Assessing increasingly sophisticated advanced driver assistance systems and autonomous driving technologies requires manufacturers and testing organizations to analyze large volumes of camera, sensor and vehicle data. During on-road testing, engineers have traditionally reviewed footage, identified road signs and conditions, compared them with dashboard data and assessed the vehicle’s response. This manual process is difficult to scale without increasing engineering resources.

IVEX was developed to address this challenge using computer vision, AI and cloud-based data processing to automate aspects of on-road vehicle safety testing. The platform can interpret the vehicle’s surroundings, including speed limits and road signs, read information displayed to the driver and compare these data sources. The analysis can then assess whether the vehicle recognized relevant road information and responded according to the applicable safety protocol.

The company uses a human-in-the-loop model, enabling engineers to review its output and retain control over the final assessment. The technology can reduce engineering time spent analyzing data by more than 95%; in one example, it processed 400 hours of multi-sensor driving data using eight hours of engineering time.

“Vehicle safety testing generates enormous amounts of data, but much of its analysis still depends on repetitive manual work,” said Mario Torres, co-founder and CEO of IVEX. “AI and rigorous mathematical algorithms allows us to make the testing process significantly faster and more scalable while keeping engineers in control and ensuring that every result remains traceable to the original data. With this funding, we can deepen the capabilities of our platform and bring IVEX technology to North America and Asia.”

IVEX’s long-term goal is to develop an AI-powered platform for automotive safety testing, with different safety protocols, testing requirements and vehicle functions operating as applications. The company says its underlying models will improve as the platform processes data from additional vehicles and driving environments, supporting the development of further safety-testing applications.

Related news, Keysight and University of York partner on AI safety validation for SDVs

In computer vision and robotics, ensuring that AI systems remain reliable under real-world conditions is a growing challenge. Deep neural network-based vision systems are increasingly used in safety-critical applications such as autonomous driving, where misinterpreting a traffic sign could potentially lead to unsafe decisions. Everyday wear and tear can subtly alter traffic signs, raising questions about whether naturally occurring damage could also expose vulnerabilities in AI-based recognition systems.

To understand and address the problem, a research team led by associate professor Seong Tae Kim from Kyung Hee University and assistant professor Hong Joo Lee from SeoulTech developed Adversarial Wear and Tear (AdvWT), a framework that exploits natural wear and tear as an adversarial signal.

“We focused on traffic signs because they are exposed to weather and environmental damage throughout their lifetime, and their accurate recognition is essential for safety-critical applications,” explained Dr Lee. “Unlike temporary optical attacks, natural deterioration can persist until a physical object is repaired or replaced.”

To create AdvWT, the team trained a generative image-to-image translation model to learn the visual characteristics of damaged and undamaged traffic signs. The model, based on StarGAN-v2, learned a latent “damage style” representation that can reproduce diverse forms of realistic deterioration while preserving the identity and meaning of the sign.

By progressively adjusting the damage representation, they generated signs that looked naturally degraded but were more likely to be misclassified by an AI system. In a human study involving 32 participants, AdvWT-generated images received high naturalness ratings, closely matching the perceived realism of real damaged traffic signs.

The framework was evaluated against two traffic-sign datasets and tested against eight recognition architectures. Across the evaluated models and datasets, AdvWT achieved near-perfect attack success rates on lightweight CNNs such as ResNet-18 and MobileNet, while also remaining effective against transformer-based models. AdvWT also achieved the highest average transferability across most tested model combinations, suggesting that the adversarial perturbations could transfer across different model architectures.

To test the framework in a physical setting, the researchers printed clean and adversarial speed-limit signs and photographed them under different distances, viewing angles as well as indoor and outdoor conditions. The resulting images remained effective at misleading the traffic-sign classifier, showing that the adversarial effect persisted after printing and recapturing under varied physical conditions.

The researchers also demonstrated that the same bidirectional model could be used to restore naturally damaged traffic signs, suggesting potential applications beyond adversarial testing. Importantly, training models with AdvWT-generated damaged signs improved their ability to generalize to real-world damaged traffic signs, suggesting that simulated natural deterioration could be used to identify and strengthen weaknesses in vision systems.

