Robot Safety

Industrial environments, such as warehouses and factory floors, can now take advantage of advanced robot safety that combines both “inside-out” protections on the robot itself and “outside-in” systems that monitor the broader workspace. These capabilities are powered by NVIDIA IGX™ platform and safety AI agents that can perform higher-level scene understanding, helping robots understand what’s happening around them and react safely. Together, these technologies let robots slow, stop, or reroute when people or hazards are detected, improving safety without sacrificing productivity.

Running NVIDIA IGX Thor™ at the edge lets robots execute perception, planning, and safety AI locally with industrial‑grade reliability, enabling “inside‑out” safety for real-time decision-making. IGX Thor combines powerful AI compute, high‑bandwidth sensor processing, enterprise security, and functional safety with NVIDIA AI Enterprise and Isaac™ robotics software. This lets robots fuse rich multimodal sensor data, understand human and machine behavior, and continuously adapt their motion to avoid hazards in dynamic industrial environments.

Both traditional and smart robots rely on onboard (inside-out) sensors and cameras to see, perceive, and act based on the conditions of the world around them. However, onboard sensors have a limited field of view when dealing with occlusions like static walls, payloads, and moving objects such as workers and vehicles. 

As a result, robots are forced to operate at lower speeds and reduced efficiency. In constantly changing factory and warehouse environments, this lack of flexibility leads to costly downtime for reconfiguration and reprogramming. 

NVIDIA Halos for outside-in functional safety agents extends perception beyond a robot’s sensors to include sensors installed in the surrounding environment. This provides functional safety and operational awareness for autonomous robots and shifts reactive safety to proactive situational awareness. By using infrastructure cameras to their full potential, these AI functional safety agents add always-on continuous monitoring and adaptive reasoning that extends to robot fleets, logistics, and workers who move dynamically through spaces.

The NVIDIA Halos AI Systems Inspection Lab helps industrial robot deployments meet rigorous safety and AI integrity requirements, providing hardware and platform companies with compliance recognition for global robot and machinery safety standards. The lab fully aligns with expert credentialing organizations such as ANAB and works alongside well-known certification bodies like TÜV and other third‑party agencies. This gives operators confidence that their robots, sensors, and platforms are evaluated against industry‑recognized best practices for functional safety and secure integration.

• 360° Situational Awareness: Augment onboard sensors and reduce blind spots with infrastructure-based cameras.
• Proactive Safety: Predict and prevent incidents in real time. Reduce safety incidents and near misses through proactive detection of hazardous interactions and line-of-sight occlusions.
• Enhanced Productivity: Automatically modulate robot efficiency across dynamic environments to optimize actions.
• Human-Robot Collaboration: Operate at higher robot speeds when zones are clear of people, with automatic slow-down or stopping when workers enter.
• Functional Safety Standards Compliance: Implement solutions developed, tested, and inspected by TÜV Rheinland to comply with current and emerging functional safety standards.

Running on the NVIDIA IGX platform, Halos outside-in functional safety agents fuse low-latency detections with safety monitoring and decision logic to supervise multiple robots simultaneously. It slows or stops them the moment a person enters a protected area, while allowing higher speeds and closer collaboration when zones are clear. This approach helps manufacturers reduce safety incidents and false stops, increase robot throughput, and more easily navigate occlusions. 

NVIDIA Metropolis Blueprint for video search and summarization (VSS) and the NVIDIA Cosmos Reason reasoning vision language model delivers tremendous benefits for operations teams. They can now simplify compliance with functional safety and emerging AI safety standards through predesigned architectures, inspection reports, and detailed safety event logging.

Outside-in cameras establish a virtual fence and monitor the trailer interior and docking zones for occlusion alerts. Robots operate at full speed or high-efficiency mode when no workers are in the region of interest, and engage safety functions the moment a person enters to prevent incidents from occurring.

Virtual tripwires and dynamic zones manage multiple AMR robots and people in shared aisles and intersections, especially around blind corners or high-rack areas. Especially in areas with material handling, alerts and occlusions detection are imperative when a person or robot enters a hidden zone or is obscured by an object, maintaining safety even in blind spots.