The internet of things, part of the cyber-physical systems revolution, is, among other things, making our buildings, cities and grids more efficient. The array of connected objects that make up these systems are in constant communication with one another, and with central processing units, to create a data-rich picture of the world around them.

"If you can’t measure it, you can’t improve it," William Thomson, Lord Kelvin

Once these systems have created this digital, or cyber, version of the their environment, they can enact physical changes to bring about greater efficiency. Occupancy data, for example, tracks the movement of people within a smart building, turning off unnecessary systems behind them. As we race into this future, however, who or what is making sure that the cyber-physical systems themselves are as streamlined as they can be?

The somewhat surprising reality is that the cyber-physical systems technologies were developed within manufacturing systems that were not designed to accommodate them. In the face of a revolutionary technological shift we seem to have neglected the opportunity to revolutionise the way we create the technology itself.

"The register-transfer-level (RTL) design flow for digital circuits is one of the major success stories in electronic design automation," a new paper highlights while asking, "will a durable design methodology, such as the RTL design flow, emerge for cyber-physical systems?"

The answer, it suggests, will depend on how well cross-disciplinary teams learn to manage heterogeneous and dynamic technologies across large scales while also accounting for human users.

The paper ‘Design Automation of Cyber-Physical Systems’ was put together by researchers from Michigan Technological University, Boston University, University of California, Berkeley, and University of California, Riverside. Published in IEEE Transactions in CAD, it hopes to lay the groundwork for better design in cyber-physical systems.

This is not as straightforward as it may seem. Cyber-physical systems are not just an interface between mechanical and digital components; it is therefore not enough to simply separate the two before applying new design methodology. Cyber-physical systems create an emergent space between these two elements, a space with a new set of properties and design challenges.

A cake is more than the sum of the sugar and flour in its recipe, and in the same way, cyber-physical systems are more than their mechanical and digital components. A smart grid is more than a power station, transmission, distribution, a building’s energy meter and grid software. The nuances of the network and the connectivity of each element creates a space in which prediction, adjustment as well as production and consumption assessments can happen in close to real-time.

"Sensors transmit data to the physical system, which reads the data and takes action," says Shiyan Hu, an associate professor of computer engineering at Michigan Tech and one of the papers co-authors. Typically, such transfers are streamlined and efficient, but being a cyber security specialist, Hu recognizes that the exchange is a weak link.

The very attribute, connectivity, that has brought about the revolutionary elements of this new technological movement also create vulnerabilities for the user and the system itself. Privacy and security concerns are the primary stumbling blocks of the internet of things, an unresolved issue nagging away as it tries to move forward. Despite what way seem like a rapid modernization, the cyber-physical transformation is dragging the weight of these issues along with it.

“Security and privacy have become two of the foremost design concerns for cyber-physical systems today", the paper states, "and they cannot just be bolted on as an afterthought.” As we also explored in a webinar and article last year.

Hu recommends hiring specialized experts at each stage of the design and manufacturing process. “There is no one-size-fits-all kind of cyber-security,” he claims. Consider the self-driving car, for example, we must tailor design and software of the central operating system all the way to the smartphone Bluetooth connection to the anti-lock brakes, to truly ensure a secure, streamlined system.

"Cybersecurity is part of this, but more – cyber-physical security impacts not just the network, it impacts the whole system, including the physical objects," explains Hu.

The key to making safe, dependable and innovative technologies, Hu and his team put forward, is to embrace big data. Through their paper they advocate combining model-based design with data-based learning or, to put it another way, they want to merge two existing paradigms into one practice.

They hope their results could help establish the equivalent of RTL design flow but for cyber-physical systems. This “streamlining” would also incorporate machine learning, with real-time sensors, effective communication interfaces and human-centric strategies. Before we unleash the system that may redesign society, let’s make sure that system is well designed and future proofed.

[contact-form-7 id="3204" title="memoori-newsletter"]