The development of embodied AI (EAI) critically depends on efficient data acquisition, yet faces persistent challenges including high costs, limited training scenarios, and lack of standardized datasets. We present AIRSPEED, an open source data production platform designed to address these bottlenecks through three core innovations. First, AIRSPEED achieves hardware–software decoupling via unified robot and simulation interfaces, enabling seamless integration with diverse data collection devices and simulation platforms. Second, it supports comprehensive data production methods spanning teleoperation and teaching approaches, as well as synthetic data generation through data synthesis and virtual teleoperation. Third, AIRSPEED automates pyramid-structured dataset construction compatible with both HDF5 and LeRobot formats, significantly reducing manual overhead. Experimental validation demonstrates substantial efficiency gains, achieving up to 35.6× acceleration in dataset construction and 6.0× overall speedup compared to manual workflows. With end-to-end latency as low as 3 ms and compression throughput exceeding 296 MB/s, AIRSPEED establishes a scalable foundation for EAI data production. AISPEED is open sourced on this website: URL .

Inertial parameter identification is crucial for accurate robot control, but existing methods for fixed-base manipulators are insufficient for floating-base systems. To address this, we propose the Decomposed Inertia Identification (DII) framework, which utilizes inertia transfer theory and the Recursive Parameter Null Space Algorithm (RPNA) to decompose base parameters into fixed-base and residual subsets. This approach reduces optimization complexity and enables symbolic parameter identification. Inspired by animal spinning behaviors, we use spinning trajectories to excite leg dynamics, overcoming high-DoFs challenges. The Covariance Matrix Adaptation Evolution Strategy (CMA-ES) optimizes parameters under physical consistency constraints. The method was validated on a Unitree Go1 quadruped robot, achieving 98.1% parameter convergence within 200 iterations during spinning locomotion (0.5–2 rad/s yaw velocity). Updating inertial parameters reduced tracking errors by 63% in body posture control and improved straight-line locomotion accuracy by 89% under payload variations on leg. The DII framework bridges fixed- and floating-base systems, enabling the application of mature fixed-base methodologies to floating-base robots and advancing self-model identification for real-world applications.

In-Context Reinforcement Learning (ICRL) enables agents to learn automatically and on-the-fly from their interactive experiences. However, a major challenge in scaling up ICRL is the lack of scalable task collections. To address this, we propose the procedurally generated tabular Markov Decision Processes, named AnyMDP. Through a carefully designed randomization process, AnyMDP is capable of generating high-quality tasks on a large scale while maintaining relatively low structural biases. To facilitate efficient meta-training at scale, we further introduce decoupled policy distillation and induce prior information in the ICRL framework. Our results demonstrate that, with a sufficiently large scale of AnyMDP tasks, the proposed model can generalize to tasks that were not considered in the training set through versatile in-context learning paradigms. The scalable task set provided by AnyMDP also enables a more thorough empirical investigation of the relationship between data distribution and ICRL performance. We further show that the generalization of ICRL potentially comes at the cost of increased task diversity and longer adaptation periods. This finding carries critical implications for scaling robust ICRL capabilities, highlighting the necessity of diverse and extensive task design, and prioritizing asymptotic performance over few-shot adaptation.

This study introduces a novel climbing strategy, reconfigurable parallel-type cable-driven climbing designed for long-span, large-scale bridge stay cable robotic applications, which has the potential to revolutionize the stay cable inspection and maintenance practice. The proposed methodology features the development of a Collaborative Climbing Robot Squad (CCRobot-S), which builds upon the design principles of the previous CCRobot series. In this study, CCRobot-S implements a parallel-type cable-driven manipulation design, allowing for reconfigurable kinematic morphology by its movable anchor bases and realizing the capacity of crossing over the stay cables for its flying platform. The collaborative robot squad design liberates the dimensions and scales of the robot's reachable workspace and moves the part of the robotic system that indeed needs to be moved, enhancing the working efficiency and climbing agility. This strategy also utilizes controllable adhesion instead of friction to interact with the bridge cable surface for the flying platform, realizing force multiplication for forceful manipulation. Toward bringing high efficiency and heavy-duty capacity, we propose the applicable climbing frameworks (zero-downtime climbing gait for cable inspection and spider-like climbing gait for cable maintenance) and the optimization frameworks (optimal anchor configuration for the movable anchor bases and optimal grasp arrangement for the flying gripper). This article includes the exploration of the design and climbing gaits of CCRobotS, the formulation of the CCRobot-S model, a comprehensive analysis of its workspace, and its climbing strategy and optimization. Extensive experiments have assessed the proposed climbing strategy's effectiveness and showcased CCRobot-S' capabilities.

