Open X-Embodiment: Robotic Learning Datasets and RT-X Models
The project aggregates robot demonstration data across institutions and studies models trained across different robot embodiments.
Machines, companies, technologies, industries, applications, components, people, research, rules, glossary terms and attributed news, ranked in one list.
132 results for “robot learning”
Also matched: machine, autonomyThe project aggregates robot demonstration data across institutions and studies models trained across different robot embodiments.
The work studies closed-loop grasping learned from a large collection of robot interaction data.
A three-fingered, 9-DOF research hand with current-controlled joints and omnidirectional tactile sensors, controlled over CAN.
A low-cost ROS mobile manipulator built on an iRobot Create 3 (earlier Kobuki) base with an Intel NUC, RealSense camera and optional 4–6 DOF arm.
The paper describes vision-language-action models that represent robot actions alongside visual and language inputs.
Configurable bipedal humanoid platform from PAL Robotics for locomotion and reinforcement-learning research, offered with 4-, 5- or 7-DoF arms and ROS 2 software.
arXiv:2611.23575v1 Announce Type: new Abstract: Low-cost tactile arrays have made touch a practical input for robot learning, but piezoresistive designs sense only…
Experts from Eli Lilly and Purdue University will reveal findings about the perception of robots on the manufacturing floor. The post Eli Lilly, Purdue to share field…
The live webinar, "Robots in the Supply Chain," will take place Wednesday, November 18, 2026, at 12 P.M. ET. The post Learn how AI is changing supply chain robotics…
Pieter Abbeel is a UC Berkeley professor working at the intersection of robotics and machine learning.
Franka Research 3 is a 7-DoF torque-sensing robot arm aimed at research labs, offering 1 kHz low-level control and integration with ROS 2 and MATLAB.
Third-generation bipedal humanoid from Mentee Robotics, an Israeli startup now owned by Mobileye, shown in February 2025 with camera-only perception and a hot-swappable…
Cloud-connected companion robot for children aged about 5–10 with an animated face, used for conversation and social-emotional games. Embodied shut down in December…
arXiv:2610.36681v1 Announce Type: new Abstract: Animals exhibit diverse shape-changing behaviors, from the undulatory swimming of eels to the arching movements of…
Machine learning develops models whose behavior is fitted from data rather than specified only through fixed rules.
Dieter Fox studies robotics, perception and state estimation; the institutional page is the authority for current roles.
Unitree's compact bipedal humanoid, about 1.32 m tall and 35 kg, sold in a base version and an EDU version that adds dexterous hands, extra waist/wrist joints and…
A child-sized bipedal humanoid from Galbot aimed at entertainment and storefront interaction, driven by a motion foundation model. Pre-orders opened in September 2026.
Leju's full-size bipedal humanoid offered as a research and service platform, available in 22-, 30- and 40-DOF configurations. Leju's spec sheet lists hot-swappable…
A 160 cm full-size bipedal humanoid from LimX Dynamics introduced in May 2026 for customer-facing and entertainment roles in malls, theme parks and showrooms.
A soft upper-body humanoid from Toyota Research Institute for research on carrying large objects with its arms and chest as well as its hands. Soft tactile covers go…
A lower-body bipedal platform from EngineAI for research and education, released in July 2024 with open-source software access. It has 12 degrees of freedom (six per…
The work applies diffusion models to generate robot action sequences from observations.
Sony’s robot dog companion with OLED eyes, cameras, touch sensors and cloud-connected learning; the ERS-1000 is the current generation of the aibo line.
A ROS 2 learning and research robot built on the iRobot Create 3 base, with a Raspberry Pi 4, 2D lidar and OAK-D camera; a Lite version omits the top deck and uses a…
arXiv:2611.06196v1 Announce Type: new Abstract: Ensuring safety in advanced robot control methods such as Model Predictive Control (MPC) and Reinforcement Learning…
arXiv:2610.32093v1 Announce Type: new Abstract: Behavior cloning for robotic manipulation is highly sensitive to the coverage of its training data, yet collecting…
arXiv:2611.13307v1 Announce Type: new Abstract: Purpose: Robot-based haptic dyads, where two users exchange forces through robots, have been proposed for robot-assisted…
arXiv:2611.13685v1 Announce Type: new Abstract: In this paper, we investigate how human egocentric videos can benefit robot policy learning without action supervision…
arXiv:2611.19196v1 Announce Type: new Abstract: Imitation learning for visuomotor policies typically relies on teleoperated robot demonstrations, which are costly…
arXiv:2610.41348v1 Announce Type: new Abstract: Robots deployed in unstructured environments are susceptible to actuator-authority degradation, which reduces the…
arXiv:2611.19616v1 Announce Type: new Abstract: Reinforcement learning (RL) enables robots to acquire complex skills, but standard RL formulations do not guarantee that…
The new approach, known as FLEX, could lead to more intuitive, interactive household and assistive robots.
