.png "Reynolds & Moore Launches NVIDIA IGX Training Path")
Reynolds & Moore Launches Training Path for Physical AI Safety Based on NVIDIA IGX™
As physical AI moves into factories, warehouses, homes, and other real-world environments, engineering teams need to understand how AI, edge computing, and functional safety integrate to support safe deployment.
Reynolds & Moore has launched a new Training Path for NVIDIA IGXto help teams develop safer AI-enabled systems. The training combines functional safety principles, AI safety concepts, and practical knowledge of NVIDIA IGX architecture to equip engineering teams to incorporate safety into their AI-enabled system designs.
NVIDIA IGX is an industrial-grade edge AI computing platform designed for safety-critical and industrial applications. It combines high-performance AI compute with connectivity, security, and safety features to support rapid real-time AI processing. This makes the platform well suited for robotics safety, energy-constrained use cases, and industrial edge scenarios where real-time AI processing enables fast, reliable decisions close to the point of operation.
Supporting the Growth of Physical AI
The NVIDIA IGX Training Path is developed to guide engineering teams through the integration of NVIDIA IGX into their safety architecture design. The training focuses on the practical considerations engineers need to make when incorporating AI-enabled compute into a safety-related system. Teams gain the knowledge to make informed architecture decisions early and accelerate the path toward deployment.
Physical AI includes robots and intelligent machines that perceive, reason, and act in the physical world. The market is growing rapidly as these technologies move from research and prototypes toward real-world deployment. As physical AI becomes more capable and more integrated into operational systems, teams must address a broader range of safety considerations. These include risks associated with hardware, software, datasets, AI models, sensors, compute platforms, human interaction, and changing real-world conditions. Ensuring these elements work together safely requires engineering approaches that account for both traditional functional safety and the unique challenges introduced by AI.
Our Role in the NVIDIA Halos Ecosystem
Reynolds & Moore is part of the NVIDIA Halos ecosystem and a member of the NVIDIA Halos AI Systems Inspection Lab, the first ANSI National Accreditation Board (ANAB) accredited inspection lab for AI-driven physical systems. NVIDIA Halos is a safety platform for physical AI that helps developers build safety into each layer of their autonomous robotics solutions.
Reynolds & Moore combines functional safety expertise with hands-on experience designing safety architectures for AI-enabled systems, and the foundational accelerated computing platforms are becoming an increasingly important part of that work. Selecting and integrating the right platform early in the development process can reduce integration complexity, support a more streamlined safety architecture, and help teams prepare for certification.
NVIDIA IGX Training Path Modules Include:
Module 1: Fundamentals of Functional Safety
Duration: 4.5 hours
By the end of this module, you will be able to interpret safety requirements from the standards governing your industry and understand how they apply to your design. You will also be able to explain where safety activities fit within the development lifecycle and identify the integrity level required for a given safety function.
- What functional safety is, why it exists, and how a safety lifecycle differs from conventional product development
- The V-model development lifecycle and where safety activities fit within it
- Core safety concepts, including safety integrity levels, performance levels, safe states, redundancy, and diagnostic coverage
- Failure modes, failure rates, and the resulting design requirements
- Key topics from functional safety standards, including IEC 61508 (Functional Safety of Electrical/Electronic/Programmable Electronic Safety-Related Systems) and ISO 13849 (Safety of Machinery – Safety-Related Parts of Control Systems)
Module 2: Fundamentals of AI in Functional Safety
Duration: 4.5 hours
By the end of this module, you will be able to initiate a safety case for systems that incorporate AI. You will understand the types of faults AI components can introduce, evaluate potential mitigation strategies, and position a project within the ISO/IEC TS 22440 framework.
Certified System Integrator Program
Set Yourself at the Forefront of the Global Vision Market
.jpg)
Vision system integrators certified by A3 are acknowledged globally throughout the industry as an elite group of accomplished, highly skilled and trusted professionals. You’ll be able to leverage your certification to enhance your competitiveness and expand your opportunities.
- Why AI components are treated differently from conventional software
- What non-deterministic logic is and why it can be difficult to interpret and verify
- ISO/IEC TS 22440 fundamentals, including AI component classification, hazard and risk analysis, and the three-phase AI/ML realization process
- Why datasets and dataset coverage form an essential part of the safety argument and the role of data throughout the model development lifecycle
- Optional training add-on: The ML FMEA process and how it maps to an AI/ML development project
Module 3: NVIDIA IGX Fundamentals
Duration: 7.5 hours
By the end of this module, you will be able to identify an architecture for your use case, identify failure modes, and assign NVIDIA IGX Thor diagnostics to mitigate them. You will be able to select the appropriate NVIDIA IGX software stack for a given application, connect sensors through the correct data pathways, implement a basic safety function using the Safety Extensions Package (SEP), and explain the function of the Functional Safety Island (FSI) and safety microcontroller (sMCU) in your application.
- The purpose of the Functional Safety Island (FSI), CPU Complex (CCPLEX), Safety Microcontroller (sMCU), and Board Management Controller (BMC), and how they communicate
- An overview of the NVIDIA Holoscan Sensor Bridge for sensor fusion
- How to implement safety functions on the FSI RTOS using common architectures
- Implementing safety functions on the Functional Safety Island (FSI) using the Safety Microcontroller (sMCU), and Safety Extensions Package (SEP)
- Where to use an inside-out and outside-in approach in your safety system
Start Building Safer Physical AI Systems
The NVIDIA IGX Training Path connects AI innovation with practical functional safety engineering, from foundational safety concepts to NVIDIA IGX architecture and implementation.
You can also meet us at NVIDIA GTCBerlin, Booth #7040, to discuss the training, our role in the NVIDIA Halos AI Systems Inspection Lab ecosystem, and how we support the development of safer physical AI systems.
Reynolds & Moore
Reynolds & Moore is a functional safety engineering firm dedicated to bridging the gap between groundbreaking innovation and uncompromising safety in complex industries. Whether it is a novel <a href="https://myappsplus.com/this-is-the-way-to-deploy-the-latest-nvidia-technology-much-faster-nvidia-partner/” title=”This is the way to deploy the latest Nvidia technology 'much faster': Nvidia partner”>technology or a first-ever application, we build entirely new safety cases.
Discover how Reynolds & Moore can support your automation journey with their complete range of solutions and expertise.
Visit Company Website
« Back To News
Strengthening Software Quality and Safety: Welcoming Qt Group’s Axivion Tool to the R&M Safety Partner Ecosystem
Reynolds & Moore
06/17/2026
3 minutes
Qt Group joins the Reynolds & Moore R&M Safety Partner Ecosystem, combining Axivion code and architecture verification with R&M’s functional safety engineering for AI-driven and autonomous systems.
Numurus Solves a Critical AI Technology Gap with NEPI Software for Robotic, Smart Sensing and Edge-AI Development
Numurus LLC
02/14/2023
4 minutes
Numurus LLC, a leading provider of smart system solutions, today announced NEPI
GXO Plans 20,000 Robots in 2026. None of Them Will Be Humanoids.
Association for Advancing Automation
09/23/2026
4 minutes
After 45 humanoid pilots, GXO still sees no production deployments in 2026 as speed, dexterity and ROI remain barriers.
