Add BrightSurf on Google Email

New AI method helps robots learn from the past without being trapped by it

A new AI method called OMPO (Occupancy-Matching Policy Optimization) has been developed to help robots learn from past experiences without being trapped by changing conditions. OMPO uses a combination of historical data and recent interactions to stabilize reinforcement learning and adapt to policy and physical changes.

SourceScience China Press·JournalNational Science Review·TypeComputational simulation/modeling·DateSep 1, 2026

Seoul National University of Science and Technology develops AI framework to test and improve vision-system robustness

A new AI framework, AdvWT, uses realistic fading, cracks, and corrosion to test vulnerabilities in AI vision systems. It achieves near-perfect attack success rates on lightweight CNNs and transformer-based models, and can also be used to restore naturally damaged traffic signs.

SourceSeoul National University of Science & Technology·JournalIEEE Transactions on Dependable and Secure Computing·TypeExperimental study·DateAug 31, 2026

Pusan National University develops adaptive multi-expert framework for dynamic 3D reconstruction

Researchers develop adaptive multi-expert framework for dynamic 3D reconstruction, combining strengths of multiple motion representations to improve reconstruction quality. The framework leverages complementary experts to handle heterogeneous dynamics, enabling more accurate reconstruction of complex scenes.

SourcePusan National University·JournalIEEE Transactions on Pattern Analysis and Machine Intelligence·TypeExperimental study·DateAug 27, 2026

Six-legged robot learns to walk from a stick insect

Researchers trained AI on stick insect walking cycle to find optimal walking strategy, resulting in a six-legged robot that can navigate treacherous terrain and adapt to missing limbs. The approach allows for cheaper and faster robot production, enabling potential disaster response applications.

SourceTohoku University·JournalBioinspiration & Biomimetics·DateAug 27, 2026

Scripps Research and collaborators awarded $19.5 million to establish an open-access autonomous chemistry laboratory

Scripps Research will establish an open-access autonomous chemistry laboratory with a $19.5 million NSF award, advancing AI-driven discovery and education in chemistry. The lab will utilize AI and automation to streamline chemical reaction discovery and optimization, making it more efficient and accessible to a broader community.

Five-stage roadmap brings magnetic soft robots closer to clinical use

Researchers mapped a five-stage roadmap for magnetic soft robots to advance minimally invasive medicine, including untethered and tethered designs for various applications. The review highlights the importance of standardized testing, biocompatibility validation, and clinical validation to bring these promising robots from laboratory p...

A smartphone navigation app for people with blindness and low vision

A new smartphone app called Mobilio uses AI, machine learning, and personalized audio cues to provide turn-by-turn directions, path guidance, and obstacle avoidance for people with blindness or low vision. The app completed outdoor navigation tasks 13% faster and reduced obstacle contact by 41% compared to Google Maps and a white cane.

SourceHarvard John A. Paulson School of Engineering and Applied Sciences·JournalNature Biomedical Engineering·TypeExperimental study·DateAug 24, 2026

Robotic assisted knee replacement gives the same patient outcomes as a surgeon only procedure

The RACER-Knee trial found that robotic-assisted knee replacement surgery resulted in similar patient outcomes as traditional surgery, including recovery, mobility, and pain levels. Despite increased precision, the technology did not translate to improved patient outcomes within the first year after surgery.

SourceUniversity of Warwick·JournalThe Lancet·TypeRandomized controlled/clinical trial·DateAug 21, 2026

AI framework predicts network latency trends for safer telesurgery

Researchers developed a deep reinforcement learning approach to predict network latency trends, enabling telesurgical systems to anticipate and adapt to changing communication conditions. The model achieved high accuracy in predicting direction and value of future latency changes, providing valuable information for latency mitigation.

SourceJournal Center of Harbin Institute of Technology·JournalSmartBot·TypeExperimental study·DateAug 18, 2026

One step closer to efficient autonomous dental surgery

A new robotic system demonstrates superior accuracy and geometric fidelity compared to traditional methods, preserving more surrounding bone and reducing unnecessary removal. The technology has potential to make tooth autotransplantation more predictable and less dependent on operator experience.

SourceEditorial Office of West China School of Stomatology, Sichuan University·JournalInternational Journal of Oral Science·TypeExperimental study·DateAug 17, 2026

Could this simple robot outmaneuver a human hand?

A novel robot, BioflexBot, successfully mimics fundamental hand motions and exceeds human performance in precision and range of mobility. The BioflexBot can grasp complex objects, reach long distances, and deliver objects in confined spaces.

SourceWiley·JournalAdvanced Science·DateAug 13, 2026

Pusan National University researchers develop 3D printing method for switchable soft actuators

A new study developed a 3D printable liquid crystal elastomer that switches molecular alignment to produce opposite motions on demand from a single material. This breakthrough enables the creation of programmable shape-changing materials for various applications, including soft robotics and wearable technologies.