arXiv:2610.41320v1 Announce Type: new Abstract: Recent multimodal foundation models exhibit strong perception, reasoning, and in-context learning capabilities, but…
arXiv:2611.23563v1 Announce Type: new Abstract: Many learning-based approaches for deformable linear objects (DLOs) rely on simplified deformation models that neglect…
arXiv:2611.19324v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have demonstrated strong capabilities in multi-task robotic manipulation. However…
arXiv:2610.32225v1 Announce Type: new Abstract: Recent robot manipulation policies commonly predict continuous action chunks directly from observations, without…
Sanctuary AI's general-purpose humanoid with hydraulic, high-dexterity hands, first unveiled in 2023 and iterated through several generations. In 2026 the company…
A dual-arm humanoid upper body on an omnidirectional wheeled base, revealed by Shenzhen's Astribot (Stardust Intelligence) in April 2024. It is designed for fast…
Bipedal variant of Humanoid's HMND 01 Alpha, positioned as an R&D platform toward home and service applications and controlled by a reinforcement-learning whole-body…
A small 3D-printed, servo-driven open-source humanoid for sim-to-real and reinforcement-learning experiments, with a published bill of materials starting around $350…
A set of methods that fit model behavior from data for tasks such as prediction, classification, perception or control.
Eka Robotics is building a vision-force-action, or VFA, model to develop an intuition of physics through reinforcement learning at scale. The post Eka Robotics raises…
arXiv:2611.05667v1 Announce Type: new Abstract: This paper presents Proactive Human-in-the-Loop (PHIL), a framework for proactive policy correction and preference-based…
arXiv:2610.36270v1 Announce Type: new Abstract: Complex whole-body loco-manipulation requires executing diverse maneuvers, such as crouching, grasping, and locomotion…
arXiv:2611.05979v1 Announce Type: new Abstract: Manipulating diverse objects with multi-fingered dexterous hands is challenging due to the high dimensionality and…
Autonomous tractor using cameras and machine learning to perform bounded tillage work under remote monitoring.
Camera- and machine-learning-equipped spraying system that identifies crops and weeds for targeted application.
arXiv:2611.09030v1 Announce Type: new Abstract: Agentic robotics asks coding agents to implement and iteratively improve robot policies from simulated and real-world…
arXiv:2610.33227v1 Announce Type: new Abstract: The last decades have seen a revolution in autonomous robotics. Deep learning approaches and their hardware…
arXiv:2610.31845v1 Announce Type: new Abstract: Scaling policy learning requires broad experience and reliable supervision. Robot demonstrations provide high-quality…
arXiv:2611.05571v1 Announce Type: new Abstract: Robot imitation learning is limited by the scarcity of action-labeled data, while action-free demonstrations such as…
arXiv:2610.40902v1 Announce Type: new Abstract: Flapping-wing simulation must reconcile high-frequency wingbeat aerodynamics with the slower dynamics relevant to flight…
arXiv:2610.41336v1 Announce Type: new Abstract: Learning dexterous manipulation from human videos remains an open challenge, particularly in the context of transferring…
arXiv:2610.36374v1 Announce Type: new Abstract: Robotic agents operating in real-world environments must adapt continuously from a stream of multimodal demonstrations…
arXiv:2611.17658v1 Announce Type: new Abstract: Imitation policies for robot manipulation are often trained on noisy object positions, and they learn to discount the…
arXiv:2610.31801v1 Announce Type: new Abstract: Segmenting robot demonstration data into disjoint temporal and semantic subsegments is valuable for learning, cleaning…
arXiv:2611.23502v1 Announce Type: new Abstract: We introduce TAG (Testbed for Autonomous Games), a compact, low-cost robotic platform for studying learning and control…
arXiv:2611.23628v1 Announce Type: new Abstract: Noise steering improves pretrained generative robot policies by learning to select their initial noise without updating…
arXiv:2611.17631v1 Announce Type: new Abstract: Visuomotor imitation learning has become a predominant paradigm for real-world robotic manipulation, yet it remains…
arXiv:2611.23567v1 Announce Type: new Abstract: Symbolic methods make robot designs interpretable but often overlook continuous geometry; learned methods handle geometry…
arXiv:2610.31812v1 Announce Type: new Abstract: Preference-based reinforcement learning (PbRL) has emerged as a promising paradigm for learning robot behaviors aligned…
arXiv:2611.08576v1 Announce Type: new Abstract: As robots are deployed in the real world, they will inevitably make mistakes. Human interventions provide valuable…
arXiv:2611.05838v1 Announce Type: new Abstract: Assessing the deployment performance of learned robot policies during training is challenging, as directly measuring task…
arXiv:2611.23731v1 Announce Type: new Abstract: Diffusion and flow-matching policies have become a standard choice for visuomotor imitation learning, owing to their…
arXiv:2610.36685v1 Announce Type: new Abstract: Bimanual robot manipulation plays an important role in real-world robotic applications, but the high cost of collecting…
arXiv:2610.31829v1 Announce Type: new Abstract: Real-world reinforcement learning has enabled contact-rich manipulation tasks to be trained directly on physical robots…
arXiv:2611.08857v1 Announce Type: new Abstract: Learning locomotion directly on a humanoid robot is a promising direction to avoid the sim-to-real gap. However…
arXiv:2605.05236v3 Announce Type: replace Abstract: In the field of precision manufacturing in complex constrained environments, the role of soft robots is increasingly…
arXiv:2611.08698v1 Announce Type: new Abstract: Recent advances in robot foundation models have highlighted the potential of in-context learning (ICL) for adapting robot…
arXiv:2610.31834v1 Announce Type: new Abstract: Existing approaches for robot learning in simulation often require substantial manual effort to construct scenes and…
Fei-Fei Li is a Stanford professor whose work includes computer vision and human-centered artificial intelligence.
