%52%6f%62%6f%74
-
KAIST Develops Robot That Judges Its Surroundings and Walks, Runs, and Jumps Like an Animal
An era in which robots decide "how to walk" on their own has arrived. A four-legged robot has been developed that, much like a person or an animal, autonomously chooses the appropriate gait strategy for its surroundings — changing its gait on stairs, leaping over gaps, and keeping its balance on forest trails.
KAIST (President Choongsik Bae) announced on the 16th of July that a research team led by Professor Hae-Won Park from the Department of Mechanical Engineering has developed a core control technology for four-legged robots that lets a single controller select and switch in real time among walking, running, jumping, and other locomotion skills, allowing the robot to move quickly and stably, even in real outdoor environments.
Four-legged robots move on four legs, giving them an advantage over wheeled robots on rough terrain. But in real outdoor settings, obstacles such as stairs, ledges, stepping stones, gaps, and tree branches appear one after another in different forms, meaning the ability to simply walk and run fast is not enough.
Existing four-legged robots have excelled at running quickly across flat ground or clearing simple obstacles, but they have struggled to maintain both speed and stability in real-world environments where obstacles combine in complex ways. Because walking, running, jumping, and other gaits had to be controlled individually, the robots were also limited in how naturally they could switch between them as conditions changed.
To overcome these limitations, the research team developed a new learning-based control technology called APT-RL (Action Pretrained Transformer-based Reinforcement Learning).
APT-RL is a control technology designed to enable a robot to first learn a range of locomotion skills — such as walking, running, and jumping — and then freely combine and transition among them in real-world environments as the situation demands.
Rather than filming the movements of real people or animals, the team generated 15.5 hours of training data covering a variety of gaits using computer simulations alone, in just eight minutes. That data was used to teach the robot basic movement capabilities, drawing on robot dynamics (a mathematical model of how a robot moves) and trajectory optimization (a technique for calculating the efficient path of movement). The approach is far faster and more efficient than earlier methods that relied on motion capture, a technology that records human or animal movement using sensors.
The team then applied reinforcement learning — an artificial intelligence technique in which an agent learns optimal behavior through repeated trial and error — so the robot could autonomously select and switch gaits suited to complex three-dimensional terrain such as stairs, ledges, and gaps. Finally, the team combined a depth camera (which measures the distance to objects in order to obtain three-dimensional information) with LiDAR (Laser Detection and Ranging, a sensor that uses lasers to measure the distance and shape of the surrounding environment in three dimensions), enabling the robot to recognize its surroundings and target speed in real time and choose the most appropriate walking strategy.
The team tested the control technology on its own four-legged robot, 'KAIST HOUND.' The experiments were conducted not only on an indoor obstacle course but also in real outdoor environments, including KAIST’s campus and forest trails.
KAIST HOUND moved stably across urban terrain that included stairs, grass, and slopes, as well as irregular natural terrain such as fallen trees, exposed roots, and paths covered in fallen leaves, switching gaits in real time to match the conditions. In rugged terrain with obstacles, the robot reached a peak instantaneous speed of six meters per second (about 22 kilometers per hour), demonstrating that it can achieve both fast movement and stability in real outdoor environments.
The experiments showed that KAIST HOUND autonomously selected and switched between a trot (alternating diagonal legs) and a bound (a leaping gait using the front and back leg pairs together) depending on the terrain and target speed, and that it could integrate walking, running, jumping, and ledge-clearing into a single controller.
Professor Hae-Won Park said "We expect this to become a foundational technology that expands the potential uses of physical-AI-based walking robots in rugged environments such as disaster sites, defense missions, and industrial facility inspections."
Jun-Gill Kang (affiliated with the Agency for Defense Development (ADD) at the time of the research) and Jaehyun Park, a Ph.D. candidate in KAIST's Department of Mechanical Engineering, are co-first authors of the study. Professor Hae-Won Park and Professor Seungwoo Hong from Korea University are co-corresponding authors. The research was selected as the cover paper for the July issue of Science Robotics, the world's leading academic journal in robotics, and was published on July 15 (U.S. Eastern time).
Paper title: Agile perceptive multi-skill locomotion for quadrupedal robots in the wild
DOI: 10.1126/scirobotics.adz7397
Authors: Jun-Gill Kang (the Agency for Defense Development at the time of the research, co-first author), Jaehyun Park (KAIST, co-first author), Hae-Won Park (KAIST, corresponding author), Seungwoo Hong (Korea University, corresponding author)
Related Video: https://drive.google.com/drive/folders/1306_hddGZGh7xwvWFc4B-9lLXwYisirN
This research was supported by funding from the Ministry of Trade, Industry and Resources (MOTIR) and the Korea Planning & Evaluation of Industrial Technology (KEIT) (RS-2024-00427719), as well as by the Agency for Defense Development's Future Challenge Defense Technology R&D program (912768601).
2026.07.16 View 91 -
KAIST Begins Developing the World’s First Brain-to-Robot Technology, Moving Robots by Thought and Sending Sensation Back to the Brain
KAIST researchers have begun developing a next-generation brain-robot interface platform that uses human brain signals to control an exoskeleton in real time and sends the tactile and force information sensed by the robot back to the brain.
KAIST, led by President Kwang-Hyung Lee, announced on the 25th that research teams led by Professors Kyoungchul Kong and Jung Kim of its Department of Mechanical Engineering, together with Angel Robotics Co., Ltd., have launched the world’s first bidirectional “Brain-to-Robot” system as a flagship initiative of the Korea Medical Device Development Fund (KMDF). The project runs from April 2026 to December 2032.
Professor Kyoungchul Kong is a world-renowned wearable-robotics researcher who founded Angel Robotics, a developer of walking-assist exoskeletons, and led his team to back-to-back gold medals at Cybathlon, the international competition for assistive technologies for people with disabilities. Professor Jung Kim is a globally recognized researcher who received the Scientist and Engineer of the Month Award for his work on robotic skin. Together, the two teams have formed a consortium to develop a Brain-to-Robot platform that merges neural interfaces with exoskeleton robotics.