Automatic 3D shape analysis is heavily influenced by the pose of input 3D models, as the continuous nature of pose space introduces complexities that usually exceed the encoding capacities of standard deep learning frameworks. To tackle this challenge, we present Upright-Net+, an enhancement of our previous model, Upright-Net, specifically developed for estimating upright orientation in 3D point clouds. Our approach is grounded in the design principle that "form ever follows function," treating the natural base of an object as a functional structure that stabilizes it in its typical pose, influenced by physical laws and geometric properties. We reformulate the continuous orientation problem into a discrete classification task, focusing on learning the points that constitute the natural base of a 3D model. The upright orientation is determined by aligning the normal orientation of this base towards the mass center. To mitigate over-smoothing in the global feature embeddings from stacked graph convolutional layers, we introduce a Global Positional Encoding Module using Relative Distance Histogram Statistics Embedding (GPE-RDHS), which reduces structural ambiguity and enhances orientation estimation. We also enhanced a weighted residual loss term to penalize false positive predictions, enhancing overall model performance. Our method demonstrates exceptional performance in upright orientation estimation and reveals that the learned orientation-aware features significantly benefit downstream tasks, particularly in classification.

While industrial anomaly detection (AD) technology has experienced notable advancements in recent years, the challenges of generating realistic anomalies and acquiring a deep understanding of normal patterns persist. In this research, we introduce FractalAD, an end-to-end industrial anomaly detection method. The training dataset is curated by synthesizing fractal images and patches derived from normal samples. This unique fractal anomaly generation method is meticulously crafted to capture the intricate morphology of anomalies comprehensively. Furthermore, we have devised a knowledge distillation structure, aimed at extracting inherent knowledge present in normal samples. To bridge the gap between a teacher and a student model, discrepancies are translated into anomaly attention through a cosine similarity attention module. Notably, our proposed method facilitates the utilization of an end-to-end semantic segmentation network for anomaly detection, seamlessly integrating into the existing framework without introducing additional trainable parameters to the backbone and segmentation head. It exhibits clear advantages over alternative methods in terms of both training efficiency and inference speed. The outcomes of ablation studies underscore the efficacy of both fractal anomaly generation and backbone knowledge distillation. Performance experiments substantiate the competitive standing of FractalAD, showcasing commendable results on the MVTec AD dataset and MVTec 3D-AD dataset when compared to other state-of-the-art anomaly detection methods.

The concentric cable-driven manipulator (CCDM) has the characteristics of high dexterity, light weight, and safe movement, making them widely used in confined spaces. However, there are difficulties in the fine control with proper stiffness of CCDMs due to their flexible structures and various configurations. This article proposes an analytical variable-stiffness method for the fine control of CCDMs. First, the stiffness model is established by taking into account key factors, including the middle elastic backbone, cable tensions, configurations, and external loads. Then, the stiffness mesh is generated based on the stiffness model, which visually represents changing trends of its stiffness. Simultaneously, the stiffness of CCDMs can be accurately adjusted by optimizing their configurations and cable tensions. Therefore, the fine control with high or low stiffness of CCDMs can be realized in practical applications. Finally, experiments are conducted to verify the analytical variable-stiffness method of CCDMs. Results indicate that the average error of the stiffness model is 5.42%. It also confirms the effectiveness of the proposed method for achieving analytical variable-stiffness control of CCDMs. Furthermore, the proposed method is also applicable to cable-driven manipulators with similar structures.

