Fennec Engineering earns T2 qualification for Advanced Safety Acceleration Platform

Fennec Engineering has worked with clients including Amazon Robotics. Source: Fennec Engineering

Functional safety is critical for robots and related technologies to earn trust and adoption, and it must be considered as early as possible in development, according to Fennec Engineering. The company today said it has received independent certification from TÜV Rheinland for its Advanced Safety Acceleration Platform, or ASAP.

The year-long Tool 2 (T2) qualification process demonstrated that ASAP meets the rigorous requirements of IEC 61508 and ISO 13849 for software tools used in safety-critical development. Steven Marks, director of product at Fennec, led the process of assessing the platform’s methodology and end-to-end traceability for both standards.

In addition, the company codified the full process for IEC 61508 certification to create a roadmap for its customers to use going forward.

“Establishing this process is obviously beneficial for the manufacturer, but what I learned during this qualification process was how beneficial it was to the assessor,” said Marks. “Having a cohesive, traceable safety narrative makes it easier to demonstrate that safety has been systematically engineered into the product. This is just as important for building public trust as it is for achieving certification.”

“I think of our SaaS offering as the TurboTax of safety — it’s the first tool of its kind,” he told The Robot Report . “We’ve built proprietary test cells showing how safety is a competitive advantage instead of a block to innovation.”

A test cell for safety-critical development. Source: Fennec Engineering

“It’s a matter of culture versus tools,” said Justin Croyle, chief product officer at Fennec. “You have to build in performant safety systems from the beginning. Instead of thinking of functional design — ‘What are the actions I need the device to do?’ — we need to change the culture to focus on functions and hazards.”

“Physical guarding of machinery evolved into E-stop, and then from the 1980s, it became more complex with silicon,” he noted. “With software, robotics, and AI, we now have even more complex systems. We want to see a change from safety as a rudimentary system bolted on to a control that can unlock performance.”

“For example, there was a mobile robot program that I got hired into, and it was building code for actuation,” recalled Croyle. “The team didn’t have a coding standard and relied on static analysis. It had to do a massive rewrite and hire dozens of engineers. They thought they had a fault-tolerant architecture, but it turned out they didn’t. Instead of being run once every eight hours, the diagnostics had to run within 100 milliseconds. How can you do that without overwhelming your compute? That was an ‘aha’ moment for me.”