Brain interface technologies that let users move a cursor or operate a smartphone with brain signals have already reached the stage of human clinical trials, and U.S. companies such as Neuralink and Synchron are accelerating their development. Existing approaches, however, have struggled to link actual movement and sensory feedback at the same time. They have also concentrated largely on advancing signal decoding itself, without clearly defining the target of control, namely what the brain signals actually drive and what kind of sensory information is returned.
Brain-to-Robot is designed to overcome these limitations head-on. It sets the exoskeleton itself as the control target: brain signals read the user’s movement intentions to drive the robot, and at the same time the robot’s sensory readings are delivered back to the brain. These readings include ground reaction force (the force the floor exerts on the foot), joint torque (rotational force at the joints), and tactile information. The aim is a complete bidirectional interface.
According to the research team, no fully bidirectional Brain-to-Robot system that combines exoskeleton control with sensory feedback has yet been reported anywhere in the world, and the project is expected to mark a turning point in brain interface technology.
Within this system, the KAIST teams are responsible for the core technologies. Professor Kong’s team will develop wearable-robot control and AI-based interpretation of movement intention, and will design a somatosensory interface, a system for transmitting bodily sensory information, that delivers the robot’s sensory data accurately to the Brain Chip, the semiconductor that processes brain signals.
Professor Kim’s team will develop robotic skin that senses in place of impaired sensation for people with disabilities, along with AI-based interpretation of somatosensory information.
The teams will also develop AI-based encoding and decoding algorithms that turn brain signals into robot commands and send the robot’s sensory information back to the brain. A key challenge is processing hundreds of channels of cortical signals, the neural signals generated in the cerebral cortex, while stably maintaining an ultra-low-latency closed loop, a control cycle in which signals are exchanged continuously in real time.
Commercialization of the flagship project will be led by Angel Robotics (KOSDAQ: 455900), the company founded by Professor Kong. The team plans to pursue full-cycle commercialization, from regulatory approval by the Ministry of Food and Drug Safety through to real-world deployment.
“If this technology succeeds, it will open a new rehabilitation paradigm in which people with quadriplegia can move beyond the hospital to walk on their own, pick up objects, and even feel sensation at their fingertips in everyday life,” Professor Kong said.
The research team stressed that, because this is an unprecedented and highly complex convergence technology never attempted at home or abroad, long-term safety, clinical validation, and a regulatory approval framework must advance in parallel with the technology itself. To reach the global market, they added, safety and efficacy testing, the accumulation of clinical evidence, a risk-management system, protection of brain-signal data, cybersecurity, and ethical review must all be addressed in an integrated way.
Meanwhile, KAIST is conducting a wide range of fundamental research in the field of brain interfaces. A research team led by Professor Hyung-Soon Park of the Department of Mechanical Engineering is studying wearable rehabilitation robot technologies based on neural intention-recognition interfaces, which identify users’ movement intentions from brain signals, for the effective treatment of neurological disorders. A research team led by Professor Sungho Cho of the School of Computing is developing AI-based brain-signal interpretation technologies.
A research team led by Professor Jihoon Lee of the Department of Brain and Cognitive Sciences is conducting next-generation brain–machine interface research focused on ultra-low-power bio/neural interface circuits, which connect and process biological and neural signals with low power consumption; wireless neural signal measurement technologies, which measure neural signals without wires; and on-device AI-based closed-loop neuromodulation technologies, which use cyclical control structures to exchange signals in real time.
In addition, a research team led by Professor Hyunjoo Lee of the School of Electrical Engineering is conducting research on high-resolution neural signal measurement and precision brain stimulation based on ultra-miniaturized multimodal neural electrodes, which can simultaneously measure and stimulate multiple types of neural signals. A research team led by Professor Minkyu Je of the Department of AI Semiconductor Systems is studying AI-based semiconductor integrated circuits and system technologies for next-generation neural interfaces. A research team led by Professor Jae-Woong Jeong of the School of Electrical Engineering is conducting research on high-precision brain-signal measurement, which precisely measures neural signals generated in the brain, and neuroengineering based on neural stimulation.
“This Brain-to-Robot flagship project is a world-class, highly challenging convergence research initiative led by the teams of Professors Kyoungchul Kong and Jung Kim,” said KAIST President Kwang-Hyung Lee. “KAIST has a wide range of researchers studying fundamental technologies in brain interfaces, AI, semiconductors, and robotics, and based on this foundation, we will lead innovation in next-generation Brain-to-Robot technologies.”
2026.06.25 View 1225 -
KAIST Develops Robot Learning Technology Capable of Precisely Imitating Even “Rough” Demonstrations
Robots with increasingly precise dexterity are becoming essential in everyday life and industrial settings, from assembling tiny smartphone components to assisting doctors in surgery. However, teaching robots delicate human movements has traditionally required collecting vast amounts of data at extremely fine time intervals, resulting in significant costs and time burdens. KAIST researchers have developed a robot artificial intelligence technology that can perform sophisticated tasks by autonomously adjusting precision according to the situation, even when trained only on coarsely (sparsely) sampled demonstrations.
KAIST, led by President Kwang Hyung Lee, announced on the 24th that a research team led by Professor Daehyung Park of the School of Computing has developed DiSPo, a multi-granularity manipulation model that generates fine-grained robot motions tailored to a user’s desired level of precision, even from rough human demonstrations.
Existing robot learning methods, such as Behavior Transformer and Diffusion Policy, are limited by their dependence on the time intervals of the data used during training. As a result, learning precision manipulation tasks such as screw fastening or component insertion has required collecting large volumes of high-frequency data at very short time intervals. This has significantly increased data collection costs and slowed down the inference speed of robot AI models.
To overcome these limitations, the research team combined Mamba, a state-space model capable of predicting time intervals, with a diffusion model that enables rich action representation. The team also introduced a new Step-scale factor mechanism, which allows users to directly control the time intervals used by the robot.
As a result, even when trained on only low-frequency (coarse) demonstration data, the robot can generate high-precision motions during inference without additional training by autonomously subdividing actions through a discretization process.
DiSPo achieved up to an 81% higher task success rate compared to state-of-the-art models in simulation environments. In real-world experiments using a collaborative robot, DiSPo stably performed challenging tasks such as passing a clamp through a narrow gap with only a 2.5 mm radial clearance and accurately pressing a small shutter button on a smartphone. This performance was up to four times higher than that of existing AI models.