A mass of experiments shows that the pose of the input 3D models exerts a tremendous influence on automatic 3D shape analysis. In this paper, we propose Upright-Net, a deep-learning-based approach for estimating the upright orientation of 3D point clouds. Based on a well-known postulate of design states that “form ever follows function”, we treat the natural base of an object as a common functional structure, which supports the object in a most commonly seen pose following a set of specific rules, e.g. physical laws, functionality-related geometric properties, semantic cues, and so on. Thus we apply a data-driven deep learning method to automatically encode those rules and formulate the upright orientation estimation problem as a classification model, i.e. extract the points on a 3D model that forms the natural base. And then the upright orientation is computed as the normal of the natural base. Our proposed new approach has three advantages. First, it formulates the continuous orientation estimation task as a discrete classification task while preserving the continuity of the solution space. Second, it automatically learns the comprehensive criteria defining a natural base of general 3D models even with asymmetric geometry. Third, the learned orientation-aware features can serve well in downstream tasks. Results show that our network outperforms previous approaches on orientation estimation and also achieves remarkable generalization capability and transfer capability.

This paper presents CCRobot-V, the fifth version of CCRobot, a cooperative serial multi-robot system for bridge cable inspection and maintenance that uses silkworm-like locomotion to climb the entire length of super-long stay cable at high speeds while carrying heavy inspection/maintenance equipment. CCRobot-V consists of one climbing precursor robot, one inspection/maintenance robot, several cable-carrying robots, and a power-tethered cable guiding system. The pre-cursor robot is the “head,” which leads the affiliated sub-robots along the bridge cable. Every sub-robot possesses a pair of self-locking palms. When a sub-robot grips on the bridge cable with its palms, it becomes a fixed anchor point that allows the adjacent sub-robots in front and back to use winches and steel wires to pull themselves upward. With this cooperative multi-robot system, cable inspection/maintenance tasks can be divided into several functional units, with each inspection/maintenance equipment installed separately on a customized sub-robot. Thus, CCRobot-V provides a complete mobile inspection/maintenance work line for a bridge cable. The experimental and field tests demonstrate CCRobot-V's high climbing speed, high payload capacity, and full-length cable moving capability. It has the potential application value for the actual bridge cable inspection/maintenance.

In typical data-based grasping methods, a grasp based on parallel-jaw grippers is parameterized by the center of the gripper, the rotation angle, and the gripper opening width so as to predict the quality and pose of grasps at every pixel. In contrast, a grasp is represented using only two contact points for contact-points-based grasp representation, which allows for fusion with tactile sensors more naturally. In this work, we propose a method using contact-points-based grasp representation to get a robust grasp using only one contact points quality map generated by a neural network, which significantly reduces the complexity of the network with fewer parameters. We provide a synthetic dataset including depth image and contact points quality map generated by thousands of 3D models. We also provide the method for data generation, which can be used for contact-points-based multi-fingers grasp. Experiments show that contact points quality network can plan an available grasp in 0.15 seconds. The grasping success rate for unknown household objects is 94%. Our method is also available for deformable objects with a success rate of 95%. The dataset and reference code can be found on the project website: https://sites.google.com/view/cpqnet.

Superior dexterity and extreme flexibility are typical advantages for concentric cable-driven manipulators working in confined spaces. However, its inverse kinematics and configuration planning are very complicated. In this article, we propose a spatial biarc method for the above problem. The distinguishing feature of this method is that input parameters are two positions and two direction vectors in three-dimensional (3-D) space, and the output is a reasonable spatial biarc for controlling a concentric cable-driven manipulator in 3-D space. This method has the following three advantages. First, the positions and direction vectors of the base and inner distal tip are considered simultaneously. In addition, the length and ratio of the overlapped section and separated section can be adjusted by changing the length of the direction vectors. Furthermore, by judging the angular value of the direction vectors, one can predetermine whether the spatial configuration of the entire arm is C- or S-shaped. The proposed method realizes the parameterization of a concentric cable-driven manipulator, which makes it convenient to intuitively control the manipulator to achieve interference-free motion trajectory planning in confined spaces. Finally, trajectory tracking inspections are simulated and experimentally executed. It can be seen from results that the proposed spatial biarc method can provide reasonable solutions for concentric cable-driven manipulators. The method is especially favorable in terms of 3-D-pose-determination problem and trajectory-planning problem. It can also be applied to other manipulators with similar configurations. Without loss of generality, when the given points and direction vectors are coplanar, the proposed spatial biarc method can be transformed to a planar biarc method.

