n TECHNOLOGY October 2026 www.drivesncontrols.com 20 A GREEK DEVELOPER of industrial IoT controllers has announced a technology that allows an AI agent to test and prove control code that it has generated on a real controller costing €155, and to catch any errors before the code reaches a real plant. Athens-based Erqos says that the native hardware-in-the-loop (HIL) support for its EQSP32 microPLC tackles an obstacle to using agentic AI for industrial controls: How do you know that the generated code actually works? The development avoids the need for expensive simulation rigs and test benches. All a user needs is the microPLC, a USB cable and an AI agent, whose cost may be a few tens of dollars or – with a small or locally hosted model – a few cents, or nothing at all. A proving run can replace several days of skilled engineering work. With HIL, an AI agent injects input states via a serial connection and reads the resulting outputs back the same way, while the controller runs its actual production program. The agent has to prove the behaviour of its code on real hardware against a defined test plan, rather than simply generating the code and assuming it will work. The agent writes the control program, loads it onto the PLC, applies operating conditions and faults, observes the hardware outputs, compares them against expected behaviour, corrects failures – and repeats the tests until every check is passed. According to Erqos, this is reduces the risk posed by subtle coding errors, before the controller is connected to real machinery. A hallucinated setpoint, an invented register or an off-by-one debounce fails the hardware check in exactly the way a human typo would, and the agent must fix it before the test report is complete. The mechanism is simple: the agent controls what the firmware senses, and observes what it does. Through the PLC’s built-in HIL layer, the agent injects the state of every sensor and reads back the state of every output (such as pumps, valves and alarms). The control logic itself executes unmodified on silicon, with real task scheduling, real timing, and the same background network activity that it will encounter in production. Because the stimuli are injected rather than physical, the agent reaches conditions that a wired plant cannot safely provide – for example, a sensor wire breaking mid-cycle, or an over-pressure fault at its exact threshold – deterministically and repeatably. The deeper consequence is not speed, but coverage. Writing a test plan, building a harness and running dozens of cases by hand can take days, much of it spent waiting for production-length timers. Under the pressure of deadlines, a few obvious faults may get tested, while hysteresis bands, minimum-time interlocks and fault-reset paths are quietly dropped. When the agent generates the test plan from the specification it has already written, and time compression collapses the cost of running it, one more state or edge case costs almost nothing. Injection also handles cases that are difficult, dangerous or impossible on a real plant. Verifying an over-pressure, dry-run or overtemperature trip on equipment such as a pump means driving the plant towards the failure that the logic exists to prevent – at a risk to the pump and the operator. Under HIL, it is a single injected value. Broken-wire and out-of-range faults, which would otherwise require sabotaging the loop, are injected directly, and exact trip points can be applied repeatably in a way a drifting real transmitter never can. The protection layer – the part of a controller that matters most and is tested least – becomes fully and safely verifiable. The whole suite re-runs in less than a minute after any change. Hardware-in-the-Loop is well established in industrial automation, but it has typically required dedicated simulation environments, digital twins, fieldbus emulators or specialised test hardware. This confines it to large programs and safety-critical projects. Erqos says that it reduces the same closed-loop development and verification process to a €155 PLC, a serial connection and an AI agent. HIL is available for all of Erqos’ EQSP32 controllers, and comes free with the product – not a paid tier, subscription or add-on. https://erqos.com/agentic-hil p Avnu Alliance and OPC Foundation have signed an agreement to advance testing, certification and integration of OPC UA FX over TSN (Time-Sensitive Networking), with the aim of enabling testable and certified devices that deliver multi-vendor interoperability in real industrial automation applications. The agreement combines Avnu’s work on TSN conformance, certification and interoperability, with OPC UA FX, for secure, reliable industrial communications. The two groups plan to help suppliers, integrators and users to move towards interoperable systems based on certified TSN components conforming with the IEEE/IEC 60802 TSN Profile for Industrial Automation. p Kuka claims it is the first robot manufacturer to achieve certification for Security Level 2 in accordance with IEC 62443-4-2 – for its iiQKA.OS2 robot systems and KR C5 controllers. It says it is setting a new standard for securing networked production environments and addressing the challenges of such environments – from controlled access to secure communication. With its new iiQK.OS2 operating system, it adds, it is creating the foundation for a new generation of industrial automation: more intuitive, more connected, and more secure. p The UK vehicle drivetrain developer Advanced Electric Machines (AEM) has developed a rare-earth-free, heavy-duty electric motor for the demanding commercial applications. The 200kg Titan motor is the largest and highest torque motor in AEM’s HDRM family and delivers 700kW peak power and 5,000Nm peak torque (500kW and 3,500Nm continuous). It reaches a top speed of 4,000 rpm. New elements in the motor include a compressed coil technology and improved thermal management, an extruded housing design – allowing for multiple lengths – and changeable output shafts. p Visual Components, the Finnish specialist in 3D manufacturing simulation and robot offline programming, has launched a viewport bringing Nvidia Omniverse libraries and OpenUSD into its 3D manufacturing simulation platform. FactoryLens allows users to visualise large-scale factory and warehouse layouts in real time, control materials and appearance, record video, and export configurations using Universal Scene Description (OpenUSD), all without leaving Visual Components’ software. This will help manufacturers to turn factory layouts they already use for simulation into more physically realistic digital twins. p The US real-time automation specialist RTA has upgraded its RTConnect Historian for PLCs. The new version offers expanded storage, broader PLC support, smarter data transformation and more flexible publishing. It supports Allen-Bradley FlexLogix, Micro820, Micro850, Micro870, CompactLogix, ControlLogix, SLC, MicroLogix and PLC-5E systems, and offers multiple publishing options via SQL (MySQL, Microsoft SQL Server), HTTP/JSON, WebSockets, MQTT, email, FTP and USB export. A revamped dashboard lets operators filter, visualise and export graphs quickly. www.rtautomation.com TECHNOLOGY BRIEFS AI agent proves control code on a €155 controller, before it reaches a plant Erqos’ technology proves its AI-generated control code on a real controller costing €155
RkJQdWJsaXNoZXIy MjQ0NzM=