Embodied AI company (星尘智能, Stardust Intelligence) established in Shenzhen in late 2022. It builds the S1 wheeled dual-arm humanoid and its own robot learning models.
ETH Zurich spin-out, renamed from Swiss-Mile to RIVR in January 2025, that builds wheeled-legged quadruped robots with learned control for last-mile delivery; it raised…
arXiv:2610.30810v1 Announce Type: new Abstract: Loco-manipulation extends the workspace of conventional manipulation by combining mobility with physical interaction…
arXiv:2611.25215v1 Announce Type: new Abstract: Autonomous navigation in diverse real-world environments remains one of the greatest challenges in robotics. It demands…
arXiv:2610.40799v1 Announce Type: new Abstract: Across coding, web, and tool-use tasks, language-model agents are now routinely trained with the same recipe: supervised…
arXiv:2611.25232v1 Announce Type: new Abstract: Ferrofluids are liquids containing magnetic nanoparticles, enabling shape control through external magnetic fields. This…
arXiv:2610.32143v1 Announce Type: new Abstract: Robot-assisted dressing is particularly challenging due to narrow openings and articulated body parts. Even small errors…
arXiv:2611.19318v1 Announce Type: new Abstract: We address tracking a moving object on a conveyor with a Franka Emika Panda robot arm, despite an imprecise digital twin…
arXiv:2611.18365v1 Announce Type: new Abstract: Automation of heavy machinery offers the opportunity to improve productivity in mechanized industries such as…
arXiv:2610.36726v1 Announce Type: new Abstract: This paper presents an approach to achieve side-by-side autonomous motion with a pair of personal mobility vehicles…
Boston Dynamics combines classical motor control with whole-body reinforcement learning to power dynamic, wheeled locomotion for Atlas. The post Boston Dynamics adds…
Dream Labs says its Dream-1 AI model works across robots from different manufacturers and will allow robots to transfer their learning. The post Dream Labs raises $90M…
Attendees will learn how integrated actuators can reduce engineering effort by simplifying system integration. The post Harmonic Drive to share how integrated actuators…
This White Paper gives robotics researchers and engineers an overview of a new large-scale motion capture dataset built to close the data gap limiting humanoid robot…
arXiv:2610.36079v1 Announce Type: new Abstract: End-to-end autonomous driving has predominantly relied on open-loop imitation learning for training, which presents two…
Computational methods that extract information from images or video, including detection, tracking, segmentation and depth estimation.
Sensor fusion combines measurements from multiple sensors to estimate a system or environment more reliably than a single stream alone.
arXiv:2611.19047v1 Announce Type: new Abstract: Diffusion policies, which generate short action sequences by denoising random noise, are a leading way to learn robot…
arXiv:2611.05776v1 Announce Type: new Abstract: Adaptive sampling can substantially improve the value of sparse ocean observations by directing mobile robots toward…
arXiv:2610.40922v1 Announce Type: new Abstract: End-to-end policies, from task-specific controllers to robot foundation models, learn complex behavior from raw…
arXiv:2611.19190v1 Announce Type: new Abstract: Reinforcement learning (RL) can teach robot teams to navigate shared spaces, but trial and error causes collisions that…
arXiv:2611.05929v1 Announce Type: new Abstract: Off-policy deep reinforcement learning trains on random samples from a replay buffer of past experience. In physical…
Members of the student-run MIT Combat Robotics Club learn through destruction.