The technology is expected to make a significant contribution to automation in a wide range of everyday and industrial service fields that require high precision, including precision component assembly, cable connection, medical surgery, and precision machining.
“This study demonstrates that robots can learn precise motions from coarse demonstrations and autonomously adjust their level of precision according to the task situation,” said Professor Daehyung Park. “Moving forward, this technology is expected to dramatically reduce data collection costs while serving as a general-purpose robot learning technology for various industrial fields, including precision assembly and medical applications.”
The study was led by Nayoung Oh, a master’s student at the KAIST Graduate School of AI, as the first author, and was presented on June 1 at the 2026 IEEE International Conference on Robotics and Automation, or ICRA 2026, one of the world’s most prestigious robotics conferences, held in Vienna, Austria.
Paper Title: DiSPo: Diffusion-SSM based Policy Learning for Coarse-to-Fine Action Discretization
DOI: https://doi.org/10.48550/arXiv.2409.14719
2026.06.24 View 915 -
KAIST Teams Win Both International Challenges at ICRA 2026 and CVPR 2026
Two research teams from KAIST have claimed first place in international challenge competitions held at the world’s premier robotics and computer vision conferences.
KAIST (President Kwang-Hyung Lee) announced that the ACDC-K Team and the Curaytor Team, both from the laboratory of Prof. Hyun Myung in the School of Electrical Engineering, won first place in international challenge competitions held in conjunction with the IEEE International Conference on Robotics and Automation (ICRA 2026) and the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2026), respectively.
The achievement highlights the global competitiveness of KAIST’s robotic perception and spatial intelligence technologies, with two teams from the same laboratory securing victories in leading international competitions across distinct research fields.
The ACDC-K Team won first place among more than 60 participating teams in the SLAM (Simultaneous Localization And Mapping) category of the Hilti×Trimble SLAM Challenge 2026, held during the Open Challenges in Robotics for Asset Inspection (OCRAIM) Workshop at ICRA 2026 in Vienna, Austria, from June 1 to 5.
Jointly organized by Hilti, Trimble, and the University of Oxford, the challenge evaluates robotic localization and mapping performance using sensor data collected from real construction sites. Participants were required to address practical challenges frequently encountered in construction environments, including non-overlapping front and rear fisheye camera configurations, low-texture indoor scenes, and rapid camera motion.
To tackle these challenges, the ACDC-K Team developed a robust visual-inertial SLAM system that fuses front and rear fisheye camera data with inertial measurements. By integrating feature-point and feature-line observations with adaptive constraints and correction mechanisms, the team achieved highly reliable localization and mapping performance in complex construction environments.
Meanwhile, the Curaytor Team won first place among eight participating teams in the Nothing Stands Still (NSS) Challenge 2026, held during the Computer Vision for the Built World (CV4AEC) Workshop at CVPR 2026 in Denver, Colorado, from June 3 to 7.
Jointly organized by Stanford University, ETH Zurich, and Oregon State University, the NSS Challenge evaluates 3D point cloud registration technologies for construction and industrial environments that evolve over time.
The Curaytor Team developed a novel multi-registration framework capable of aligning multiple LiDAR scans collected across different times and locations. The framework integrates feature extraction, correspondence estimation, robust global registration, registration confidence assessment, and change-aware refinement techniques. As a result, the team achieved highly accurate registration performance even in environments containing structural changes and dynamic objects.
“This achievement demonstrates the robustness of our visual-inertial SLAM and 3D LiDAR registration technologies in complex and constantly changing real-world environments,” said Prof. Hyun Myung. “It is particularly meaningful that our students secured first-place finishes in highly competitive international challenges hosted at two of the world’s most prestigious conferences in robotics and computer vision.”
Prof. Hyun Myung’s laboratory has consistently demonstrated excellence in spatial intelligence research. The laboratory previously won first place in the LiDAR track and ranked first among academic teams in the vision track of the Hilti SLAM Challenge in 2023. In addition, the Curaytor Team successfully defended its title in the NSS Challenge, securing back-to-back championships in 2025 and 2026.
2026.06.19 View 1873 -
KAIST Illuminates the Eyes of Humanoid Robots with Minimal Memory
<CVPR 2026 poster session. From left to right: Minseok Seo (KAIST, first author), Mark Hamilton (MIT and Microsoft, second author), and Prof. Changick Kim (KAIST, corresponding author)>
From facial recognition on smartphones to humanoid robots, computer vision technology, which serves as the eyes of artificial intelligence (AI), is widely utilized in our daily lives. A joint research team from KAIST and international institutions has developed a technology that allows AI to see the world more clearly with minimal memory, increasing GPU (Graphics Processing Unit) memory efficiency by up to 16 times. This achievement is evaluated as a core technology that will accelerate the era of humanoid robots and on-device AI.
<Overview of Upsample Anything. Given a high-resolution image, it is first downsampled to a low-resolution image and then reconstructed through test-time optimization (TTO). During this process, pixel-wise anisotropic kernel parameters are learned. The learned kernels are subsequently applied to low-resolution foundation feature maps to generate high-resolution feature maps. These feature maps are then used to perform pixel-wise anisotropic Joint Bilateral Upsampling, enabling high-quality reconstruction at high resolution>
KAIST announced on June 17th that a research team led by Professor Changick Kim from the School of Electrical Engineering, through joint research with researchers from MIT and Microsoft in the United States, has developed 'Upsample Anything,' a universal technology that can enhance the visual performance of AI even with limited GPU memory.
Following its acceptance to 'CVPR 2026,' the world's most prestigious conference in the field of artificial intelligence and computer vision, this achievement was awarded the 'CVPR Compute Gold Star' in recognition of its efficient utilization of computational resources. It was also selected as the 'Transparency Champion,' ranking first overall in the category of research process transparency and reproducibility. This is an accomplishment that widely recognizes the core elements of responsible AI research, including research performance, computational resources used, code disclosure, and experimental reproducibility.
Recently, humanoid robots, autonomous driving systems, and AI based on world models (AI models that learn and predict the physical environment and changes of the real world) have been compressing input images into low-resolution features (core information extracted from images by AI) to increase computational speed and reduce memory usage.