“Building reliable AI requires more than improving average performance. It requires continuously identifying where AI systems fail, understanding why those failures occur, and using those insights to make the systems more robust,” said said Dr Kim. “Over the next five to 10 years, research in this direction could play an important role in developing AI systems that can be deployed more reliably in real-world, high-stakes domains such as healthcare and finance.”

Recent news, AI co-driver turns spoken observations into structured test data

This next phase of Nigel AI introduces prompt-based code generation within NI LabView and automated sequence generation in NI TestStand, enabling engineers to move from design to validation more quickly while maintaining the rigor required in mission-critical applications.

As product complexity increases and validation cycles grow more demanding across industries, engineering teams face pressure to deliver higher throughput without sacrificing quality.

Emerson’s approach embeds context-aware AI directly into the test environments engineers already use, helping streamline development while ensuring robust security, alignment with regulatory requirements such as the Cyber Resilience Act (CRA) and strong engineering oversight.

“In one year, Nigel has evolved from an AI advisor to AI author, showing the strength of the NI platform and its ability to integrate the most in-demand technologies quickly for the benefit of our customers,” said Ritu Favre, president of Emerson’s test and measurement business.

“By enabling purpose-built AI to generate code and test sequences within workflows, we’re accelerating productivity while ensuring engineers remain in control of validation and decision-making.”

With this release, Nigel AI is now available across the broader NI LabView+ Suite, bringing AI-driven workflows to NI InstrumentStudio, NI FlexLogger and NI VeriStand. The expansion brings context-aware intelligence into interactive measurements, sensor-based datalogging and hardware-in-the-loop test environments. Engineers can apply AI capabilities consistently across the full test lifecycle – from initial development through system validation and deployment.

Nigel AI is purpose-built for engineering environments and operates directly within structured test workflows. It leverages modular instrumentation, open software systems, configurable workflows, engineering code and structured data to produce outputs aligned with real-world system constraints.

Related news, Applied Intuition launches Dana agentic platform for physical AI

With these tools, agents can write MATLAB code, run it in a live session, examine outputs or errors and iterate toward a correct result, while engineers remain responsible for validating results and applying their expertise.

The new capabilities are aimed at MATLAB users, AI engineers building agent-driven workflows and platform teams managing AI-assisted engineering environments. Because agents run code directly in MATLAB, their outputs are based on deterministic computation, numerical analysis and executable models rather than only language-model-generated text. Engineers can check results by reviewing outputs, comparing them against expected behavior and refining workflows through iteration – standard practices in engineering development.

“As organizations adopt agentic AI in model-based design and engineering, the focus is shifting from code generation to reliable execution within established multi-disciplinary toolchains,” said Diego Tamburini, AI practice director at CIMdata.

“Engineers remain responsible for defining problems, validating outcomes and maintaining oversight, while AI agents increasingly handle iterative and repetitive tasks – augmenting human efficiency and effectiveness. This reinforces the importance of human-in-the-loop workflows, where real execution and validation underpin trust in AI-driven engineering processes.”

MATLAB MCP Server and MATLAB Agentic Toolkit are open-source packages, enabling developers and organizations to inspect, extend and incorporate agent‑based workflows with MATLAB in their own environments using agentic tools such as Claude Code, GitHub Copilot, OpenAI Codex and Gemini CLI. This supports interoperability across diverse AI agent frameworks and positions MATLAB as foundational infrastructure for emerging agent‑based engineering ecosystems.

“AI agents are most effective in engineering when they can directly interact with the tools used for design, simulation and analysis,” said Seth DeLand, generative AI product manager at MathWorks.

“By enabling agents to execute and iterate MATLAB workflows, we’re connecting AI-driven iteration to the same computational environment engineers use to develop and validate their work. This allows teams to move from LLM generated code to executable, testable results within a consistent engineering framework.”

Recent news, Toyota Gazoo Racing to enter hydrogen fuel cell Hilux in 2027 Dakar Rally

The system combines an FPGA core for real-time data processing with interchangeable SerDes (serializer/deserializer) interface devices, enabling engineers to reconfigure test setups as display and camera requirements change.

The launch is the latest addition to Emerson’s NI PXI platform, an open, modular test system, and extends its capabilities for automotive testing applications. The system includes the NI PXIe-1480 FlexRIO multiprotocol vision carrier module, which supports interchangeable interface devices for testing automotive displays and cameras. The first available interface devices support FPD-Link IV, a high-speed automotive video interface used in ADAS and in-vehicle display applications.

