Industry Briefing
A single destination for timely, editor-curated robotics news from around the world.
Cambridge Team Develops AI System to Generate 10,000 3D Models in One Day
A collaborative research team from Cambridge and Oxford has developed an innovative AI system named Articraft, which can produce more than 10,000 interactive 3D models within a 24-hour timeframe. This groundbreaking technology employs a specialized software development kit (SDK) that enables large language models to generate code for 3D object creation directly, eliminating the need for conventional modeling software. This advancement not only enhances efficiency but also significantly reduces costs associated with 3D modeling.
3D Modeling AI Technology Machine Learning Robotics
DJI Modify update makes 3D drone data easier to use
DJI has announced a significant update to its DJI Modify software, enhancing its capabilities for users involved in the creation and management of large-scale 3D models. This update, which was released recently, focuses on streamlining workflows to facilitate the processing, editing, and sharing of complex 3D data. By implementing these improvements, DJI aims to address the challenges faced by professionals in various industries who rely on precise and efficient 3D modeling. The enhancements are expected to provide users with a more intuitive experience, ultimately increasing productivity and collaboration in projects that require detailed 3D representations.
News
Chinese Company Wins International AI Competition with Innovative FAM-1.3 Model
At the CVPR 2026 Embodied AI Workshop, ZK Fifth Epoch, a Chinese technology company, achieved a significant victory by winning the ARNOLD Challenge with its innovative FAM-1.3 model. Competing against prominent teams from Nvidia, UC Berkeley, and Stanford, ZK Fifth Epoch's model stands out for its introduction of a '3D World-Action Model' paradigm. This advancement enhances the safety and intelligence of robotic operations by enabling the prediction of actions in three-dimensional space and identifying potential risks prior to execution. The event, held recently, showcased cutting-edge developments in embodied AI, highlighting the growing competition and innovation within the field.
Embodied AI Robotics 3D Modeling Machine Learning
Cornell’s insect-inspired 3D model could allow flapping-wing robots to fly stably
Researchers at Cornell University have unveiled a groundbreaking 3D computational model designed to decode complex physical phenomena. This innovative model, which was developed over the past year, aims to enhance our understanding of various scientific processes by simulating intricate interactions within physical systems. The research team, led by a group of physicists and engineers, conducted extensive experiments and simulations to refine the model's accuracy and applicability. The development of this model is particularly significant as it addresses longstanding challenges in the field of physics, providing a tool that can potentially revolutionize how scientists approach problem-solving in areas such as material science, fluid dynamics, and even climate modeling. By leveraging advanced algorithms and high-performance computing, the researchers were able to create a more precise representation of physical interactions, which could lead to new discoveries and innovations. This work not only showcases the capabilities of modern computational techniques but also underscores the importance of interdisciplinary collaboration in advancing scientific knowledge. The findings of this research are expected to be published in a leading scientific journal, contributing to ongoing discussions and developments in the field.
POp-GS: Next best view in 3D-Gaussian splatting with P-Optimality
Researchers have developed a new algorithm aimed at enhancing the capabilities of 3D Gaussian Splatting (3D-GS) by quantifying uncertainty and information gain through a method known as P-Optimality. This advancement addresses a significant limitation of 3D-GS, which, despite its effectiveness in producing high-quality rasterizations, lacks the ability to measure uncertainty and information. This deficiency presents challenges for practical applications, particularly in the field of 3D-GS Simultaneous Localization and Mapping (SLAM). The team’s innovative approach seeks to improve the reliability and applicability of 3D-GS in real-world scenarios by providing a systematic way to assess information gain, thereby enhancing its utility in various technological applications. The findings were detailed in a recent paper, contributing to the ongoing development of more sophisticated modeling techniques in three-dimensional environments.
Computer vision
How Your Virtual Twin Could One Day Save Your Life
In May 2019, a cardiac surgeon at Boston Children’s Hospital successfully performed a complex heart surgery on a child with a severe congenital defect, utilizing advanced virtual twin technology. This innovative approach involved creating a detailed 3D model of the child's heart and vascular system from MRI and CT scans, allowing the surgical team to simulate various strategies and predict outcomes before the operation. The procedure was critical due to the unique nature of the child's heart condition, which had no established surgical manual. The Living Heart Project, initiated in 2014, has since guided nearly 2,000 surgeries by employing virtual twin modeling, which combines engineering principles with medical expertise to enhance surgical precision and patient outcomes. This project, now involving over 150 organizations globally, aims to revolutionize medical treatment by providing a dynamic, predictive tool that can simulate the human body's responses. The technology not only aids in surgical planning but also has the potential to streamline clinical trials. By creating virtual patient cohorts, researchers can test treatments more efficiently, reducing the time and costs associated with traditional trials. The FDA has recognized the significance of this approach, collaborating with the project to establish guidelines for in silico clinical trials, marking a significant shift in how medical innovations are developed and validated. As virtual twins expand beyond cardiac applications to other organs, they promise to transform healthcare by enabling personalized medicine and fostering a deeper understanding of patient physiology, ultimately improving treatment outcomes and patient engagement in their health management.
