FORT Robotics, the “Trust Layer” for physical AI, has announced joining the Nvidia Halos for Robotics ecosystem that’s bringing safety to autonomous robots.
FORT will demonstrate an agentic safety application built with the open source Nvidia Halos Outside-In Safety Blueprint this week at the Automate conference in Chicago and presenting in the Humanoid Robotics Pavilion with Nvidia on June 23rd.
The Nvidia Outside-In Safety Blueprint combined with the FORT Trust Layer extends robot perception beyond onboard sensors by using external infrastructure sensors and visual AI Agents to deliver real-time, safety-certifiable functional safety to maximize operational throughput.
Leveraging Nvidia IGX Thor and Nvidia Holoscan Sensor Bridge for AI compute and sensor connectivity, the solution enables robots to safely operate alongside workers at high efficiency modes while dynamically adapting to complex environments.
This offering provides significant value beyond traditional inside-out functional safety systems that are limited to onboard sensors and conservative operating constraints:
Enhanced productivity
Traditional safety systems were built for predictable machines in bounded settings, and lack the flexibility needed for mobile robot systems in constantly changing warehouses and factories.
Outside-In Safety automatically modulates robot efficiency across dynamic environments, reducing costly robot slowdowns and optimizing both safety and productivity.
Worker safety and accident prevention
As more worksites adopt autonomous systems and physical AI, safety frameworks must adapt to protect workers in mixed human-robot environments.
By providing proactive situational awareness, Outside-In Safety can prevent safety incidents and protect workers in real time. This can help address safety challenges across multiple industries.
Maximize return on investment
Nvidia Halos Outside-In Safety Blueprint will help FORT’s new and existing customers across warehousing, manufacturing, and other automated industries to leverage both robots and existing infrastructure for more value.
For example, building-mounted cameras can be leveraged to unlock new cost savings by maximizing throughput and optimizing processes such as trailer truck unloading, inventory replenishment, product assembly, and more.
A robotics safety ecosystem built for scale
FORT is a member of the Nvidia Halos AI Systems Inspection Lab, the world’s first ANSI National Accreditation Board (ANAB)-accredited inspection lab designed specifically for physical AI and autonomous systems.
It provides a unified framework to verify functional safety, cybersecurity, and AI compliance for autonomous vehicles, robotics, and sensor technologies. This collaboration reflects FORT’s ongoing work with Nvidia to making physical AI trustworthy at industrial scale.
“Safety has always been the precondition for scale – you can’t deploy robots broadly if you can’t guarantee they’ll operate safely around people and valuable infrastructure,” said Samuel Reeves, CEO of FORT.
“What collaborating with Nvidia gives us is enhanced perception that makes safety genuinely intelligent. Agentic robots that understand their environment and respond in real time aren’t just safer, they’re more productive. That’s the combination the industry has been waiting for.”
FORT has long been the industry standard for safety-certified control, providing autonomous systems with the essential hardware and software backbone required to mitigate real-world operational risk. Outside-In Safety extends FORT’s Trust Layer for Physical AI to be even broader.
- Outside-In Safety (new): Reduces costly robot slowdowns and improves safety without sacrificing productivity, by automatically modulating robot efficiency across dynamic environments.
- Onboard Active Safety: Onboard perception technology, either embedded or bolted-on, enables machines to actively detect, anticipate, and respond to their environments in real time. This predictive approach allows autonomous vehicles to execute smart, real-time planning and contingency maneuvers, a meaningful leap beyond traditional reactive safety architectures.
- Human-in-the-Loop control: Safe and reliable remote operation including both line-of-sight control and remote operation and intervention.
Facts Only
* FORT Robotics announced joining the Nvidia Halos for Robotics ecosystem.
* FORT will demonstrate an agentic safety application built with the open source Nvidia Halos Outside-In Safety Blueprint at the Automate conference in Chicago.
* FORT will present in the Humanoid Robotics Pavilion with Nvidia on June 23rd.
* The Nvidia Outside-In Safety Blueprint extends robot perception beyond onboard sensors using external infrastructure sensors and visual AI Agents.
* This blueprint delivers real-time, safety-certifiable functional safety by maximizing operational throughput.
* The solution leverages Nvidia IGX Thor and Nvidia Holoscan Sensor Bridge for AI compute and sensor connectivity.
* The offering enhances productivity by automatically modulating robot efficiency across dynamic environments.
* The solution aims to prevent safety incidents by providing proactive situational awareness in mixed human-robot environments.
* FORT is a member of the Nvidia Halos AI Systems Inspection Lab, an ANAB-accredited inspection lab for physical AI and autonomous systems.
* Outside-In Safety reduces robot slowdowns and improves safety without sacrificing productivity by modulating efficiency.
* The system includes Onboard Active Safety, Human-in-the-Loop control, and remote operation/intervention capabilities.
Executive Summary
FORT Robotics is joining the Nvidia Halos for Robotics ecosystem, focusing on bringing safety to autonomous robots through the FORT Trust Layer. This integration involves demonstrating an agentic safety application using the open source Nvidia Halos Outside-In Safety Blueprint at the Automate conference and presenting with Nvidia in the Humanoid Robotics Pavilion on June 23rd. The Outside-In Safety Blueprint extends robot perception beyond onboard sensors by utilizing external infrastructure sensors and visual AI Agents to provide real-time, safety-certifiable functional safety. This solution leverages Nvidia IGX Thor and Holoscan Sensor Bridge for compute and sensor connectivity, enabling robots to operate efficiently alongside workers in dynamic environments.
The offering provides benefits over traditional inside-out safety systems by enhancing productivity through automated efficiency modulation, preventing accidents via proactive situational awareness, and maximizing return on investment by leveraging existing infrastructure for process optimization. Furthermore, the collaboration establishes a robotics safety ecosystem built for scale through FORT’s membership in the Nvidia Halos AI Systems Inspection Lab, an ANSI National Accreditation Board (ANAB)-accredited inspection lab for physical AI systems.
Full Take
The narrative centers on shifting the paradigm of robotics safety from reactive, onboard constraint management to proactive, distributed environmental awareness—an "Outside-In" approach. This move fundamentally challenges the assumption that operational scale requires limiting safety to the machine itself; instead, it posits that scale is contingent upon understanding the system's context relative to the physical world and human workers. The pattern observed is the integration of external, verifiable infrastructure (sensors) with AI agents to generate functional safety guarantees, positioning trust as an emergent property derived from environmental perception rather than solely from internal state monitoring.
The implication for industrial deployment is that current safety systems are inherently limited by their bounded settings, creating a bottleneck where productivity gains cannot be realized without sacrificing cautious operation. The push toward Agentic safety suggests that true operational efficiency in dynamic environments is not achieved by slowing down robots or imposing rigid rules, but by allowing the system to intelligently modulate performance based on real-time environmental risk assessments. This mirrors broader trends in complex systems engineering where managing interactions between distributed agents (robots and humans) requires a shared perception framework, moving beyond siloed control architectures toward holistic accountability across the physical domain.
Bridge Questions: If safety is maximized through external perception, what are the new verifiable standards required for certifying the integrity of those external infrastructure sensors? How does this "outside-in" dynamic change liability frameworks when autonomous decisions are mediated by environmental data rather than solely onboard computation? What mechanisms must be in place to ensure that modulating robot efficiency remains aligned with established legal and ethical safety protocols across diverse industrial contexts?
Sentinel — Human
The text appears to be a well-structured piece of journalistic reporting or a company announcement, focusing on synthesizing complex technical integration for business benefit.