What Are Cyber Physical Systems?

Cyber physical systems (CPS) are engineered systems that integrate computational processes with physical processes through tight feedback loops between sensing, computing, and actuation. As defined by NIST Special Publication 800-171r3, they comprise interacting digital, analog, physical, and human components engineered for function through integrated physics and logic. The computational layer monitors and controls physical behavior in real time, while the physical environment in turn constrains and shapes computation, making the two inseparable at a design level.

CPS emerged as a recognized field around the mid-2000s as embedded computing, wireless networking, and sensor technology matured enough to support real-time feedback across physical systems at scale. The discipline draws on control theory, embedded systems design, networking, and formal methods for verification. While the Internet of Things (IoT) and CPS share many enabling technologies, the emphasis in CPS is on tight timing guarantees and safety-critical control, whereas IoT frameworks often prioritize data collection and connectivity.

Sensing, Computation, and Actuation

The architecture of a CPS centers on a closed loop: physical sensors convert real-world measurements into digital signals, embedded processors or networked computers execute control algorithms, and actuators translate computational outputs back into physical effects. Sensors may measure temperature, pressure, velocity, current, or chemical concentration depending on the application domain. Actuators include electric motors, hydraulic valves, heating elements, and robotic joints. The control loop's cycle time, often measured in microseconds for industrial machinery or milliseconds for vehicle systems, determines the real-time requirements imposed on the computing substrate.

Embedded processors in CPS must satisfy both functional correctness and temporal correctness. A control output that is logically right but delivered late can cause instability or physical damage. This makes deterministic execution, predictable interrupt latency, and real-time operating systems design considerations that do not arise in conventional enterprise computing.

Networked Control and Feedback

CPS at scale connect many physical subsystems through communication networks, introducing challenges that do not appear in isolated embedded control. Network delay, packet loss, and jitter can degrade closed-loop stability in ways that classical control theory must be extended to address. Networked control systems research addresses these problems through design techniques such as event-triggered control, where actuator updates are sent only when state variables cross defined thresholds, reducing communication load while maintaining stability bounds.

NIST's Framework for Cyber-Physical Systems identifies trustworthiness as a cross-cutting concern spanning security, safety, reliability, resilience, and privacy. In large CPS deployments such as smart grids and water treatment networks, the failure of one networked node can propagate across many interconnected subsystems, requiring architectural isolation and graceful degradation strategies.

Security and Resilience

The convergence of computation and physical infrastructure makes CPS an attractive target for adversarial interference. A successful attack on a CPS does not merely compromise data; it can cause physical damage, service disruption, or harm to personnel. The Stuxnet worm, discovered in 2010, demonstrated that software attacks could physically destroy industrial equipment by manipulating programmable logic controllers. Security measures for CPS must account for the constrained computing resources of embedded devices, the long operational lifetimes common in industrial and infrastructure settings, and the safety implications of delayed updates or patches that could momentarily disrupt control functions. Research on CPS security published through IEEE Xplore addresses detection methods and resilient architectures designed to maintain safe physical operation even under cyber attack.

Applications

Cyber physical systems are deployed across a wide range of sectors, including:

• Smart electrical grids, integrating sensor networks with automated generation and demand management
• Autonomous and semi-autonomous vehicles, where real-time sensing drives braking, steering, and powertrain control
• Industrial automation and robotics in manufacturing environments
• Building automation systems for HVAC, lighting, and access control
• Medical devices and patient monitoring systems requiring reliable physiological feedback loops