However, during the compression process, a problem occurs where important visual information, such as small objects, thin structures, and minute defects, is lost. Conversely, processing all images at high resolution from the beginning requires massive GPU memory and computational resources, making real-time processing difficult. This has remained an unresolved challenge for a long time in situations where small devices like smartphones or robots, where mobility is crucial, must precisely perceive their surrounding environment.
To overcome these limitations, the research team developed a training-free (requiring no additional data training) upsampling technology that restores low-resolution feature information into high resolution by utilizing the edge and structural information of the input image.
Existing technologies required a separate retraining or complex optimization process to be applied to new environments or data. In contrast, 'Upsample Anything' developed by the research team can find the optimal restoration method using just a single input image, allowing it to be immediately applied to various environments.
In addition, by compressing and utilizing only core information instead of storing and processing all visual information at high resolution, GPU memory usage was significantly reduced. Based on a 224×224 size image (approximately 50,000 pixels) widely used in AI research, the research team restored visual information close to the original with a short calculation of about 0.4 seconds, achieving a performance that improves GPU memory efficiency by up to 16 times.
This means that artificial intelligence can perceive its surrounding environment more precisely even with limited computational resources. Therefore, this technology is expected to be widely used in various next-generation artificial intelligence fields, such as small devices like smartphones, as well as humanoid robots that need to accurately identify and manipulate small objects, autonomous driving systems, and on-device AI.
<Comparison image illustrating the performance gap with conventional methods (AI-generated). Conventional vision foundation models understand a scene by converting the input image into low-resolution features at a small patch level (left). Upsample Anything restores these low-resolution features to the original resolution level, enabling the AI to comprehend the scene's structure and boundaries with significantly higher precision (right)>
Professor Changick Kim said, “This technology is an algorithm that can significantly increase the visual precision of artificial intelligence with fewer resources, and it is expected to accelerate the commercialization of humanoid robots and on-device AI.” He added, “It is even more meaningful because it was recognized at CVPR not only for its performance but also for its computational efficiency and research transparency.”
This research was participated in by KAIST PhD student Minseok Seo as the first author, and this achievement was presented on June 7 at 'CVPR 2026,' the world's most prestigious conference in the field of artificial intelligence and computer vision.
※ Paper Title: Upsample Anything: A Simple and Hard to Beat Baseline for Feature Upsampling, DOI:10.48550/arXiv.2511.16301
※ Author Information: Minseok Seo (KAIST, First Author), Mark Hamilton (MIT, Microsoft, Second Author), Changick Kim (KAIST, Corresponding Author)
2026.06.17 View 1746 -
"Age of Robots Making Human-Like Judgments, KAIST Solves Key Challenge in Physical AI
< (From left) Professor Chang D. Yoo, Tung M. Luu (PhD candidate, first author) at the back center, and Hwanhee Kim (M.S candidate, second author) at the front right >
“Robots that make judgments like humans are coming faster than we think.” A core technology that will accelerate the era where robots understand human intentions and choose the correct actions on their own has been developed in South Korea. KAIST researchers solved a key challenge in the commercialization of physical AI by developing a technology where AI learns human judgment criteria on its own with just a few videos. KAIST announced on June 10th that a research team led by Professor Chang D. Yoo from the School of Electrical Engineering has developed 'VOTP (Video-based Optimal TransPort Preference)' for the first time in the world, a new technology that allows AI to learn human intentions and judgment criteria using just a few preference videos instead of thousands to tens of thousands of human evaluation data points.
< VOTP Overview Diagram >
The research team's paper has been accepted to ICML (International Conference on Machine Learning) 2026, the world's most prestigious AI conference, which will be held at COEX in Seoul this July. It was selected for an Oral presentation, an honor given to only the top 0.7% (168 papers) out of all submitted papers (23,918 papers), recognizing the excellence of the research. ICML is considered one of the most influential international conferences in the fields of AI and machine learning. Recently, AI technology is rapidly evolving beyond generative AI that writes text and draws pictures into the era of 'Physical AI,' which moves actual machines and acts in the real world. Representative examples include robots that perform dangerous tasks in factories instead of humans, autonomous vehicles that judge road situations on their own, and medical robots that perform delicate surgeries. However, there was a barrier that had to be overcome for the practical application of physical AI. It is the problem of learning human-level evaluation criteria to judge whether the actions performed by a machine match human intentions and which actions are more desirable. For example, when a surgical robot performs suturing or an autonomous vehicle passes through a complex intersection, the AI must choose the most appropriate action among numerous options. To achieve this, a 'Reward Function' that reflects human preferences and judgment criteria is required. However, until now, humans had to directly evaluate thousands to tens of thousands of action data points to build this, which required an enormous amount of time and cost. The research team focused on the way humans learn new tasks after seeing just a few demonstrations. VOTP, developed by the research team, helps AI understand human-preferred action patterns on its own with just a few videos of good and bad examples. Even without humans evaluating a vast amount of data one by one as before, AI can understand human judgment criteria and expand its learning to various situations. The core idea of this research is that intelligent machines such as robots or autonomous vehicles can quickly grasp human intents with only a small number of videos containing human preferences. The algorithm developed for this purpose proved its effectiveness and generalization performance through extensive experiments across various environments and tasks. This method can significantly reduce human feedback and data construction costs required for physical AI development. Since robots, autonomous vehicles, and industrial machinery can learn actions that meet human expectations with only a small number of examples, it is expected to drastically shorten development time and costs. The technology can be widely applied not only to robot arm control, humanoid robots, autonomous vehicles, smart factories, drones, and surgical robots, but also to AI agents that directly operate computers. In particular, it is expected to be utilized as a core foundational technology for all physical AI systems that need to learn human intention and satisfaction.
< VOTP Research Image (AI Generated) >
Professor Chang D. Yoo said, "The core of physical AI is making machines understand human intentions and choose the correct actions," and added, "Since VOTP can learn human judgment criteria with only a small number of videos, it is a core technology that will accelerate the era of robots making human-like judgments." This research, in which PhD student Tung M. Luu from the School of Electrical Engineering participated as the first author, was selected as an Oral presentation paper at ICML (International Conference on Machine Learning) 2026, the world's most prestigious AI conference. ※ Paper Title: Video-Based Optimal Transport for Feedback-Efficient Offline Preference-Based Reinforcement Learning, Paper File: https://sanctusfactory.com/data/file/publications/202606091714078906.pdf This research was conducted with support from the Institute for Information & Communication Technology Planning & Evaluation (IITP) and the National Research Foundation of Korea (NRF), funded by the Ministry of Science and ICT."