A Flying-type cable climbing robot, CCRobot-IV-F, is presented in this paper. It is a climbing precursor of the fourth version of CCRobot, designed to surpass the abilities of previous robots with high climbing speed and obstacle-crossing capability. CCRobot-IV-F weighs less than 10 kg and a no-load speed of up to 4.5 m/s, which significantly exceeds that of other climbing robots. A dynamic model integrated with a cable-fixed coordinate system is developed, and a cascaded controller designed for stabilizing hover and climb with grippers, when a Global Positioning System and magnetometer are unavailable, is shown to work reliably in practice. Experimental results show that CCRobot-IV-F significantly improves the locomotive performance of CCRobot-IV, exhibiting fast speed, good payload capacity, and excellent obstacle-crossing capability. Moreover, CCRobot-IV-F is applied to a cable-stayed bridge in the field.

This study focuses on the multiphase flow properties of crowd motions. Stability is a crucial forewarning factor for the crowd. To evaluate the behaviors of newly arriving pedestrians and the stability of a crowd, a novel motion structure analysis model is established based on purposiveness, and is used to describe the continuity of pedestrians’ pursuing their own goals. We represent the crowd with self-driven particles using a destination-driven analysis method. These self-driven particles are trackable feature points detected from human bodies. Then we use trajectories to calculate these self-driven particles’ purposiveness and select trajectories with high purposiveness to estimate the common destinations and the inherent structure of the crowd. Finally, we use these common destinations and the crowd structure to evaluate the behavior of newly arriving pedestrians and crowd stability. Our studies show that the purposiveness parameter is a suitable descriptor for middle-density human crowds, and that the proposed destination-driven analysis method is capable of representing complex crowd motion behaviors. Experiments using synthetic and real data and videos of both human and animal crowds have been conducted to validate the proposed method.

This paper investigates the cooperative target enclosing of multiple unicycle-type mobile robots subject to input disturbances. The objective is to make all robots orbit around a given stationary target, and maintain evenly spaced along a common circle. The network of the multirobot systems is set in a cyclic pursuit manner. A dynamic control law is developed for the cooperative target enclosing of the multirobot systems, while tackling the heterogeneous input disturbances generated by linear exogenous systems. The proposed control law requires each robot to use the relative displacement measurements with respect to the target and its neighbors. It is shown that global asymptotic stability of the closed-loop multirobot systems can be guaranteed in the presence of a large class of input disturbance signals. Finally, simulation results illustrate the effectiveness of our approach.

Fast and accurate road damage detection is essential for the automatization of road inspection. This paper describes our solution submitted to the Global Road Damage Detection Challenge of the 2020 IEEE International Conference on Big Data, for typical road damage detection in digital images based on deep learning. The recently proposed YOLOv4 is chosen as the baseline network, while the effects of data augmentation, transfer learning, Optimized Anchors, and their combination are evaluated. We propose a novel road damage data generation method based on a generative adversarial network, which can generate multi-class samples with a single model. The evaluation results demonstrate the effectiveness of different tricks and their combinations on the road damage detection task, which provides a reference for practical application. The code of our solution is available at https://github.com/ZhangXG001/RoadDamgeDetection.git.

This paper presents a novel Cable Climbing Robot CCRobot-III, which is the third version designed for bridge cable inspection tasks, aiming at surpassing previous versions in terms of climbing speed and payload capacity. Benefiting from Split-type Wire-driven design, CCRobot-III can climb along a 90-110mm diameter bridge cable in inchworm-like gait at a speed of up to 12m/min, and carrying more than 40kg payload at the same time. CCRobot-III consists of a climbing precursor and a main-body frame. The two parts are connected and driven by steel wires. The climbing precursor, acting as a mobile anchor, moves quickly on a bridge cable. The mainbody frame, acting as a mobile winch, carries payload and pulls itself to a certain position with steel wires. Both parts have one or two pairs of palm-based gripper, which is the key component for providing strong adhesion to support the robot climbing. Experimental results have shown that CCRobotIII possesses outstanding climbing performance, high payload capacity, and good adaptability to complex conditions of cable surface. Moreover, it has potential engineering applications on the cable-stayed bridge for fieldwork.

