Secure Cyber-Physical Robotic Automation for Energy-Efficient Industry 5.0

In the modern industrial world, factories are no longer just rows of machines humming in isolation. They are becoming cyber-physical systems, where the digital brain of a business process talks directly to the physical muscle of a robot. This connection is the heart of what experts call Industry 5.0, a vision that seeks to blend human ingenuity with automated precision. However, as these systems grow more complex, they face two persistent challenges: they often waste energy by running inefficiently, and they remain vulnerable to subtle digital attacks that can trick a robot into seeing things that aren't there or making poor decisions. For manufacturers, the goal is to create a system that is not only fast and accurate but also energy-conscious and secure against these hidden threats.

A team of researchers has proposed a new framework designed to solve these problems simultaneously. They call their approach ORPABPM, a system that tightly integrates business process management with robotic automation. In simple terms, business process management is the method companies use to map out their workflows, ensuring every step makes sense and follows the rules. Robotic process automation is the technology that uses software or physical robots to carry out those steps without human hands. While these two fields have existed side by side, they have rarely been joined in a way that allows a robot to understand the broader business goal while also managing its own energy use and defending itself from cyber threats. The researchers built a closed-loop system where the business plan guides the robot, the robot reports back on its performance, and the system learns from that feedback to get better over time.

To test this idea, the team constructed a hybrid testbed that mimics a real manufacturing floor. They connected a programmable logic controller, which acts as the factory's central nervous system, to computerized numerical control machinery and industrial sensors. Using specialized simulation software, they created a digital environment where they could run thousands of tasks and introduce specific types of attacks to see how the system would react. The experiments involved workloads of up to 72,000 tasks, allowing the researchers to measure how accurately the robots performed, how much electricity they consumed, and how well they handled errors when under pressure.

The results showed that the integrated approach significantly outperformed existing methods. When the system was running, it achieved a task-execution accuracy of 99.3 percent, meaning it completed nearly every job correctly without needing human intervention. In terms of energy, the framework proved highly efficient; it managed to complete up to 46,000 tasks while consuming only 135 kilowatt-hours of power. This level of efficiency was maintained even as the workload increased, with the system keeping its throughput efficiency at approximately 95 percent even after 40,000 tasks were finished. These numbers suggest that by aligning the business strategy with the physical execution of the robot, companies can do more work with less energy.

Security was another critical area of focus. The researchers specifically tested the system against two common types of cyber attacks: "blind-spot" attacks, which try to hide information from the robot's sensors, and "planner" attacks, which attempt to corrupt the robot's decision-making logic. In previous systems, such attacks often caused a sharp rise in errors. In this new framework, the error rates remained incredibly low, limited to just 0.035 percent for blind-spot attacks and 0.053 percent for planner attacks. The system achieved this by constantly monitoring the robot's sensory inputs and planning behavior, instantly detecting when something was wrong and adjusting its actions to stay on course.

The study also included a detailed breakdown of why the system worked so well. By removing specific parts of the framework one by one, the researchers found that the energy-saving component was responsible for the largest gains in power efficiency, while the security component was the key to maintaining resilience against attacks. This confirmed that the system's success relied on the combination of all its parts working together. The researchers noted that while the technology is robust, it does not yet solve every human challenge, such as the need for staff training or organizational changes required to adopt such advanced tools.

Ultimately, this work demonstrates that it is possible to build industrial robots that are not just automated, but also intelligent, energy-aware, and secure. By treating the business process and the physical robot as a single, connected entity, the researchers have shown a path toward manufacturing that is both sustainable and resilient. The findings suggest that future factories could operate with a level of precision and efficiency that was previously difficult to achieve, provided the systems are designed to listen to their environment and protect themselves from digital interference.