2026.06.10 View 1602 -
Humanoid Robot Pilot PIBOT Wins Best Paper Award at the World’s Most Prestigious Robotics Magazine
< (From left of the award recipients) Ph. D candidate Sungjae Min, Ph. D candidate Gyuree Kang, Professor David Hyunchul Shim, Ph.D candidate Hyungjoo Kim >
KAIST announced on June 5th that a paper proposing an aircraft autonomous piloting framework based on the humanoid robot pilot ‘PIBOT,’ developed by a research team led by Professor David Hyunchul Shim of the School of Electrical Engineering, was selected as the Best Paper Award among the papers published in the IEEE Robotics & Automation Magazine (IEEE RAM) in 2025.
< The proposed PIBOT system framework capable of piloting based on aviation manuals and voice communication without modifying existing aircraft >
This award is highly meaningful as it signifies that grassroots research based entirely on domestic, independent initiatives has been recognized as a world-class achievement in robotics. The award ceremony took place in Vienna, Austria, on June 4, 2026 (local time) during the International Conference on Robotics and Automation (ICRA 2026). IEEE Robotics & Automation Magazine (IEEE RAM) is a prestigious academic magazine published by the IEEE Robotics and Automation Society (RAS), under the umbrella of IEEE, the world's largest technical professional organization. It is well known for delivering the latest research achievements, industry trends, and tutorials in the fields of robotics and automation, widely conveying robot technologies applicable to actual industrial sites to researchers in both industry and academia. As of 2025, IEEE RAM recorded an Impact Factor (IF) of 7.1, holding the second highest impact among IEEE publications in the field of robotics. In particular, it presents the Best Paper Award to research that has a significant academic and industrial impact among the papers published after undergoing rigorous peer review. This study was selected as a Future Challenge Defense Technology Research and Development Project by the Agency for Defense Development (ADD) in 2021 and was conducted based purely on domestic technology with support of approximately 5.7 billion won over five years. The research team received high praise for implementing Physical AI technology at an exceptionally high level, enabling a humanoid robot to systematically and adaptively perform complex tasks such as piloting aircraft based on artificial intelligence, going beyond simple walking or carrying items. Recently, humanoid robot technology has been developing rapidly in terms of athletic performance, such as tumbling or implementing complex movements. However, in the industrial sector, the applicability to actual industrial sites is drawing attention as a more critical factor. The pilot robot ‘PIBOT’ being developed by Professor David Hyunchul Shim's research team is designed to acquire specialized knowledge required for aircraft operation and to recognize and respond to actual flight situations in real time, going beyond simple repetitive tasks or logistics processing. Accordingly, it is evaluated as presenting a new direction for the utilization of humanoid robot technology, termed as Expert Physical AI.
< The research team's PIBOT sitting in an actual aircraft (KLA-100) and operating the instruments and control stick >
The research team has successfully completed Phase 1 of the research since the project launched in 2021, and since 2024, they have been developing Phase 2 of the pilot robot, which features a human-like physique and joint structure suitable for actual aircraft piloting. In addition, they are pursuing collaborative research with relevant organizations to expand and apply this technology to various mobile vehicle piloting fields, such as ground vehicles and ships, as well as aircraft.
< PIBOT performing piloting in an aircraft simulator device >
Professor David Hyunchul Shim said, “It is very meaningful that the pilot robot technology, proposed for the first time in the world by Korean researchers, has been recognized as a world-class research achievement thanks to the support of a large-scale national project. We will further develop our research in a direction where humanoid robots can help humans in real-world environments and safely operate complex systems.” In this study, PhD students Sungjae Min, Gyuree Kang, and Hyungjoo Kim participated as co-first authors, and Professor David Hyunchul Shim served as the corresponding author. The paper can be found through IEEE Xplore. ※ Paper Title: “Toward Fully Autonomous Aviation: PIBOT, a Humanoid Robot Pilot for Human-Centric Aircraft Cockpits”, Paper Links: https://doi.org/10.1109/MRA.2024.3505774, https://ieeexplore.ieee.org/document/10798973/ Meanwhile, this research was conducted with support from the Agency for Defense Development's Future Challenge Defense Technology Research and Development Project.
2026.06.05 View 1137 -
Robot Valley Project Activation of the Korean style Robot and AI Startup Ecosystem Fully Underway
< From left: Top Excellence Award winner Robolight (Pre-startup Founder Han-seol Choi), Top Excellence Award winner Coils (CEO Seong-ryeol Heo), Professor Jung Kim of KAIST, Grand Prize winner Noman (CEO Jung-wook Moon), Professor Kyoungchul Kong of KAIST, CEO Dae-hee Park of Daejeon Creative Economy Innovation Center, Excellence Award winner Gigaflops (CEO Min-tae Kim), Excellence Award winner BLUE APEX (Pre-startup Founder Na-hyeon Kwon) >
KAIST announced on December 10th that KAIST Holdings (CEO Hyeonmin Bae), a specialized technology commercialization investment institution, successfully held the '2025 KAIST Hu-Robotics Startup Cup' on the 9th at the main building of Daejeon Startup Park. This was held as part of the Robot Valley Project, aiming to discover and foster promising startup teams in the robotics field and establish a robot scale-up ecosystem based on a technology platform.
This competition was conducted as a core program of the Robot Valley Project (Deep-Tech Scale-up Valley Fostering Project), which is promoted by the Ministry of Science and ICT and supported by Daejeon Metropolitan City. The competition proceeded through a meet-up day with KAIST Mechanical Engineering researchers, robotics companies like Angel Robotics and Twinny, and startup experts such as Bluepoint, leading to the final round. Throughout this process, a support system for the scale-up of robot startups was established, linking technology verification, strengthening entrepreneurial capabilities, and investment linkage.
KAIST Holdings and the Deep-Tech Valley Project Group (hereinafter referred to as the Project Group) stated that this competition marks the beginning of 'establishing a Korean-style Robot and AI startup ecosystem.' Their goal through the Robot Valley Project is to create a Korean-style robot scale-up ecosystem centered around Daejeon and KAIST, and furthermore, to build a technology circulation structure utilizing verified technology platforms.
