Weekly supplier brief
What changed this week, and what to check before your next collection.
The last seven days
2026-09-29 – 2026-10-06 · UTC
Source-linked developments, grouped by collection relevance.
Broader tasks and environments
Collection consideration: Record task, object, environment and embodiment variation. Preserve each collection’s provenance rather than merging everything into one hours total. Research basis ↗
Robotics — Paper and dataset web discovery · · First discovered; source date unavailable
Vision-Language-Action in Robotics: A Survey of Datasets and Data Infrastructure
This survey paper examines the data infrastructure challenges in Vision-Language-Action (VLA) models for robotics. It categorizes datasets by embodiment diversity, modality composition, and action space formulation, identifies limitations in simulation-based and video-reconstruction paradigms, and outlines four open challenges: representation alignment, multimodal supervision, reasoning assessment, and scalable data generation.
Original source ↗LeRobot — X ·
The robotics data ecosystem keeps getting stronger 🦾
The author explores the LeRobot community dataset, using embeddings and FiftyOne to analyze 497 episodes across 50 robot types. They identify near-duplicates, filter sessions, and reveal cross-embodiment patterns like cloth folding. A curated subset is available for further exploration.
Original source ↗
Question to resolve: Can you describe the distinct tasks and settings, and deliver consistent robot states and action definitions?
Human video with action supervision
Collection consideration: Plan wrist and hand-action annotations, camera calibration and alignment with the target robot. Ordinary RGB footage alone does not provide the supervision used in this study. Research basis ↗
Robotics — Paper and dataset web discovery · · First discovered; source date unavailable
Physical AI at Scale: Why Robotics Needs a New Data Infrastructure
The paper explores the challenges and opportunities in scaling Physical AI for robotics, focusing on the need for real-world data, the limitations of simulation, and the importance of a continuous data flywheel. It discusses the role of reinforcement learning, human demonstration learning, and Edge AI in enabling scalable robotic deployments. Qualcomm's capabilities in XR, edge computing, and AI infrastructure are highlighted as key enablers for this emerging field.
Original source ↗arXiv Robotics — research abstracts ·
SoTa: Soft Tactile Skins for Dexterous Manipulation
This research presents SoTa, a low-cost capacitive tactile skin that provides full-hand coverage for both humans and robots. The sensor features a shared layout of 202 taxels across finger and palm regions, enabling human-robot co-training with a common tactile encoder. Tactile observations improve success rates in dexterous manipulation tasks, with human demonstrations more than doubling mean success across eight evaluation conditions.
Original source ↗Skild AI — Blog · · First discovered; source date unavailable
Learning from Human Videos for Robotics
Skild AI proposes using human video data to overcome robotics' data bottleneck. By observing human actions, robots can learn new tasks with minimal direct interaction.
Original source ↗arXiv Robotics — research abstracts ·
Towards a General Humanoid Loco-Manipulation Model via Egocentric Whole-Body Human Data Pretraining
The paper presents HumanVerse-500, a 500-hour dataset of human loco-manipulation behaviors collected with a lightweight wearable system. It introduces λ₀, a whole-body humanoid vision-language-action policy trained through three stages, achieving state-of-the-art performance on real-world tasks and analyzing how human data supports downstream control.
Original source ↗
Question to resolve: Which action representation can you deliver, and who will provide the paired robot data?
Failures, interventions and recovery
Collection consideration: Preserve failed attempts, intervention timing and recovery actions alongside successes. Agree on task outcomes and intervention labels before recording. Research basis ↗
Chris Paxton — X Robotics ·
DITTO-X: Bidirectional Teleoperation for Dexterous Robotic Hands
DITTO-X is a framework that enables both forward and reverse teleoperation for dexterous robotic hands. In forward teleoperation, humans control robots with haptic feedback, while in reverse teleoperation, the robot moves the human's hand to match its state, allowing for smooth intervention. This improves policy robustness through DAgger data collection.
Original source ↗Robotics — Paper and dataset web discovery · · First discovered; source date unavailable
Physical AI at Scale: Why Robotics Needs a New Data Infrastructure
The paper explores the challenges and opportunities in scaling Physical AI for robotics, focusing on the need for real-world data, the limitations of simulation, and the importance of a continuous data flywheel. It discusses the role of reinforcement learning, human demonstration learning, and Edge AI in enabling scalable robotic deployments. Qualcomm's capabilities in XR, edge computing, and AI infrastructure are highlighted as key enablers for this emerging field.