As vehicle architectures shift toward centralized electronic control units, engineers must validate multiple high-speed interfaces on increasingly complex systems. Conventional display test methods, often relying on cameras or visual inspection, limit scalability and introduce variability. By providing a direct interface to ECU and display SerDes input/output (I/O), the modular PXI-based approach improves test coverage through automated, deterministic testing. The result is a single, flexible platform that streamlines validation and improves test repeatability.

“The increasing complexity of automotive electronics requires a more flexible and scalable approach to validation,” said Sam Burhans, chief product manager with Emerson’s test and measurement business. “This modular architecture allows engineers to adapt their test systems to evolving interface standards without redesigning their entire setup.”

The NI PXIe-1480 carrier module serves as the foundation for this architecture, supporting interchangeable interface devices tailored to specific testing requirements.. Emerson plans to extend the platform later this year with additional interface devices supporting a broader range of camera and display protocols.

The solution is designed for engineers in automotive validation and production test environments, including Tier 1 suppliers and original equipment manufacturers, supporting applications across in-vehicle infotainment, ADAS and autonomous driving systems.

Recent news, Beamr partners with Intempora to bring ML-safe video data to RTMaps

As modern vehicles become more electronically complex, traditional OBD2 code readers are often insufficient for advanced diagnostics. The increasing number of electronic control modules and evolving vehicle communication systems have created challenges for technicians who need deeper access to vehicle systems beyond basic fault code reading. The Foxwell NT919 BT offers an integrated solution combining fault identification, component testing, system diagnostics and repair assistance.

From fault codes to fault verification

Many vehicle issues do not immediately generate clear diagnostic trouble codes but instead appear as abnormal component responses, reduced performance or intermittent failures. Relying only on fault code reading can make it difficult to determine the actual cause.

The bidirectional scan tool NT919 BT supports more than 3,000 bidirectional active tests, enabling users to send commands to supported vehicle components and observe their actual responses. Technicians can test components such as cooling fans, fuel injectors, ABS pumps, relays and lights without removing parts, helping verify potential faults more efficiently.

VIN recognition and full-system diagnostics

The NT919 BT features auto VIN scanning, which quickly identifies vehicle information and loads the corresponding diagnostic menu, reducing manual vehicle selection steps for technicians who frequently work on different vehicles.

The device supports diagnostics across multiple vehicle control systems, including engine, transmission, ABS, SRS, BCM, TPMS, HVAC and steering systems. Users can read generic and manufacturer-specific trouble codes and view live data to gain a more complete understanding of vehicle conditions.

ECU coding and maintenance services for advanced repairs

Modern vehicles increasingly rely on software-controlled systems, making module configuration, adaptation and calibration important parts of professional repair procedures.

The NT919 BT supports ECU coding, enabling users to perform vehicle configuration adjustments, module matching and related settings on supported vehicles. It also provides more than 34 maintenance services, including oil reset, EPB service, ABS bleeding, battery registration and steering angle sensor calibration.

Support for next-generation vehicle communication protocols 

The NT919 BT supports CAN-FD and DoIP communication protocols to work with newer vehicle electronic architectures. DoIP support covers vehicles including BMW F/G chassis models, Volkswagen, Volvo, Land Rover and Nissan, while CAN-FD support covers brands including GM, Chrysler, Porsche, and Volkswagen.

For certain FCA (Chrysler, Dodge, Jeep and Ram) and Renault vehicles from 2018 onward, the device offers gateway access through in-app registration; manufacturers charge a separate access fee for this, with Foxwell providing only the registration platform.

The NT919 BT features an 8in touchscreen, Android operating system, 4GB RAM, 64GB storage and Bluetooth 5.0 connectivity, providing flexible operation for workshop and mobile diagnostic environments.

Recent news, RUF debuts 1,000hp boxer-eight engine at Goodwood Festival of Speed 

The MagicCAR Compact Plus covers network, infotainment, functional, quality and endurance testing for ECUs, supporting OEMs and suppliers across the development process from initial validation through to series production.

The compact benchtop test system features an integrated, programmable power supply with switchable terminal signals and precise current measurement. At the same time, it provides realistic residual bus simulations, including diagnostics and SecureOnboardCommunication (SecOC), across all relevant automotive communication protocols: CAN-FD, LIN, FlexRay and Automotive Ethernet.

