in deploying Self-Driving Cars?
Edge cases, sensor failures, adversarial conditions, and incomplete regulatory readiness pose significant risks.
Category: Risks
How does V2X enhance a Self-Driving Car?
Vehicle-to-everything communication augments perception and coordination, improving safety and efficiency.
Category: Example
What data is essential for training Self-Driving Cars?
Diverse sensor data across weather, geography, and traffic scenarios; well-labeled edge cases are crucial.
Category: Requirements
How do Self-Driving Cars handle uncertainty?
They use probabilistic modeling for perception and prediction, with conservative planning and fallback strategies.
Category: Best practices