Tech & AI

Researchers developed an AI system that optimizes how mobile charging stations serve autonomous electric vehicles, improving charging efficiency to 91.5% and reducing the number of vehicles that can be serviced by a quarter. The breakthrough suggests fleets could deploy cheaper vehicles with smaller batteries, lowering total ownership costs and accelerating EV adoption.EN

Researchers have developed a method using large language models to automatically detect runtime failures in complex software systems—problems that standard testing misses. For industries like rail and aviation where system failures carry high costs and safety risks, the approach offers a practical way to add real-time safety checks without redesigning existing development processes.EN

Researchers have mapped out a comprehensive methodology for validating prestressed concrete railway sleepers—the foundation of modern rail infrastructure—using rigorous standards and real-world testing. The work provides rail operators and manufacturers a clearer path to certification while identifying a critical material factor that changes dramatically in the first six months of use.EN

Researchers have developed an artificial intelligence framework that monitors elderly residents' daily routines through smart home sensors and automatically optimizes heating, water, and waste systems to reduce carbon emissions. As the global population over 60 doubles to 2.1 billion by 2050, the technology offers cities a scalable way to improve senior care while meeting sustainability targets.EN

Researchers have created a roadmap for manufacturers shifting to Industry 5.0, a model that balances automation with worker wellbeing and sustainability. The framework identifies which management strategies work best at each stage of adoption, potentially accelerating the transition while reducing costly implementation failures.EN

Scientists have developed a cost-effective method to manufacture high-entropy alloys—materials used in aerospace and power generation—by repurposing commercial ferroalloys and removing harmful impurities through slag refining. The breakthrough could lower production costs while maintaining the purity standards required for high-performance applications.EN

Researchers have introduced transposed Hom-Poisson algebras, a new mathematical structure that extends how scientists model dynamic systems with multiple interacting components. The advance could improve how engineers and data scientists design algorithms for optimization problems in finance, logistics, and AI systems.EN

Researchers at HRI 2026 found that robots using interactive questioning to understand human feedback outperform those that blindly follow instructions. The advance matters because it makes deploying robots in real-world settings—warehouses, hospitals, homes—faster and cheaper by reducing the need for expert trainers.EN

Researchers have mapped out how to combine solar collectors with underground thermal storage to reduce biomass dependency in district heating systems—a critical issue as fuel prices climb. The study quantifies costs and performance across real-world constraints, giving utilities concrete guidance on where and how to invest in this transition.EN

Researchers have released EXACT, a user-friendly tool that combines anomaly detection with explainable AI for industrial operations. The breakthrough matters because manufacturers can now identify equipment failures and anomalies without needing data science expertise—and critically, understand *why* the AI flagged a problem, enabling faster fixes and reducing costly downtime.EN

Researchers have developed a system that converts sensor data into images that existing AI models can analyze to detect equipment breakdowns. The approach cuts costs and downtime by catching faults early, addressing a major industrial pain point where delayed detection wastes energy and money.EN

A new analysis reveals why Sweden's industrialized construction sector thrives while America's manufactured housing languishes, despite similar technology. The study identifies four specific policy levers—from building code alignment to government-industry coordination—that could unlock cheaper housing production and address labor shortages plaguing U.S. builders.EN

Researchers propose reimagining how AI systems interact with users—shifting from rigid transaction logic to dynamic, adaptive relationships. The approach could transform product design across gaming, customer service, and robotics by creating systems that respond to context rather than simply optimize for predetermined goals.EN

A new study shows interactive systems perform better when designers set up conditions rather than pre-program outcomes. Researchers observed how users naturally adapted gameplay rhythms when constraints shifted, suggesting companies could build more flexible, responsive products by embracing emergence over rigid control.EN

Researchers found that curved image sensors reduce measurement errors by 45% compared to standard flat sensors in 3D imaging systems. The breakthrough could reshape how manufacturers design scanners for industrial inspection, autonomous vehicles, and medical imaging—potentially lowering costs by eliminating expensive lens workarounds while improving accuracy.EN

Machine learning models trained on single-vehicle data fail to reliably detect when trucks are overtaking, according to new research from Volvo Group. The finding suggests that AI-powered driver assistance systems need diverse training data across multiple vehicles to work safely—a gap that could delay the rollout of autonomous safety features in commercial fleets.EN

A new analysis of aluminum casting defects in EV manufacturing shows that air entrainment and oxide contamination—not hydrogen—are the primary culprits limiting component strength and durability. The finding reshapes quality control priorities for automakers racing to scale up gigacasting production, potentially steering billions in R&D spending toward different mitigation strategies.EN

Researchers have developed a chemical-free method to blend mixed-colored polyester scraps into uniform fabrics using pressurized carbon dioxide. The breakthrough could slash water use and chemical costs in textile recycling, a major environmental burden for an industry processing millions of tons of waste annually.EN

Researchers have pinpointed the sweet spot for blending recycled aggregates into concrete: a 12.5% replacement ratio that actually improves strength and slashes carbon emissions by over a quarter. The finding could reshape construction economics by simultaneously reducing landfill waste, lowering production costs, and meeting climate targets—without sacrificing structural performance.EN

Researchers have cracked how to turn tofu and tempeh wastewater into valuable acids used in chemicals and biofuels, opening a circular economy opportunity for food manufacturers. The process could simultaneously solve a waste disposal problem and create a revenue stream from byproducts that currently burden producers.EN

Researchers have developed a machine learning architecture that automatically recalibrates faulty sensors with greater accuracy than existing approaches. The technique could reduce costs and downtime across industries relying on sensor networks—from air quality monitoring to manufacturing—by eliminating manual calibration work.EN

Researchers have developed a simpler alternative to suspension plasma spray that can coat turbine parts with exotic materials at lower cost and greater reliability. The technique could accelerate adoption of advanced coatings in aerospace and power generation, where thermal protection is critical to engine performance and longevity.EN

Engineers have long struggled to predict how concrete fails when pressure concentrates on small areas—a critical problem for building foundations, bridge supports, and anchors. New experimental research shows that the shape and size of the steel plate transferring that load dramatically affects concrete's local strength, providing design guidance that could prevent costly structural failures.EN

A Swedish transport agency found that management teams can adopt the same agile practices used by their developers to improve decision-making speed and quality. The discovery challenges the assumption that agile tools only work for engineering teams, potentially reshaping how organizations structure their leadership operations.EN

Swedish researchers tested how many electric vehicles and rooftop solar panels a residential grid can safely handle simultaneously—and found that charging strategy matters enormously. The findings offer utilities and regulators a practical roadmap for planning infrastructure that won't buckle under the weight of mass EV adoption and distributed solar generation.EN