In combination with expandable fault modules, targeted interface and robustness tests can be carried out. Critical communication states can be simulated and analyzed in a reproducible manner to verify the behavior of ECUs under realistic fault conditions. And for extended test scenarios, additional external devices such as loggers, multimeters, oscilloscopes or signal generators can be integrated into the system.

The MagicCAR Compact Plus incorporates an integrated onboard logging function, recording communication data from ECU interfaces in real time to an internal SSD. This enables continuous recording at high data rates while reducing the need for external logging hardware, particularly for data-intensive applications such as automotive Ethernet or domain controllers.

Related news, Kvaser launches PCIe 4xLIN four-channel LIN interface

Designed for desktop, industrial and hardware-in-the-loop (HIL) environments, the low-profile PCIe card integrates galvanic isolation with synchronized timestamping, giving engineers reliable, system-level visibility when testing and debugging LIN networks.

The Kvaser PCIe 4xLIN provides four individually galvanically isolated LIN channels through a single PCIe card. FPGA-based Kvaser LIN-IP technology enables communication up to 20 kbaud with 1µs timestamp resolution, enabling developers to capture timing-sensitive events with high precision and improve visibility during debugging and validation. The interface also supports simultaneous operation with multiple Kvaser interfaces, enabling test setups to scale as system complexity grows.

Kvaser CCO Thomas Roberts said, “LIN continues to play a critical role in modern vehicle architectures, particularly as the number of distributed electronic functions grows. As systems scale, the challenge isn’t just accessing LIN data but in capturing and correlating it reliably across channels and interfaces. That’s exactly what the PCIe 4xLIN is designed to address.”

Faster integration with existing tools 

The tool is fully compatible with applications built on the standardized Kvaser LINlib, and integrates into existing setups without requiring software changes. Applications developed for other Kvaser LIN hardware can run without modification.

The interface supports both Windows and Linux and works with the Kvaser CANlib SDK, which includes LINlib and other APIs, examples for C, C++, C#, Python, Delphi and Visual Basic. It is installed directly in a PCI Express slot and powered through the host system, reducing cabling and simplifying integration in compact systems and test racks.

Accurate, system-wide time synchronization 

The tool supports Kvaser’s software-based TimeSync, helping engineers maintain synchronized timestamps across multiple devices and systems. In distributed test environments, this improves event correlation across LIN and CAN networks while reducing manual log analysis.

On Linux platforms, the interface exposes a PHC device for time synchronization to a host-referenced clock source, using standard tools such as phc2sys. This provides a robust foundation for synchronized logging, automated testing and accurate event analysis across connected systems.

Designed for demanding environments 

The Kvaser PCIe 4xLIN has been engineered for industrial and embedded systems, and operates in temperatures from –40°C to +85°C. Its low-profile PCIe design includes both short and long mounting brackets for flexible installation.

A single 26-pin high-density D-SUB connector provides access to all four LIN channels, enabling compact system design. Typical applications include ECU development, automotive body electronics, HIL testing, multi-channel logging and industrial LIN communication systems.

Related news, Göpel Electronic introduces universal control solution for single and endurance tests of vehicle components and interiors

To address this requirement, Göpel Electronic has developed the SmartController – a universal control solution designed for demanding individual or endurance tests, presentations and functional demonstrations, as well as laboratory applications that combines modern hardware, flexible software architecture and intuitive operating concepts in a compact system.

The SmartController is based on the proven Serie62 G CAR 6281 hardware platform and offers a modular architecture with versatile interfaces. With this, Göpel Electronic has created a universal hardware foundation that can be used across projects and customized as needed.

The SmartController supports up to eight bus interfaces, providing comprehensive support for automotive communication standards such as CAN, CAN-FD, LIN and Automotive Ethernet, including residual bus simulation, diagnostics and monitoring. The integrated Ethernet and wi-fi connectivity enables seamless networking and easy integration.

The generic software framework architecture ensures that future projects and new requirements can also be implemented efficiently. This makes the SmartController suitable for both long-term endurance tests and dynamic development and presentation environments.

A key feature is the tablet control framework, which enables modern and convenient operation. Its uniform structure enables use in various projects while maintaining a consistent user experience. Features such as the Sticky Keys input assistance (Touch ‘n’ Click/Make ‘n’ Break) support ergonomic and safe operation, Göpel said.