Cardiology Digital-twins Personalized-medicine Virtual-heart Generative-ai
Accelerating Chipmaking Innovation for the Energy-Efficient AI Era
Applied Materials is set to revolutionize semiconductor research and development with the opening of its EPIC Center in 2026, marking the largest investment in advanced semiconductor equipment R&D in U.S. history at approximately $5 billion. This initiative aims to address the pressing demands of the AI era, where companies are racing to enhance AI system performance while managing energy consumption. The EPIC Center will facilitate a collaborative environment where engineers and technologists work together from the outset, streamlining the traditional R&D workflow that has become too slow for the rapid advancements required in AI technology. By integrating atomistic modeling, process development, and validation within a shared space, EPIC aims to accelerate the transition from research to high-volume manufacturing. As AI workloads increasingly depend on efficient data movement, the center will focus on optimizing three interconnected domains: logic, memory, and advanced packaging. Innovations such as 3D devices, high-bandwidth memory, and hybrid bonding will be prioritized to enhance energy efficiency and performance per watt. The EPIC model emphasizes the need for a new approach to innovation, breaking down silos and fostering collaboration among industry leaders and academic institutions. This shift is essential to meet the challenges posed by the angstrom era, where the complexity of semiconductor manufacturing demands a more integrated and rapid development process. Through EPIC, Applied Materials aims to lead the charge in delivering the next generation of energy-efficient AI technologies.
Chipmaking Artificial-intelligence Materials-science Semiconductors
Doosan Robotics Unveils Award-Winning AI Robot Solution, as the Beginning of a New Path at CES 2026
Doosan Robotics Inc., a leader in collaborative robotic solutions, is set to showcase its latest innovations at CES® 2026, taking place from January 6 to 9 at the Las Vegas Convention Center. The company will present the Scan & Go Autonomous Robotic Solution, which has garnered two CES Innovation Awards, including “Best of Innovation” in the AI category. This AI-driven system revolutionizes large-scale composite repair by utilizing cognitive learning and laser-based 3D vision to interpret complex geometries in real time, thereby eliminating the need for traditional CAD modeling. Deployed on an autonomous forklift, Scan & Go is designed to perform intricate tasks such as sanding and grinding on large structures, enhancing production efficiency and workplace safety. In addition, Doosan Robotics will unveil its AI-Powered Depalletizing demonstration, which features real-time motion planning using Nvidia cuMotion technology. This automated solution employs 3D vision for box recognition and obstacle avoidance, significantly improving operational efficiency in logistics environments. Kevin Kim, CEO of Doosan Robotics, emphasized the company's commitment to providing cutting-edge solutions that enhance efficiency and safety across various industries. Earlier this year, Doosan Robotics announced a strategic shift towards becoming an AI robotic solutions company, focusing on future growth and technological competitiveness. The company plans to launch its AI Palletizing Solution in the first half of this year, further solidifying its position in the robotics market.
Doosan Robotics Wins Two CES Innovation Awards® 2026 Including “Best of Innovation” in AI Category
Doosan Robotics has received prestigious accolades at the CES Innovation Awards, being recognized as the Best of Innovation in the Artificial Intelligence category and as an Honoree in the Robotics category for its Scan & Go Autonomous Robotic Solution. These awards, presented by the Consumer Technology Association (CTA), highlight exceptional design and engineering in technology products ahead of CES 2026, scheduled for January 6–9 in Las Vegas. The Scan & Go solution, developed in partnership with Canada-based Maple Advanced Robotics Inc. (MARI), is an AI-driven robotic platform designed to enhance large-scale manufacturing and composite repair processes. Utilizing cognitive learning and 3D perception, the system interprets complex geometries in real time, generating optimized tool paths without the need for CAD modeling or programming. Mounted on a self-driving forklift, it autonomously navigates and operates on large, irregular structures, performing tasks such as sanding, grinding, and inspection with minimal preparation. This innovation aims to automate hazardous and labor-intensive maintenance tasks, addressing safety concerns related to noise, dust, and repetitive strain injuries. With advanced safety features, including torque-sensing collision detection, Scan & Go enhances workplace safety and productivity, earning recognition from leading wind energy manufacturers. Kevin Kim, CEO of Doosan Robotics, emphasized the company's commitment to developing immediate, on-site AI robotic solutions, marking Scan & Go as the beginning of a broader initiative to tackle real-world industrial challenges. This recognition at CES 2026 signifies Doosan's strategic transformation into an AI Robot Solution company, aimed at bolstering its global competitiveness in the robotics sector.
TARS DexHand Makes Global Debut at ICRA: Redefining Dexterous Operations with Mind-Hand Integration
At the ICRA conference held in Vienna, TARS DexHand was unveiled, showcasing an innovative 'mind-hand' integration system designed to mimic human hand movements with 21 degrees of freedom. This advanced technology enhances task execution by utilizing sophisticated data alignment and modeling capabilities. During the demonstration, the system performed real-time interactions and complex tasks, underscoring its significant potential for industrial applications and the advancement of embodied intelligence. The event highlighted the ongoing evolution in robotics and its implications for various sectors.
Dexterous Robotics Embodied Intelligence Robotics Automation AI Technology
Robot Talk Episode 140 – Robot balance and agility, with Amir Patel
Amir Patel, an Associate Professor of Robotics and AI at University College London, discussed his innovative research focused on developing robots that mimic the agility and maneuverability of cheetahs. During a recent conversation, Patel explained how his work integrates advanced robotics techniques, including sensor fusion, computer vision, mechanical modeling, and optimal control. This research aims to enhance robotic performance in dynamic environments, potentially revolutionizing applications in various fields such as search and rescue, exploration, and surveillance. By studying the natural movements of cheetahs, Patel and his team are striving to create robots that can navigate complex terrains with unprecedented efficiency and speed.
Robotics needs a service framework.
RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.
