Robotaxi and Autonomous Vehicle Safety

NVIDIA Halos connects full-stack safety with evaluation and inspection to support autonomous vehicle development and deployment.

Read the Safety Report

Overview
Highlights
Technology
Safety by Layer
Certification
Research
Partners
Resources
Next Steps

Overview
Highlights
Technology
Safety by Layer
Certification
Research
Partners
Resources
Next Steps

Overview

Full-Stack Safety, From Cloud to Car

NVIDIA Halos for autonomous vehicles is a full-stack safety system that accelerates safe robotaxi development and deployment through various safety layers. Built on TÜV SÜD–certified development processes, it supports validation throughout the vehicle lifecycle.

Platform safety combines NVIDIA Hyperion’s redundant architecture with NVIDIA DRIVE AGX™ hardware protections. NVIDIA Halos OS provides system software safeguards, while algorithmic safety pairs NVIDIA Alpamayo™ AI driving model with independent safety checks. Ecosystem safety extends validation through NVIDIA Halos SAFE’s simulation-based evaluation and protocol-aligned testing. The NVIDIA Halos AI Systems Inspection Lab verifies partner integration with Halos safety elements and provides inspection reports and certificates to support independent system certification.

Safety for Robotaxis Must Be Engineered End to End

Halos spans safety-engineered compute, trusted system software, domain-specific applications, E2E models, evaluation and independent inspection—giving robotaxi developers one safety foundation from development through fleet deployment.

Read the Blog

NVIDIA Halos SAFE: Evaluation and Validation for AVs

Halos SAFE turns safety requirements into repeatable scenarios, tests, KPIs, simulation, and credible evidence, helping AV teams understand whether autonomous systems behave as intended.

Watch the demo

Highlights

Autonomous Vehicle Safety Leadership

NVIDIA Halos is the result of continuous investment in autonomous vehicle safety—from research to engineering to active engagement with international safety standards—validated by independent third-party assessments.

18,600+

Engineering years invested in vehicle safety to date

21 Billion+

Safety transistors safety assessed

7,000,000

Lines of safety-assessed code

2,000,000

Daily end-to-end integration tests for validation

22,000+

Platform safety monitors

25,000+

Hours of safety test data

1,000+

Patents filed

330+

Research papers published on AV safety

30+

Certificates and assessment reports issued

Technology

Safety Across the Stack

NVIDIA Halos delivers full-stack AV safety through safety-engineered hardware and software, AI safeguards, and independent inspection.

Safety By Layer

NVIDIA Halos builds safety into every layer of autonomous driving, from the platform and software to evaluation and certification.

Safety-Engineered Platform

NVIDIA Hyperion integrates NVIDIA DRIVE AGX compute and diverse sensors in a redundant safety architecture. NVIDIA Alpamayo adds reasoning-based driving, with policy evaluation and safety monitors assessing proposed actions.

NVIDIA Halos OS

Safety software foundation for autonomous driving stacks, includes Halos Core system services and middleware, and Halos Classical Stack active safety, parking, and Level 2+ capabilities that complement AI-based driving.

NVIDIA Halos SAFE

AV safety evaluation framework spanning regional NCAP protocols and L2++–L4 systems, combining on-road testing, simulation, and fleet monitoring to measure driving performance and produce traceable evidence.

NVIDIA Halos AI Systems Inspection Lab

The first ANAB-accredited inspection program for AI functional safety in physical AI to inspect integration of Halos elements before third-party certification.

Learn More

Video

Halos SAFE: A Safety Assurance Framework for Evaluation

NVIDIA Halos SAFE is a safety evaluation framework; the demo shows customers how driving scenarios are tested in simulation and assessed against safety requirements to generate evidence for AV validation.

Watch the Video

Certification

Assessed by Experts

Independent third-party safety and cybersecurity assessments of NVIDIA Halos elements demonstrate NVIDIA’s ongoing commitment to AV safety.

ANSI National Accreditation Board

ANAB accredited the NVIDIA Halos AI Systems Inspection Lab as an ISO/IEC 17020 Inspection Body. NVIDIA is the first company accredited by ANAB for an inspection plan that combines cybersecurity, AI, and functional safety.