Original source ↗Chinese robotics — Hardware and sensing · · First discovered; source date unavailable
IROS 2026 Field Observations: Long-Term Tasks, Failure Recovery, World Models
The IROS 2026 conference highlights a shift towards long-term tasks, failure recovery, and world modeling in robotics. The event showcases advancements in perception, planning, and control, with a focus on real-world applications and standardized testing. The integration of tactile and force feedback, along with data collection tools, reflects a growing emphasis on embodied intelligence and simulation-to-reality transitions.
Original source ↗arXiv Robotics — research abstracts ·
RUL-Aware RRT*: Degradation-Balanced Motion Planning for Robotic Manipulators
RUL-aware RRT* integrates joint health data into motion planning, balancing degradation and improving long-term reliability. It reduces premature actuator failure and operational downtime in extended robotic operation.
Original source ↗arXiv Robotics — research abstracts ·
Filter-Aware Fine-Tuning for Safe Humanoid Whole-Body Tracking
This research introduces CoFiT, a filter-aware fine-tuning method for pretrained whole-body trackers in humanoid robots. By addressing mismatches between tracking policies and safety filters, CoFiT reduces violation time by 91% on TWIST2 and 21% on SONIC, while requiring smaller safety filter corrections. On real hardware, it achieves 83% reduction in violation time and completes trials without operator intervention.
Original source ↗
Question to resolve: Can a buyer distinguish autonomous actions from human corrections in each episode?
Data that can enter a training pipeline
Collection consideration: Deliver a data dictionary, episode boundaries, synchronized streams and a sample loader. Specify units, coordinate frames and missing-data handling. Research basis ↗
Chinese robotics — Robot learning research · · First discovered; source date unavailable
Embodied Intelligence Education in Practice: Four-Stage Transition from Classroom to Production Line
This article presents a comprehensive guide to implementing embodied intelligence education in robotics, focusing on the integration of physical interaction, sensor data, and real-world problem-solving. It outlines a four-stage model for transitioning from classroom to industrial application, emphasizing sensor calibration, algorithm visualization, and real-world testing. The tutorial includes practical insights, common pitfalls, and real-world examples to support educators and students in developing hands-on, industry-relevant skills.
Original source ↗LeRobot — X ·
The robotics data ecosystem keeps getting stronger 🦾
The author explores the LeRobot community dataset, using embeddings and FiftyOne to analyze 497 episodes across 50 robot types. They identify near-duplicates, filter sessions, and reveal cross-embodiment patterns like cloth folding. A curated subset is available for further exploration.
Original source ↗Chinese robotics — Dataset and collection discovery · · First discovered; source date unavailable
Guidelines for Building an Embodied Intelligence Lab at the University: From Hardware Selection to Data Closed-Loop
This guide addresses the challenges of building a functional embodied intelligence lab in academia, emphasizing the importance of data pipelines over hardware procurement. It covers hardware selection, sensor configuration, data collection, simulation platforms, and software architecture, offering practical advice for avoiding common pitfalls in lab setup and operation.
Original source ↗Chinese robotics — Robot learning research · · First discovered; source date unavailable
Embodied Intelligence Technology Architecture and Learning Path: From Large Models to Robot Deployment Practice
This article explores the technical architecture and learning path of embodied intelligence, focusing on the transition from algorithmic research to engineering implementation. It discusses the integration of AI models with robotic systems, practical challenges such as simulation-to-real transfer, model inference latency, and sensor calibration. The content is grounded in real-world development experiences and industry trends, offering actionable insights for both beginners and advanced developers.
Original source ↗Chinese robotics — Dataset and collection discovery · · First discovered; source date unavailable
Teaching Robots to Work: A Dataset and Collection Discovery in Chinese Robotics
At the 2026 China International Digital Economy Expo, a demonstration showcased how robots are being trained using real-world data collected through VR and exoskeletons. The article details the construction of over 70 embodied intelligence training facilities, the challenges of data collection and annotation, and the role of platforms like Qingyan Tech in data governance and trading. The focus is on the infrastructure and ecosystem supporting the AI robotics industry.
Original source ↗
Question to resolve: Can the buyer load a sample and align observations with actions without guessing the schema?