This paper investigates the cooperative moving-target enclosing control problem of networked unicycle-type nonholonomic vehicles with constant linear velocities. The information of the target is only known to some of the vehicles, and the topology of the vehicle network is described by a directed graph. A dynamic control law is proposed to steer the vehicles, such that they can get close to orbiting around the target while the target is moving with a time-vary velocity. Besides, the constraint of bounded angular velocity for the vehicles can always be satisfied. The proposed control law is distributed in the sense that each vehicle only uses its own information and the information of its neighbors in the network. Finally, simulation results of an example validate the effectiveness of the proposed control law.

This project aims at developing a bio-inspired climbing robotic technology for cable inspection on the cable-stayed bridge. The design and implementation of a palm-based cable climbing robot: CCRobot-II with mass 25 kg, maximal payload 30kg and maximal length 1.1 m are described. CCRobot-II consists of several novel design features, including the palm-based gripping module, the alternating-sliding frame specialized for high-speed climbing, and the following wheels. With its carefully designed mechanism, climbing gait and trajectory algorithm, CCRobot-II is able to crawl along a bridge cable with a maximal speed 5.2 m/min. This speed has been hardly achieved by a existing climbing robot with such a large scale, and it is nearly the twice as the climbing speed of CCRobot-I which is designed previously. CCRobot-II also works effectively even if the bridge cable surface is attached with small obstacles. Experiments have been conducted, the results show that CCRobot-II has potential engineering applications on the cable-stayed bridge for fieldwork.

This paper investigates dynamic and compliant control based on joint output torque estimation for electrically actuated quadruped robots with large-reduction-ratio harmonic gear. Compared with position control, force control exhibits better performance of dynamics and compliance for the robot's interactions with complex environments. However, force control without direct feedbacks from torque sensors may come with poor tracking performance of joint compliance when the robot equipped with gears of high reduction. To solve this problem, we propose a new method to estimate joint torque from motor current and rotation velocity detected on each joint, using a more precise friction model of the harmonic gear. We also introduce a pre-stance phase to the whole cycle of leg alternating swing/stance based on hybrid force and position control to dynamically absorb feet impacts on the ground. Our controller performance is validated by standing experiment and walking experiment.

This paper1presents a novel shape representation approach, Finite Element Contour (FEC), based on studies of shape analysis from the perspective of Finite Element Method (FEM). We assume that an edge of a contour can be modeled as a bendable beam element. Linking finite number of beam elements end to end along the contour, we obtain a closed-loop Finite Element Contour model as an approximate physical model of the original shape. By this model, we can calculate its natural frequency, which is one of mechanical properties that directly related to the geometric shape, and employed it as the shape representation. FEC shape feature possesses translation and rotation invariant properties naturally. We also realized scale and unique affine normalization in few simple steps based on intrinsic physical properties of shape from FEM viewpoint. Experimental results validated that the proposed FEC feature is capable of identifying shape in object recognition task. It also can describe shape deformation. In the well-known MPEG 7 shape retrieval task, the enhanced FEC approach obtains Bullseye score 87.11%.

As one of the most unique types of art in Chinese culture, nowadays Chinese calligraphy is attracting increasing interests from researchers. It will be a big step if we have a robot to write Chinese calligraphy, especially in various styles, and it will form a bridge to combine science with art directly. However, there are so many different types and styles in Chinese calligraphy, and to distinguish them is the most basic quality to a green hand, but it is a big challenge for a robot to do so. For the lack of exploration about this, we conduct a lot of experiments to help the robot to accomplish it automatically. We first propose a parametric representation of calligraphic characters, and then adopt the Mahalanobis distance for similarity measurement and classification. The average accuracies of classifying types and styles of the Chinese calligraphy are 96.36% and 95.61% respectively. During the experiments, some interesting phenomena are discovered through similarity measure. Meanwhile, the parametric representation also has some potential applications, such as defining aesthetic grading standards of calligraphy and synthesizing calligraphy. Based on our research, the calligraphy robot can tell which style of calligraphy it sees for mimicking.