New Functional Safety Test Bed For Autos, 5G And Drones

Testing functional safety isn’t cheap. There are billions of dollars of labs spread out across the world testing everything from electromagnetic compatibility to braking distances to crashworthiness. The fixed and operating costs of such facilities is part of why companies emphasize upfront engineering rigor: the payoff for healthy ways of working are significant. Getting the requirements and associated architecture well-defined decreases downstream problems.

However, the costs of not testing functional safety and revising the requirements can be even more extreme. Yet another automotive supplier – Garrett Motion – filed for Chapter 11 bankruptcy in September due to asbestos exposure issues from its legacy brake pads. Producing products that supposedly protect people without confirming the true effects can damn a corporation to financial hell. So finding affordable, effective alternatives – especially for cooperative systems due to the extreme nature of the test set-up – is a best, overall solution.

In walks AERPAW, or maybe more appropriately: “in flies AERPA”. The Aerial Experimentation and Research Platform for Advanced Wireless (AERPAW) was awarded on behalf of the National Science Foundation a $24 million grant over five years to build and operate the U.S.’s first aerial, wireless experimentation platform including 5G cellular technology, vehicles and aerial systems. Researchers from various universities, municipalities and corporations are cooperating on the testbed being built on North Carolina State’s campus specifically to test system interaction of vehicles and drones for various use cases.  “To really research some of these 5G concepts is much more complex than 4G,” states Sarah Yost, the Business Development Manager for National Instruments (NI
NATI
), a heavily involved partner in AERPAW. “Therein, simulations are still meaningful to 5G, but not as impactful as they were for 4G.”

The interoperative solutions and where the flying and driving community may go are fascinating. “Fire trucks or other first responders could act as a cellular base station,” states Yost. “Those vehicles would communicate with video-enabled drones that would fly over and evaluate the situation. 5G would allow that evaluation to have better controlled drones with higher-definition video with very low latencies. All of that together would, in theory, permit a very informed decision of the danger involved before sending in the first responders.”

Low latencies are the crucial element, in fact, for most of the theorized use cases. The goal for the latest 5G standard is one millisecond (1 ms) or less. “When you are controlling drones, even a half of second of delay in controlling their flight can make all the difference.” The 5G specification has a latency specification and can allow one of the vehicles onsite to act as the tower and create micro-cells or pico-cells and act as a base station to the drones or other vehicles.

Quick, local communication has always been the underlying enabler to vehicle-to-vehicle or vehicle-to-infrastructure communication – otherwise known as V2X. Fifteen years ago, the Crash Avoidance Metrics Partnership (CAMP) – which is and was a collection of manufacturers working cooperatively with the United States Department of Transportation (US-DOT) – published a research report detailing the lives that would be saved with sub-100ms communications between vehicles, which inspired Dedicated Short Range Communications to be launched in Japan (5.8 GHz) and become protected bandwidth in the U.S. (5.9 GHz). The trouble, though, is that the business case for launching a cooperative system is very tough: on Day One the newly-manufactured vehicle (a.k.a. “first mouse to the cheese”) has hundreds of dollars of equipment talking to no one for an unlikely crash-event. 5G is different, though, since it is not constrained to safety-related functions and, therein, can launch under a variety of business cases. “I know it seems unfathomable right now given Covid-19,” muses Yost, “but imagine something flexible helping with massive crowds leaving concerts or sporting events. Here creativity with machine learning, artificial intelligence, low-latency and traffic signal algorithms might enable controllers to dynamically learn, manage and mitigate the traffic patterns.”

Additionally, Yost explains that drones open up a bunch of non-traditional use cases. “Drones are being used for campus security, threat detection and overtaking moving threats within private campuses. Current examples are airports. But privatized networks might be enabled by network slicing and allow any corporation to manage its campus and divert utilization of its local bandwidth to urgent usages when needed.” Separately theorized by other connected vehicle experts slightly further into the future: flying taxis will become more prevalent, and drones interacting locally with the elevated traffic would be foreseeable for both security and delivery use cases.

So as the future unfolds and manufacturers want to test cooperative, 5G functional-safety and non-safety applications, there will be a resource available on NC State’s campus for use in 6-12 months. A worthy note for all aspiring entrepreneurs: the testbed will be open to the public so come with a creative solution, a testbed reservation, a real-world prototype and hopefully the engineering rigor that will result in safe solutions.