Other research and data developments
Open Robotics — X ·
We just posted an update for #ROS 2 Lyrical Luth that includes 29 new packages and 265 updated packages. New packages include: 👩🎤 cloudini 👩🎤 ehukai 👩🎤 fadecandy-msgs 👩🎤 mrpt 👩🎤 rmf-demos 👩🎤 roboplan Details on Open Robotics Discourse: https://discourse.openrobotics.org/t/new-packages-for-lyrical-2026-10-05/58605
We just posted an update for #ROS 2 Lyrical Luth that includes 29 new packages and 265 updated packages. New packages include: 👩🎤 cloudini 👩🎤 ehukai 👩🎤 fadecandy-msgs 👩🎤 mrpt 👩🎤 rmf-demos 👩🎤 roboplan Details on Open Robotics Discourse: https://discourse.openrobotics.org/t/new-packages-for-lyrical-2026-10-05/58605
Original source ↗Chinese robotics — Hardware and sensing · · First discovered; source date unavailable
IROS 2026 Highlight Chinese Robotics Hardware
IROS 2026 in Pittsburgh spotlighted Chinese robotics firms showcasing advanced hardware and sensor solutions. Focus shifted to practical applications like logistics and precision tasks.
Original source ↗ROS Discourse — Community updates ·
Gazebo PMC Meeting Minutes for October 5, 2026
PMC discussed adding mentee committers, async engagement, inactive member handling, and audit phase 2. They also addressed deprecated functions and release tooling improvements.
Original source ↗Robotics & Automation News ·
Warehouse Automation Starts with Storage Planning
Effective warehouse automation begins by aligning storage systems with inventory flow and operational needs. Storage layout impacts equipment selection, movement paths, and safety, requiring integration with building constraints and handling requirements.
Original source ↗Carnegie Mellon Robotics Institute — News ·
CMU Robotics Expertise and Facilities Attract IROS 2026 Conference Attendees
Carnegie Mellon University's Robotics Institute attracted hundreds of robotics experts and industry leaders to its facilities during IROS 2026, showcasing cutting-edge research and state-of-the-art infrastructure. The event highlighted bio-inspired robotics, autonomous drones, and student-led innovations, reinforcing Pittsburgh's status as a global robotics hub.
Original source ↗Robotics & Automation News ·
SafeWorld emerges from stealth with $12.2 million to develop robot safety simulation technology
SafeWorld, an AI lab focused on robot safety simulation, has raised $12.2 million in seed funding to develop tools that enable enterprises to test and deploy robots safely. The platform offers scalable simulation solutions to address the growing need for safety validation in AI-powered robotics.
Original source ↗Robotics & Automation News ·
Teradyne invests in Bright Machines to advance AI infrastructure manufacturing
Teradyne and Bright Machines have announced a strategic investment and collaboration to integrate Teradyne's robotics and test technologies with Bright Machines' software-defined manufacturing platform, aiming to advance AI infrastructure manufacturing through data-driven, reconfigurable production systems.
Original source ↗Chris Paxton — X Robotics ·
Touch is at the frontier of robotics
A demonstration at IROS 2026 by Daimon Robotics shows a robot threading beads onto a flexible string, highlighting the challenges of tactile interaction. The Daim, TWM model uses touch to predict and adjust actions, suggesting that robots may need to learn both visual and tactile properties of objects for effective manipulation.
Original source ↗Robotics & Automation News ·
amsight leads ESA project to speed qualification of 3D-printed space components
amsight has been selected to lead the ESA DIQAM project, which seeks to accelerate the qualification of 3D-printed space components through a digital qualification framework. The project aims to make the qualification process faster, more data-driven, and reusable across future builds. It involves developing a Digital Part Qualification File and a software prototype for automated data ingestion and statistical evaluation. The framework will be validated using two industrial demonstrators, including an ArianeGroup Phase Change Material container and an ISPTech 4U propellant tank.
Original source ↗Robotics — Paper and dataset web discovery · · First discovered; source date unavailable
Vision-Language-Action in Robotics: Dataset Survey
Study reviews datasets, benchmarks, and data engines for Vision-Language-Action in robotics. Published in TMLR 2026.
Original source ↗
How this brief works
This rolling brief uses the latest published research, model, dataset and simulation coverage, up to 80 reports. Topic matches are reading aids. They do not prove a model used a dataset or that a buyer is purchasing data.
Collection considerations are RoboSignal’s interpretation of the cited research. They are not new findings from every linked report.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).