KAIST has produced successful scale-up cases in the robotics field, such as Rainbow Robotics and Angel Robotics. However, the recent robotics industry has seen a rapid increase in technological difficulty due to the convergence of mechanical engineering, AI, and control software, creating structural limitations for early-stage founders to challenge alone.
To solve this, the Project Group proposed the 'Scale-up Valley Construction Strategy,' which opens up the verified technologies of established senior companies to junior founders. This strategy focuses on supporting startups to concentrate on developing market-ready robot services and applications on top of verified technology platforms, rather than consuming excessive time on developing basic hardware like motors and controllers.
The Angel Robotics technology platform, presented as the core underlying technology of this strategy, consists of actuators, control modules, and core software. KAIST plans to gradually open up these foundational technologies for use by early-stage startup teams.
The Project Group emphasized that enabling startup teams to utilize such technology platforms from the initial stage is the core infrastructure for accelerating the Korean-style robot startup ecosystem.
A total of 21 teams participated in this competition, including pre-startup founders (Track A) and early-stage startups established within 3 years (Track B), all possessing human-centered robotics technology and convergence business models.
After fierce preliminaries, 8 teams advanced to the final round, and a total of 5 teams were finally selected: one Grand Prize winner, two Choi Woo-sung (Top Excellence Award) winners, and two Excellence Award winners.
The Grand Prize was awarded to 'Noman' for proposing an integrated system for a strawberry farm work robot and a rotating vertical cultivation module.
The Woo-sung Choi (Top Excellence Award) went to 'Robolight' and 'Coils.'
The Excellence Award was awarded to BLUE APEX and Gigaflops.
Professor Jung Kim, Head of the KAIST Mechanical Engineering Department and General Manager of the Robot Valley Project, said, "This competition has become the starting point for discovering future robot unicorns. For the next three years, we will continue to provide practical support for the growth of robot startups, and KAIST will play a leading role in building and expanding the deep-tech robot ecosystem centered in Daejeon."
< Group Photo of Award Winners >
Meanwhile, this competition was jointly hosted and organized by the Ministry of Science and ICT, Daejeon Metropolitan City, and the Research and Business Development Special Zone Foundation, as well as startup support organizations including KAIST, KAIST Holdings, Daejeon Technopark, and Daejeon Creative Economy Innovation Center.
2025.12.10 View 3776 -
KAIST Develops Multimodal AI That Understands Text and Images Like Humans
<(From Left) M.S candidate Soyoung Choi, Ph.D candidate Seong-Hyeon Hwang, Professor Steven Euijong Whang>
Just as human eyes tend to focus on pictures before reading accompanying text, multimodal artificial intelligence (AI)—which processes multiple types of sensory data at once—also tends to depend more heavily on certain types of data. KAIST researchers have now developed a new multimodal AI training technology that enables models to recognize both text and images evenly, enabling far more accurate predictions.
KAIST (President Kwang Hyung Lee) announced on the 14th that a research team led by Professor Steven Euijong Whang from the School of Electrical Engineering has developed a novel data augmentation method that enables multimodal AI systems—those that must process multiple data types simultaneously—to make balanced use of all input data.
Multimodal AI combines various forms of information, such as text and video, to make judgments. However, AI models often show a tendency to rely excessively on one particular type of data, resulting in degraded prediction performance.
To solve this problem, the research team deliberately trained AI models using mismatched or incongruent data pairs. By doing so, the model learned to rely on all modalities—text, images, and even audio—in a balanced way, regardless of context.
The team further improved performance stability by incorporating a training strategy that compensates for low-quality data while emphasizing more challenging examples. The method is not tied to any specific model architecture and can be easily applied to various data types, making it highly scalable and practical.
<Model Prediction Changes with a Data-Centric Multimodal AI Training Framework>
Professor Steven Euijong Whang explained, “Improving AI performance is not just about changing model architectures or algorithms—it’s much more important how we design and use the data for training.” He continued, “This research demonstrates that designing and refining the data itself can be an effective approach to help multimodal AI utilize information more evenly, without becoming biased toward a specific modality such as images or text.”
The study was co-led by doctoral student Seong-Hyeon Hwang and master’s student Soyoung Choi, with Professor Steven Euijong Whang serving as the corresponding author. The results will be presented at NeurIPS 2025 (Conference on Neural Information Processing Systems), the world’s premier conference in the field of AI, which will be held this December in San Diego, USA, and Mexico City, Mexico.
※ Paper title: “MIDAS: Misalignment-based Data Augmentation Strategy for Imbalanced Multimodal Learning,” Original paper: https://arxiv.org/pdf/2509.25831
The research was supported by the Institute for Information & Communications Technology Planning & Evaluation (IITP) under the projects “Robust, Fair, and Scalable Data-Centric Continual Learning” (RS-2022-II220157) and “AI Technology for Non-Invasive Near-Infrared-Based Diagnosis and Treatment of Brain Disorders” (RS-2024-00444862).
2025.10.14 View 5068 -
Next-Generation Humanoid Robot Capable of Moonwalk Developed
<From the middle of the back row, clockwise: Professor Hae-Won Park, Dongyun Kang (Ph.D. candidate), Hajun Kim (Ph.D. candidate), JongHun Choe (Ph.D. candidate), Min-su Kim (Research Professor)>
KAIST research team's independently developed humanoid robot boasts world-class driving performance, reaching speeds of 12km/h, along with excellent stability, maintaining balance even with its eyes closed or on rough terrain. Furthermore, it can perform complex human-specific movements such as duck walk and moonwalk, drawing attention as a next-generation robot platform that can be utilized in actual industrial settings. Professor Park Hae-won's research team at the Humanoid Robot Research Center (HuboLab) of KAIST's Department of Mechanical Engineering announced on the 19th that they have independently developed the lower body platform for a next-generation humanoid robot. The developed humanoid is characterized by its design tailored for human-centric environments, targeting a height (165cm) and weight (75kg) similar to that of a human. The significance of the newly developed lower body platform is immense as the research team directly designed and manufactured all core components, including motors, reducers, and motor drivers. By securing key components that determine the performance of humanoid robots with their own technology, they have achieved technological independence in terms of hardware. In addition, the research team trained an AI controller through a self-developed reinforcement learning algorithm in a virtual environment, successfully applied it to real-world environments by overcoming the Sim-to-Real Gap, thereby securing technological independence in terms of algorithms as well.
