Weld defect detection is critical for maintaining the safe operation of natural gas pipelines. While computer vision-based approaches have emerged as a promising research focus, existing deep learning models inevitably face the problem of limited detection accuracy in multiscale welding defect detection, especially in recognizing tiny welding defects. To overcome the above challenges, this article proposes a novel DySample-guided lightweight feature aggregation fusion network (DyLFA-Net) for weld defect detection in natural gas pipelines, aiming to improve detection accuracy and computational efficiency while preserving a compact architectural design. The DyLFA-Net model consists of two key components. 1) The IB-SCSA module at the P5 layer, which enhances multiscale semantic extraction via spatial-channel attention. The IB module enhances feature expressiveness through channel expansion and depthwise convolution, supplying the SCSA mechanism with semantically enriched input features. 2) A high-resolution detection head at the P2 layer to improve tiny defect recognition. Experiments show that the proposed model achieves an [email protected] of 89.9%, outperforming mainstream detectors. Additionally, with just 1.31 M parameters and 6.6 GFLOPs, the proposed model offers high accuracy and low complexity, enabling real-time deployment on resource-limited edge devices.

Control policies that can achieve high task performance and satisfy safety constraints are desirable for any system, including multi-agent systems (MAS). One promising technique for ensuring the safety of MAS is distributed control barrier functions (CBF). However, it is difficult to design distributed CBF-based policies for MAS that can tackle unknown discrete-time dynamics, partial observability, changing neighborhoods, and input constraints, especially when a distributed high-performance nominal policy that can achieve the task is unavailable. To tackle these challenges, we propose **DGPPO**, a new framework that *simultaneously* learns both a *discrete* graph CBF which handles neighborhood changes and input constraints, and a distributed high-performance safe policy for MAS with unknown discrete-time dynamics. We empirically validate our claims on a suite of multi-agent tasks spanning three different simulation engines. The results suggest that, compared with existing methods, our DGPPO framework obtains policies that achieve high task performance (matching baselines that ignore the safety constraints), and high safety rates (matching the most conservative baselines), with a *constant* set of hyperparameters across all environments.

We address the problem of safe policy learning in multi-agent safety-critical autonomous systems. In such systems, it is necessary for each agent to meet the safety requirements at all times while also cooperating with other agents to accomplish the task. Toward this end, we propose a safe Hierarchical Multi-Agent Reinforcement Learning (HMARL) approach based on Control Barrier Functions (CBFs). Our proposed hierarchical approach decomposes the overall reinforcement learning problem into two levels –- learning joint cooperative behavior at the higher level and learning safe individual behavior at the lower or agent level conditioned on the high-level policy. Specifically, we propose a skill-based HMARL-CBF algorithm in which the higher-level problem involves learning a joint policy over the skills for all the agents and the lower-level problem involves learning policies to execute the skills safely with CBFs. We validate our approach on challenging environment scenarios whereby a large number of agents have to safely navigate through conflicting road networks. Compared with existing state-of-the-art methods, our approach significantly improves the safety achieving near perfect (within $5\%$) success/safety rate while also improving performance across all the environments.

Distributed, scalable, and safe control of large-scale multiagent systems is a challenging problem. In this article, we design a distributed framework for safe multiagent control in large-scale environments with obstacles, where a large number of agents are required to maintain safety using only local information and reach their goal locations. We introduce a new class of certificates, termed graph control barrier function (GCBF), which are based on the well-established control barrier function theory for safety guarantees and utilize a graph structure for scalable and generalizable distributed control of MAS. We develop a novel theoretical framework to prove the safety of an arbitrary-sized MAS with a single GCBF. We propose a new training framework GCBF+ that uses graph neural networks to parameterize a candidate GCBF and a distributed control policy. The proposed framework is distributed and is capable of taking point clouds from LiDAR, instead of actual state information, for real-world robotic applications. We illustrate the efficacy of the proposed method through various hardware experiments on a swarm of drones with objectives ranging from exchanging positions to docking on a moving target without collision. In addition, we perform extensive numerical experiments, where the number and density of agents, as well as the number of obstacles, increase. Empirical results show that in complex environments with agents with nonlinear dynamics (e.g., Crazyflie drones), GCBF+ outperforms the hand-crafted CBF-based method with the best performance by up to 20% for relatively small-scale MAS with up to 256 agents, and leading reinforcement learning (RL) methods by up to 40% for MAS with 1024 agents. Furthermore, the proposed method does not compromise on the performance, in terms of goal reaching, for achieving high safety rates, which is a common tradeoff in RL-based methods. Project website:https://mit-realm.github.io/gcbfplus/