Recent news, Nio opens new UK R&D center in Oxfordshire

As hydrogen, methanol and ammonia gain relevance alongside conventional fuels, testing strategies must evolve to address new physical properties, safety risks and performance expectations. This shift is driving demand for modular, application-specific test systems capable of replicating real-world operating conditions with high precision.

A key challenge lies in the behavior of fuel-contact components under varying thermodynamic and media-specific conditions. Differences in viscosity, compressibility and reactivity between liquid and gaseous fuels require adaptable test environments. Modern systems must therefore combine hydraulic, mechanical and electronic testing capabilities within a single platform.

From component testing to system understanding

Advanced test systems now go beyond basic functional checks, enabling detailed analysis of flow rates, pressure behavior, leakage and long-term durability across components including valves, injectors, pressure regulators, rails and pumps. For injection systems, precise measurement of injection quantities and dynamic flow characteristics is critical.

One example is pressure-based measurement for injection analysis. Systems such as the IAV Cross Injection Analyzer determine injection rates from pressure changes in a fuel-filled channel, enabling dynamic, reproducible measurements without moving parts, reducing maintenance needs while supporting high-resolution data acquisition and real-time integration into digital test environments.

Lifetime and endurance testing also play a central role, with components subjected to mechanical and electronic stress scenarios to assess wear behavior and failure mechanisms over time; insights essential for validating designs intended for series production.

Simulating real-world conditions

A defining feature of modern test infrastructure is the ability to simulate realistic and extreme environmental conditions. Temperature-controlled test chambers enable testing across a wide range – from -76°F to +302°F (-60°C to 150°C) – while additional parameters such as humidity and ambient pressure can be adjusted to replicate specific operating environments.

The combination of environmental simulation with functional testing allows engineers to identify critical interactions early in the development process.

Complementing physical testing, feasibility studies and pre-validation in dedicated development labs help refine test procedures for new applications. This is especially important when dealing with emerging fuels, where standardized testing protocols may not yet exist.

Safety engineering as a core design principle

The introduction of alternative fuels adds significant complexity to safety engineering. Substances such as hydrogen, methanol and ammonia can form explosive atmospheres under certain conditions, making comprehensive hazard analysis essential.

As a result, modern test systems incorporate explosion protection concepts aligned with international standards such as ATEX and IECEx, alongside functional safety requirements defined by Performance Level (PL) and Safety Integrity Level (SIL).

Safety considerations extend beyond the test bench itself. Infrastructure factors such as ventilation systems, gas detection and facility layout are equally important. Effective solutions require a holistic approach that integrates test system design with the surrounding environment, often supported by early-stage risk assessments and coordination with local authorities.

Hydrogen testing and electrolyzer development

Hydrogen technologies add further complexity, particularly in electrolyzer development, where the focus shifts from individual components to entire systems, including single cells and stacks.

Testing requires combining electrical and fluid-based measurement techniques, assessing electrical performance, efficiency and degradation alongside flow distribution and thermal management – a convergence sometimes described as the merging of electronic and fuel testing.

Emerging test concepts support multiple electrolyser technologies, including PEM (proton exchange membrane), AEL (alkaline electrolysis) and AEM (anion exchange membrane). Modular architectures allow scaling across power classes and varying instrumentation levels, suiting both research and industrial pre-series validation.

Modularity as a strategic advantage

Given the uncertainty surrounding future fuel pathways, flexibility has become a key requirement. Modular test systems enable manufacturers to adapt to changing technologies without the need for entirely new infrastructure. Different test methods – such as flow measurement, leak testing and endurance testing – can be combined within a single system and expanded as requirements evolve.

This approach not only reduces investment risk but also accelerates development cycles. By enabling parallel testing and rapid reconfiguration, modular platforms support faster iteration and more informed decision-making.

Outlook

As propulsion technologies diversify, testing is becoming increasingly central to bridging innovation and industrialization. Generating reliable, application-specific data under realistic conditions is critical for performance, safety and regulatory compliance.

The trend is moving away from standardized solutions toward specialized, adaptable test environments, with modularity, safety integration and cross-disciplinary testing capabilities emerging as defining features of next-generation validation systems.

In recent news, Nio opens new UK R&D center in Oxfordshire