<Developed 'KAIST Humanoid' Lower Body Platform>
Currently, the developed humanoid can run at a maximum speed of 3.25m/s (approximately 12km/h) on flat ground and has a step-climbing capability of over 30cm (a performance indicator showing how high a curb, stairs, or obstacle can be overcome). The team plans to further enhance its performance, aiming for a driving speed of 4.0m/s (approximately 14km/h), ladder climbing, and over 40cm step-climbing capability.
<‘KAIST Humanoid’ Lower Body Platform running>
Professor Hae-Won Park's team is collaborating with Professor Jae-min Hwangbo's team (arms) from KAIST's Department of Mechanical Engineering, Professor Sangbae Kim's team (hands) from MIT, Professor Hyun Myung's team (localization and navigation) from KAIST's Department of Electrical Engineering, and Professor Jae-hwan Lim's team (vision-based manipulation intelligence) from KAIST's Kim Jaechul AI Graduate School to implement a complete humanoid hardware with an upper body and AI. Through this, they are developing technology to enable the robot to perform complex tasks such as carrying heavy objects, operating valves, cranks, and door handles, and simultaneously walking and manipulating when pushing carts or climbing ladders. The ultimate goal is to secure versatile physical abilities to respond to the complex demands of actual industrial sites.
<An Intermediate Result: A Single-Leg Hopping Robot Has Been Developed>
During this process, the research team also developed a single-leg 'Hopping' robot. This robot demonstrated high-level movements, maintaining balance on one leg and repeatedly hopping, and even exhibited extreme athletic abilities such as a 360-degree somersault. Especially in a situation where imitation learning was impossible due to the absence of a biological reference model, the research team achieved significant results by implementing an AI controller through reinforcement learning that optimizes the center of mass velocity while reducing landing impact. Professor Park Hae-won stated, "This achievement is an important milestone that has achieved independence in both hardware and software aspects of humanoid research by securing core components and AI controllers with our own technology," and added, "We will further develop it into a complete humanoid including an upper body to solve the complex demands of actual industrial sites and furthermore, foster it as a next-generation robot that can work alongside humans."
<Key Components of the Directly Developed Robot: (a) Reducer, (b) Motor Stator, (c) Motor Driver, (d) EtherCAT-CAN convert board>
The results of this research will be presented by JongHun Choe, a Ph.D. candidate in Mechanical Engineering, as the first author, on hardware development at 'Humanoids 2025', an international humanoid robot specialized conference held on October 1st. Additionally, Ph.D. candidates Dongyun Kang, Gijeong Kim, and JongHun Choe from Mechanical Engineering will present the AI algorithm achievements as co-first authors at 'CoRL 2025', the top conference in robot intelligence, held on September 29th. ※Paper Titles and Papers: Learning Impact-Rich Rotational Maneuvers via Centroidal Velocity Rewards and Sim-to-Real Techniques: A One-Leg Hopper Flip Case Study, Conference on Robot Learning (CoRL), Seoul, Korea 2025, Dongyun Kang, Gijeong Kim, JongHun Choe, Hajun Kim, Hae-Won Park, arxiv version: https://arxiv.org/abs/2505.12222 Design of a 3-DOF Hopping Robot with an Optimized Gearbox: An Intermediate Platform Toward Bipedal Robots, IEEE-RAS, International Conference on Humanoid Robots, Seoul, Korea, 2025, JongHun Choe, Gijeong Kim, Hajun Kim, Dongyun Kang, Min-Su Kim, Hae-Won Park, arxiv version: https://arxiv.org/abs/2505.12231 This research was supported by research funding from the Ministry of Trade, Industry and Energy and the Korea Institute of Industrial Technology Planning and Evaluation (KEIT) (RS-2024-00427719). ※ Related Video: https://youtu.be/ytWO7lldN4c
2025.09.19 View 5997 -
A Boom in Robot Startups: Global Ventures from the Legacy of HUBO's Creator
KAIST announced on September 16 that it is gaining attention as a "cradle of Korean robotics" as various robot startups founded on campus have recently succeeded in attracting investment.
Rainbow Robotics, founded by Professor Jun-Ho Oh of the Department of Mechanical Engineering, set a new milestone in the robotics industry by successfully going public with its world-class humanoid technology. Following this, Angel Robotics, a company specializing in rehabilitation and medical robots founded by Professor Kyung-chul Kong of the Department of Mechanical Engineering, also went public, making the achievements of KAIST-born robot startups more visible.
Following in their footsteps, a number of other startups are on a rapid growth trajectory after their founding in various technological fields, including quadrupedal, collaborative, and wearable robots, as well as autonomous walking. These include Pureun Robotics (2021, Hyunchul Ham, MS from Mechanical Engineering), Wero Robotics (2021, Yeonbaek Lee, MS from Mechanical Engineering), Raion Robotics (2023, Professor Jaemin Hwangbo, Mechanical Engineering), Triangle Robotics (2023, Jinhyuk Choi, PhD candidate in Computer Science), URobotics (2024, Byungho Yoo, PhD from Electrical Engineering), and Diden Robotics (2024, Junha Kim, PhD from Mechanical Engineering).
In particular, Raion Robotics, founded by Professor Jaemin Hwangbo of the Department of Mechanical Engineering, recently secured a Series A investment of 23 billion KRW from leading domestic investors, including SBVA, Company K Partners, FuturePlay, KDB Capital, IBK, and IBK Venture Capital.
< (Left) Raibo1, (Right) Raibo2 participating in a marathon >
Raion Robotics' flagship product, the quadrupedal robot 'Raibo,' is equipped with reinforcement learning-based AI, enabling stable walking on uneven terrain. It also boasts a distinctive performance with an 8-hour operating time. Recently, it successfully completed a full marathon (42.195 km) alongside a human, proving its durability in real-world conditions and attracting attention from the global robotics industry.
