Site icon GIXtools

Using the Power of AI to Make Factories Safer

Decorative image of a robotic assembly line with bounding boxes.

As industrial automation increases, safety becomes a greater challenge and top priority for enterprises.  Safety encompasses multiple aspects:  System…Decorative image of a robotic assembly line with bounding boxes.

As industrial automation increases, safety becomes a greater challenge and top priority for enterprises. 

Safety encompasses multiple aspects: 

The same technological solution that’s driving automation can be used to also address safety: artificial intelligence. 

AI-powered stationary outside-in safety platforms, which monitor activity across many distributed machines or robots, can predictively and proactively orchestrate consistent safety policies. Machines and robots equipped with inside-out reactive safety can detect any specific interaction within their workspace and take appropriate measures.  

Diagram shows orchestrating distributed mobile machines with stationary safety platforms.Figure 1. Inside-out and outside-in safety

IEC 61508 is an international standard published by the International Electrotechnical Commission (IEC). It consists of methods to apply, design, deploy, and maintain safety-related systems.

Using IEC 61508 general architecture, AI can either be used to implement the safety function or to assist functional safety to provide additional risk reduction measures.

Workflow diagram shows an EUC control system assisting reactive functional safety with proactive safety functions.  Figure 2. Reactive and proactive safety

The use of AI in safety-related systems is being analyzed in upcoming international standardization activities such as ISO/IEC TR 5469 and ISO/IEC AWI TS 22440 being developed by ISO/IEC joint working group JWG 4.

The examples in Table 1 describe some typical safety use cases that can benefit from AI.

Use caseSurveilled machinesDetecting position and movement of humans toward a machine safety zone.conveyor belt, high-speed door, collaborative robot, autonomous mobile robot, forkliftDetecting machines moving in the direction of a forbidden area or in the direction of other machines; machines moving at excessive speed or excessive strength.autonomous mobile robot, forklift Detecting defects or material that could cause safety hazards based on norms such as ANSI/RIA R15.08, and so on.autonomous mobile robot, train, airplane Detecting and classifying worker or object poses and movement.Factory, warehouses Table 1. Use cases and machines that can benefit from AI proactive safety functions

The following examples of responses to those proactive safety functions are ordered by increasing level of intervention:

NVIDIA IGX Orin combines industrial-grade hardware with enterprise-level software and support to implement both inside-out reactive and outside-in proactive safety in the same platform:

Safety in action example

Here’s an example of proactive and reactive safety in action, implemented in collaboration with FORT Robotics and Protex.AI. 

Workflow diagram shows warn zone, stop zone, and communication channels to safety protocol elements.Figure 3. Proactive and reactive safety in action

In the example, an HD IP camera connected with Ethernet to NVIDIA IGX running Protex AI stack detects a worker moving toward an industrial mobile machine. 

Photo of FORT robotics warn and stop zones labeled.Figure 4. Overall setting with camera and safety zones

Still of the camera feed in the safety stack. Figure 5. Setting safety zones in Protex.AI user interface

The Protex edge computer vision app uses NVIDIA DeepStream 6.2 and the Protex model has been trained with NVIDIA TAO Toolkit and quantized with NVIDIA TensorRT. 

Screenshot of Protex.AI user interface showing how easily the safety zone is defined and the camera-person-zone-event chain built up. Figure 6. Proactive safety stack and the camera-person-zone-event chain

Get started

For more information, see NVIDIA IGX Orin and be sure to sign up for the Functional Safety for Industry 4.0: Keeping Supply Chains Safe, Secure, and Efficient using AI at the Edge GTC session.

Source:: NVIDIA

Exit mobile version