TÜV SÜD

TÜV SÜD certified the core NVIDIA hardware and software process to Automotive Safety Integrity Level (ASIL) D. Under the ISO 26262 standard, NVIDIA DriveOS 6.0 is certified ASIL D conformant and Thor-X SoC is assessed as ASIL D conformant. NVIDIA also received ISO/SAE 21434 Cybersecurity Process certification for its automotive system-on-a-chip, platform, and software engineering processes.

TÜV Rheinland

TÜV Rheinland performed an independent United Nations Economic Commission for Europe safety assessment of NVIDIA DRIVE AV related to safety requirements for complex electronic systems.

Partners

Partners Using NVIDIA Halos System

Leading robotaxi companies, original equipment manufacturers (OEMs), safety pioneers, simulation and mapping companies, and software and sensor providers use NVIDIA Halos elements across the AV stack.

Automotive OEMs
Tier 1 Suppliers
Sensors Providers
Certification and Compliance
Simulation and Validation

Research

NVIDIA Research for Autonomous Vehicles

Our research and development have published 330+ research papers on autonomous vehicle safety.

NVIDIA Halos SAFE Framework

Provides tools and guidelines to build a credible safety case from L2 ADAS to L4 robotaxis, drawing on 330+ research papers and 1,000+ patents within NVIDIA Halos.

Read the Tech Brief

Alpamayo 1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail

Comprehensive evaluations with open-loop metrics, closed-loop simulation, and real-world vehicle tests demonstrate that Alpamayo 1 is state-of-the-art in multiple aspects (including reasoning, trajectory generation, alignment, safety, latency, and more).

Learn More About Alpamayo

Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation With World Foundation Models

Collecting and annotating real-world data for safety-critical physical AI systems, such as Autonomous Vehicles (AVs), is time-consuming and costly. To address this challenge, we introduce the Cosmos-Drive-Dreams, a synthetic data generation (SDG) pipeline for generating challenging scenarios to facilitate downstream tasks such as perception and driving policy training.

Learn More About Cosmos

SafeVL: Driving Safety Evaluation Using Meticulous Reasoning in Vision Language Models

Safety remains a fundamental challenge in autonomous driving, with a key step being the development of a safety evaluator that can reliably identify unsafe (i.e., collision-prone) scenarios.

Visit GitHub

Enhancing Autonomous Driving Safety With Collision Scenario Integration

HydraSafe is a framework that addresses the challenge of ensuring autonomous vehicle safety in hazardous scenarios by improving data availability and planner robustness.

Visit the Project Page

Safety Evaluation of Motion Plans Using Trajectory Predictors as Forward Reachable Set Estimators

The advent of end-to-end autonomy stacks—often lacking interpretable intermediate modules—has placed an increased burden on ensuring that the final output, i.e., the motion plan, is safe in order to validate the safety of the entire stack.

Read the Paper

For a list of additional AV Research papers, click here.

Resources

The Latest in NVIDIA Halos Resources

Blogs
Press Releases
Sessions
Live Streams

The World Is Building Robotaxis on NVIDIA DRIVE Hyperion

Global automakers, software partners, and mobility leaders are bringing Level 4-ready fleets to market on DRIVE Hyperion—NVIDIA's robotaxi-ready platform.

Read the Press Release

NVIDIA Unveils Alpamayo Super and Expands the Open AV Ecosystem

Alpamayo 2 Super is a 34 billion-parameter reasoning VLA model, built for Level 4 robotaxi-ready autonomous vehicles.

Read the Press Release

BYD, Geely, Isuzu, and Nissan Adopt NVIDIA DRIVE Hyperion for Level 4 Vehicles

Leading automakers and mobility providers are building on NVIDIA’s end-to-end platform as it advances Level 4 hardware.

Read the Press Release

View All AV Safety Sessions

Scaling Functionally Safe AVs With NVIDIA DriveOS and Hyperion

How do you scale autonomous vehicles to large fleets while meeting the world’s strictest safety standards? This livestream presented by NVIDIA safety experts shows how the DriveOS stack and Hyperion reference architecture solve the safety-scalability problem.