This trend is also evident in URobotics, a startup from Professor Hyun Myung's lab in the Department of Electrical Engineering. URobotics recently secured a 3.5 billion KRW seed investment and was selected for the 1.5 billion KRW Deep Tech TIPS program, accelerating its growth in the field of autonomous walking robots. The company is preparing to apply its technology to various industrial sites, including defense, construction, logistics, and smart cities, by internalizing its control and autonomous walking technologies and applying them to humanoids. The industry is already taking note of its high growth potential from the early stages.
< (Left) URobotics' general-purpose autonomous walking solution being tested on a quadrupedal robot, (Right) Developing core spatial intelligence technology >
< URobotics' autonomous walking solution >
Diden Robotics, a startup from Professor Haewon Park's lab in the Department of Mechanical Engineering, is leading the industrial application and commercialization of walking mobile robot technology. The company's key competitive advantages lie in its hardware design capabilities through the internalization of core components, advanced Physical AI technology based on reinforcement learning, and a special magnetic foot technology. Robots developed with this technology can move freely on vertical steel walls and ceilings to perform high-difficulty tasks like welding and non-destructive testing. Based on this technology, Diden Robotics attracted a 7 billion KRW investment in a Pre-A round and has signed supply contracts with major shipyards, proving its commercial viability.
< (Left) Diden Robotics' mobile robot DIDEN30 for shipbuilding sites (Right) Various work scenarios inside a ship block >
KAIST recently secured 10.5 billion KRW in government funding by participating as the lead institution in the Deep Tech Scale-up Valley project. With this funding, it plans to create a virtuous cycle among companies, technology, and talent in the robotics industry and emerge as a next-generation robotics hub. URobotics and Angel Robotics are also participating in this project.
Bae Hyun-min, head of the Startup Center, said, "Researchers from KAIST are entering the global stage through challenging startups. The Startup Center will actively support them to help KAIST establish itself as a 'hub for deep tech startups'."
KAIST President Kwang Hyung Lee emphasized, "KAIST is a cradle of innovation that creates social value through startups, beyond education and research. The achievements of these robot startups show that KAIST is at the center of leading the paradigm of the global robotics industry. This also aligns with KAIST's vision of preparing for the era of 'Physical AI,' which fuses artificial intelligence with the physical world. KAIST will continue to strengthen its global technological leadership through innovation that connects academia and industry.
2025.09.16 View 9072 -
KAIST to Foster a 'Robot Valley' in Daejeon with $10 Million Initiative
<Group Photo of Kick-off Meeting>
On September 3, KAIST announced the official launch of the "2025 Deep Tech Scale-up Valley Nurturing Project" with a kick-off meeting at the KAIST Department of Mechanical Engineering.
KAIST was selected for this project by the Ministry of Science and ICT and the Research and Development Special District Foundation. With this selection, the university plans to create a "Robot Valley".
Over the next three and a half years, KAIST will receive a total of 13.65 billion won (approximately $10 million) in funding. The university's goal is to intensively nurture globally competitive, innovative robotics companies based on foundational technologies and to develop Daejeon into a global hub for the robotics industry.
The initiative will leverage Daejeon's exceptional research talent and its startup and investment ecosystem to create a model for regional revitalization and to cultivate the robotics industry as a next-generation strategic sector.
KAIST's vision for this project is to develop "Human-Friendly Robots (HFR)" that are more than just automated machines; they are collaborative partners that share space, roles, and emotions with people.
The project will implement a multi-stage strategy that includes promoting the commercialization of robotics technology, supporting the startup ecosystem, securing global technological competitiveness, and developing robot commercialization platforms. This will establish a virtuous cycle of technology development, startup and investment growth, and reinvestment.
Unlike traditional startup support and scale-up programs, this project aims for the simultaneous growth of the entire robotics industry, not just individual companies. A key element is an open innovation model where leading robotics firms like Angel Robotics Inc. and EuRoBotics Inc. (led by Professor Byung-ho Yu and Professor Hyun Myung) will share common core technologies related to actuators, circuits, AI, and standardized data. This will allow startups to focus on developing robot products that directly meet customer needs.
The project team includes key KAIST robotics researchers. The project leader is Professor Jung Kim (President of the Korea Robotics Society) from the Department of Mechanical Engineering. Other participating professors include Geon-Jae Lee from the Department of Materials Science and Engineering (human augmentation sensors), Hyun Myung from the School of Electrical Engineering (winner of the QRC 2023 quadruped robot autonomous walking competition at IEEE ICRA), Kyung-Chul Kong from the Department of Mechanical Engineering (two-time champion of the Cybathlon International Competition and founder of Angel Robotics), and Suk-Hyung Bae from the Department of Industrial Design (winner of the ACM SIGGRAPH robot sketching competition).
In addition, the KAIST Technology Commercialization Office, KAIST Holdings, Global Techno Valley Lab (GTLAB), and the Daejeon Center for Creative Economy and Innovation will manage technology commercialization and valley construction. The Daejeon Technopark will also participate to provide comprehensive commercialization support.
"The strategic cooperation between Daejeon City's robotics industry nurturing plan and KAIST was the driving force behind the selection for this project," said Geon-Jae Lee, Director of the KAIST Technology Commercialization Office. "We will create a robotics innovation ecosystem based in Daejeon and systematically foster global companies to rival the likes of ABB in Switzerland and KUKA in Germany, which are considered among the top three robotics companies in the world."
< Kick-off Meeting Scene>
Project leader Jung Kim stated, "We will spearhead efforts to discover and nurture over 15 future unicorn companies by promoting the commercialization of deep-tech robotics developed at KAIST. The entire KAIST robotics research team will dedicate its full efforts to ensure that our research and development achievements lead to real-world industries and startups."
KAIST President Kwang-Hyung Lee emphasized, "As Korea's leading research-oriented university, KAIST will actively support Daejeon's growth into a global robotics hub. This project is more than just research and development; it will be a turning point for KAIST to stand at the center of the global robotics ecosystem and create a new growth engine for the region and the nation."
In collaboration with Daejeon City, KAIST plans to form an "HFR Valley Innovation Council" to share and review project outcomes, ultimately building a self-sustaining ecosystem. This initiative aims to establish Daejeon as a world-class robotics industry hub.
2025.09.03 View 5855