In an innovative study, researchers apply cross-modal interaction and selection mechanisms to robotic visual–inertial trajectory localization. This approach not only improves localization accuracy for robots under GNSS-denied conditions but also effectively enhances the robustness of the algorithm in complex environments.

Beijing’s metro could quietly carry freight during off-peak hours without new tunnels, a joint timetable–rolling-stock model shows. Flexible train compositions and selective skip-stop patterns cut operating costs by up to 25% while keeping passenger delays minimal.

Discover how human adipose tissue can be transformed into functional organoids representing all three germ layers—mesoderm, endoderm, and ectoderm—using a novel, streamlined method. This new study published in Engineering highlights the potential of adipose tissue for regenerative medicine, offering a scalable and clinically relevant alternative to traditional organoid generation techniques.

This article explores the potential of large language models (LLMs) in reliability systems engineering, highlighting their applications in improving industrial efficiency and flexibility. It also examines the challenges, including data deficiencies and complexity, and suggests future directions for integrating LLMs into complex engineering processes.

A new study in Engineering introduces a novel MRI technique enabling real-time, artifact-free navigation of magnetic robots. Using a multi-frequency dual-echo sequence, researchers achieved a 30-millisecond repetition time, allowing precise control and high-resolution imaging. Demonstrations in maze, phantom vessel, and in vivo trials highlight its potential for minimally invasive therapies.

New review shows that guiding pollutant motion with electric fields and 3D-printed hierarchical pores can slash fan energy and double removal rates, offering a practical blueprint for next-generation, low-resistance air purifiers.