Watch the Replay

Open Dataset for Autonomous Vehicle Safety Research

In this livestream, we showcase how the newly launched NVIDIA Alpamayo autonomous vehicles open dataset creates new opportunities to advance AV safety.

Watch the Replay

Accelerating Physical AI Certification With NVIDIA Halos

Learn about NVIDIA Halos AI Systems Inspection Lab—the first ANSI National Accreditation Board (ANAB)-accredited lab dedicated to physical AI systems.

Watch the Replay

Autonomous Vehicle Safety Validation Through Simulation

In this livestream, we present a general, theoretically grounded framework for AV safety validation whereby real-world tests are paired with simulated tests on corresponding reconstructed scenarios.

Watch the Replay

Next Steps

Redefining Safe Autonomy

Learn how cutting-edge AI, rigorous validation frameworks, and global standards are shaping autonomous vehicle safety.

Read the Product Brief

Automotive News

Stay Informed

NVIDIA Automotive

NVIDIA solutions give you the performance and scalability to design, visualize, develop, and simulate the future of driving.

Explore NVIDIA Automotive

Read the NVIDIA Autonomous Vehicles Safety Report

Section

Section

First Name

Last Name

Business Email Address

Organization / University Name

Industry

Job Title

Location

Preferred Language

State/Province

enterpriseOptIns hidden field

Send me the latest enterprise news, announcements, and more from NVIDIA. I can unsubscribe at any time.

NVIDIA Privacy Policy

I agree to the collection and processing of the above information by NVIDIA <span class="corporation-txt hidden">Corporation </span>for the purposes of research and event organization, and I have read and agree to <a href="https://www.nvidia.com/en-us/about-nvidia/privacy-policy/?deeplink=visiting-our-website"; target="_blank">NVIDIA Privacy Policy</a>.

I agree that the above information will be transferred to NVIDIA Corporation in the United States and stored in a manner consistent with <a href="https://www.nvidia.com/en-gb/about-nvidia/privacy-policy/?deeplink=visiting-our-website" target="_blank">NVIDIA Privacy Policy</a> due to necessities for research, event organization and corresponding NVIDIA internal management and system operation need. You may contact us by sending an email to <a href="mailto:privacy@nvidia.com">privacy@nvidia.com</a> to resolve related problems.

Robotaxi and Autonomous Vehicle Safety

Full-Stack Safety, From Cloud to Car

Safety for Robotaxis Must Be Engineered End to End

NVIDIA Halos SAFE: Evaluation and Validation for AVs

Autonomous Vehicle Safety Leadership

18,600+

21 Billion+

7,000,000

2,000,000

22,000+

25,000+

1,000+

330+

30+

Safety Across the Stack

Safety By Layer

Safety-Engineered Platform

NVIDIA Halos OS

NVIDIA Halos SAFE

NVIDIA Halos AI Systems Inspection Lab

Halos SAFE: A Safety Assurance Framework for Evaluation

Assessed by Experts

ANSI National Accreditation Board

TÜV SÜD

TÜV Rheinland

Partners Using NVIDIA Halos System

NVIDIA Research for Autonomous Vehicles

NVIDIA Halos SAFE Framework

Alpamayo 1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail

Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation With World Foundation Models

SafeVL: Driving Safety Evaluation Using Meticulous Reasoning in Vision Language Models

Enhancing Autonomous Driving Safety With Collision Scenario Integration

Safety Evaluation of Motion Plans Using Trajectory Predictors as Forward Reachable Set Estimators

The Latest in NVIDIA Halos Resources

Blogs

Press Releases

Sessions

Live Streams

The World Is Building Robotaxis on NVIDIA DRIVE Hyperion

NVIDIA Unveils Alpamayo Super and Expands the Open AV Ecosystem

BYD, Geely, Isuzu, and Nissan Adopt NVIDIA DRIVE Hyperion for Level 4 Vehicles

Scaling Functionally Safe AVs With NVIDIA DriveOS and Hyperion

Open Dataset for Autonomous Vehicle Safety Research

Accelerating Physical AI Certification With NVIDIA Halos

Autonomous Vehicle Safety Validation Through Simulation

Next Steps

Redefining Safe Autonomy

Automotive News

NVIDIA Automotive

Read the NVIDIA Autonomous Vehicles Safety Report