arXiv:2611.13482v1 Announce Type: new Abstract: Predicting long action chunks, often spanning dozens of timesteps, has become a key ingredient in increasingly capable…
arXiv:2610.40828v1 Announce Type: new Abstract: Imitation learning policies are intended to make robotic manipulation robust to variations in object pose, appearance…
arXiv:2611.05754v1 Announce Type: new Abstract: We study physics-based retargeting of human demonstrations for robot learning: finding robot controls that reproduce…
arXiv:2611.08872v1 Announce Type: new Abstract: In recent years, massively parallelized rigid-body simulation has played a central role in robotics research, enabling…
arXiv:2610.31891v1 Announce Type: new Abstract: Data-driven robots must learn to avoid collisions with as few demonstrations as possible to avoid collecting large…
arXiv:2611.17802v1 Announce Type: new Abstract: Service robots in domestic and office environments must identify users to provide personalized services based on their…
arXiv:2611.19660v1 Announce Type: new Abstract: Physics-based control is central to robot learning, but differentiating through implicit physics steps typically requires…
arXiv:2611.23916v1 Announce Type: new Abstract: Humanoid robots must recover from falls and resume commanded motion to sustain locomotion. Learning these behaviors in…
arXiv:2611.18003v1 Announce Type: new Abstract: Reinforcement learning has enabled robust humanoid locomotion, yet training recipes (rewards, architectures…
arXiv:2611.09124v1 Announce Type: new Abstract: Foundation models increasingly guide robot actions, yet sustained autonomy requires efficient control and failure…
arXiv:2610.40748v1 Announce Type: new Abstract: Achieving the broad task generalization of embodied models in physical environments remains a central challenge in robot…
arXiv:2611.18186v1 Announce Type: new Abstract: On small single-board computers (SBCs), GNSS signal acquisition is a major CPU load. This paper proposes using the neural…
arXiv:2611.17394v1 Announce Type: new Abstract: Improving a pretrained vision-language-action (VLA) model requires demonstrations that address its weaknesses and provide…
Diana 7 is a 7-axis force-sensitive collaborative arm from Agile Robots with up to 7 kg payload.
The H2515 is a high-payload member of Doosan Robotics' H-series cobots, floor-mounted and rated for 25 kg within a 1,500 mm work radius.
The M1013 is Doosan Robotics' flagship M-series cobot with torque sensors in all six joints, rated for 10 kg at 1,300 mm.
The CS612 is a six-axis cobot in Elite Robots' CS series, rated for 12 kg with a 1304 mm working radius and IP65 protection (IP68 optional).
The CS66 is a six-axis cobot in Elite Robots' CS series, rated for 6 kg with a 914 mm working radius and IP65 protection (IP68 optional).
The KR810 is a 7-axis collaborative arm from Kassow Robots combining a 10 kg payload with an 850 mm reach; all Kassow cobots offer ROS/ROS 2 and Profinet interfaces.
Serve's third-generation delivery robot is the company's production sidewalk platform for restaurant and local-commerce deliveries.
A battery robot that drives on the floor and climbs storage racks to fetch bins for Exotec's Skypod goods-to-person system.
Mobile crop-harvesting robot developed to identify and pick ripe fruit in commercial production environments.
Walking robotic arm mounted on the Russian segment of the International Space Station for payload and maintenance tasks.
Agility Robotics develops Digit, a bipedal robot designed for material-movement tasks in human workspaces.
Built Robotics develops autonomy systems for heavy construction equipment.
Carbon Robotics develops computer-vision-guided agricultural equipment for weed control.
Locus Robotics develops autonomous mobile robots and orchestration software for warehouse workflows.
Unitree develops legged robots, including G1 humanoids and Go-series quadrupeds.
Shanghai-based developer of humanoid and embodied-AI robots (A, X, G, D and C series) founded in February 2023; its legal name is AGIBOT Innovation (Shanghai) Technology…
Munich robotics company founded in 2018 by researchers from the German Aerospace Center (DLR), making cobot arms, robotic hands, mobile robots and the Agile ONE humanoid.
Builds open mobile robot chassis, educational robots and lightweight arms for developers and integrators.
AUBO builds i-series collaborative robot arms covering 3 to 20 kg payloads.
Bear Robotics builds autonomous serving and hauling robots for restaurants and other venues. LG Electronics took a controlling 51% stake in early 2025.
Beijing-based embodied-AI company set up in November 2023 that develops the Tiangong humanoid platform and the Huisi Kaiwu embodied-intelligence software; in October…
Blue Robotics sells low-cost marine components such as thrusters, sonars and enclosures, along with the BlueROV2 underwater vehicle and BlueBoat surface vessel.
Humanoid robot company founded in June 2023 focused on research, education and robot-soccer developer platforms.
Burro makes person-following and autonomous utility robots that carry crops and tow loads in farms, nurseries and vineyards. The legal entity is Augean Robotics, Inc.