SourcePusan National University·JournalNature Communications·TypeExperimental study·DateAug 10, 2026

An expressive robot has more to lose when it makes mistakes

A study by Drexel University found that people form a stronger bond with an expressive humanoid robot, but this trust can quickly dissipate when the robot makes errors. The researchers measured brain activity, hormones, and self-reported attitudes to understand how humans build and lose trust through collaboration with robots.

SourceDrexel University·JournalScience Robotics·TypeObservational study·DateAug 3, 2026

Seoul National University of Science and Technology researchers develop energy-saving TPMS feet for quadruped robots

Researchers at Seoul National University have developed a novel approach using porous triply periodic minimal surface (TPMS) feet and deep reinforcement learning controller, which significantly reduces battery power consumption in quadruped robots. The solution reduces energy consumption by up to 6.2% while maintaining stable locomotion.

SourceSeoul National University of Science & Technology·JournalInternational Journal of Precision Engineering and Manufacturing-Green Technology·TypeExperimental study·DateJul 31, 2026

Pusan National University study highlights federated and reinforcement learning for natural language processing

The review explores how integrating Federated Learning (FL), Reinforcement Learning (RL), and Natural Language Processing (NLP) can overcome modern NLP system limitations, such as protecting user privacy and adapting to changing environments. The study presents a unified framework that combines FL, RL, and NLP as three co-equal pillars.

SourcePusan National University·JournalComputer Science Review·TypeLiterature review·DateJul 28, 2026

Wang to receive NSF CAREER Award

George Mason University Assistant Professor Xuan Wang will receive funding for his project on knowledge transfer in robot learning. He aims to develop efficient approaches that enable robots to reuse prior knowledge, supporting safer and more affordable robotic technologies in various industries.

Pusan National University researchers develop a new smart, adaptive vibration isolator

Researchers have developed a novel hybrid control strategy for quasi-zero stiffness isolators, addressing payload variations and residual resonant peaks. The 'smart cushion' can adapt to changing loads in seconds, making it crucial for precision applications like chip manufacturing and robot handling of fragile goods.

SourcePusan National University·JournalMechanical Systems and Signal Processing·TypeExperimental study·DateJul 27, 2026

Chonnam National University researchers develop new AI system for highly accurate mapping and modeling of orchards

A novel cross-modal fusion framework integrates low-altitude drone RSI with ground robot LiDAR-inertial measurement unit (IMU) odometry to create accurate digital models of orchards. The system achieved localization accuracy on the order of a few centimeters, demonstrating robustness to seasonal variations and long-term drift.

SourceChonnam National University, The Research Information Management Team, Office of Research Promotion·JournalArtificial Intelligence in Agriculture·TypeExperimental study·DateJul 22, 2026

New spinning drone hides in plain sight

Northwestern University engineers created a drone called Phantom Twist that harnesses motion blur to blend into its surroundings. The drone spins up to 25 times per second, making it difficult for humans to see clearly, and can potentially monitor wildlife or inspect infrastructure with less disruption.

KAIST develops robot that judges its surroundings and walks, runs, and jumps like an animal

KAIST develops robot that can autonomously choose gait strategy for its surroundings, switching between walking, running, jumping, and other locomotion skills. The team generated 15.5 hours of training data using computer simulations alone, enabling the robot to move quickly and stably in real outdoor environments.

SourceThe Korea Advanced Institute of Science and Technology (KAIST)·JournalScience Robotics·TypeExperimental study·DateJul 15, 2026

Microrobots move closer to precision medicine after two decades of progress

Microrobots are designed for highly localized biomedical tasks, including targeted drug delivery, minimally invasive diagnosis, and cell- or tissue-level therapeutic intervention. Magnetic actuation is identified as a promising approach due to its ability to penetrate biological tissues and integrate with medical imaging technologies.

The use of robots increases productivity in SMEs, but does not boost exports

Researchers from Universitat Oberta de Catalunya found that robot adoption increases productivity in Spanish SMEs, especially smaller firms with fewer workers and less innovation. However, introducing robots does not directly translate into increased international business activity, as regional specialization plays a key role.

SourceUniversitat Oberta de Catalunya (UOC)·JournalJournal of the Knowledge Economy·DateJul 8, 2026

EleTac: An elephant-inspired soft robotic gripper with a sophisticated sense of touch

Researchers developed EleTac, a soft robotic gripper with high-resolution tactile sensing, to handle delicate objects. The gripper's innovative design enables it to adapt to various shapes and provide gentle forces, making it suitable for applications such as handling fruit, lab samples, and medical supplies.

SourceJapan Advanced Institute of Science and Technology·JournalIEEE Transactions on Robotics·TypeExperimental study·DateJul 7, 2026