In this paper, the configuration transformation of Wheel-Legged Robot (WLR) is studied, which can enable the robot to change its multilinks configuration on Inverted Equilibrium Manifold (IEM), while keeping balance with a small location drift on the floor. First of all, the general form of dynamics equation of planar Articulated Wheeled Inverted Pendulum (AWIP) with a wheel and n − 1 rigid links, is derived. The Partial Feedback Linearization (PFL) combined with a Sliding Mode Control (SMC) is used to design the inverse dynamics controller of AWIP, while considering full dynamics terms. The well-known WLR model is used as a simple example of AWIP to accomplish the configuration transformation task. An optimization based configuration transformation algorithm is proposed to realize a comprehensive optimization of the shortest path in joint space and the minimum location drift of WLR on the floor. Finally, the effectiveness of the proposed algorithm is demonstrated through simulation to implement the configuration transformation task.

Most existing imitation learning approaches assume the demonstrations are drawn from experts who are optimal, but relaxing this assumption enables us to use a wider range of data. Standard imitation learning may learn a suboptimal policy from demonstrations with varying optimality. Prior works use confidence scores or rankings to capture beneficial information from demonstrations with varying optimality, but they suffer from many limitations, e.g., manually annotated confidence scores or high average optimality of demonstrations. In this paper, we propose a general framework to learn from demonstrations with varying optimality that jointly learns the confidence score and a well-performing policy. Our approach, Confidence-Aware Imitation Learning (CAIL) learns a well-performing policy from confidence-reweighted demonstrations, while using an outer loss to track the performance of our model and to learn the confidence. We provide theoretical guarantees on the convergence of CAIL and evaluate its performance in both simulated and real robot experiments.Our results show that CAIL significantly outperforms other imitation learning methods from demonstrations with varying optimality. We further show that even without access to any optimal demonstrations, CAIL can still learn a successful policy, and outperforms prior work.

The performance of traditional hydraulic robots is often limited by their hoses across moving joints or connecting hydraulic drive units, which would reduce their mobility and impede their ability to operate in complex environment. In response to this deficiency, this paper introduces the WLR-II (the second generation of wheel-legged robot), a novel hydraulic wheel-legged robot developed by using hose-less design approach which is focused on improving the reliability of the hydraulic system and perfecting the appearance of the robot. As its notable features, seven Hydraulic Hose-less Joints (HHJ) that include a pair of high and also low pressure oil pipes based on rotary seal, Cylinder-Valve-Skeleton (CVS) integration thighs and arms which are produced by subtractive manufacturing as well as oscillating cylinders driven by gear rack transmission are included. In addition to a description of its design, experimental characterizations of rough pavement adaptability and payload capability together with the achievement of the reliability of hydraulic system are also demonstrated. As a result, we confirmed effectiveness of the hose-less design by moving on the rugged ground, climbing slope, squatting with load, dragging and picking up a heavy load. To the authors' best knowledge, this is the first time that the design of a hose-less hydraulic wheel-legged robot has been presented.

Wheel-legged hybrid robot with multi-modal locomotion can efficiently adapt to different terrain environments, as well as realize rapid maneuver on flat ground. We have developed a novel hydraulic wheel-legged robot (WLR) combined with a humanoid structural design. This robot can assist to emergency scenarios where the high mobility, adaptability and robustness are required. The paper introduces the details of the WLR, highlighting the innovative design and optimization of physical construction which is considered to maximize the mobile abilities, enhance the environmental adaptability and improve the reliability of hydraulic system. Firstly, maximizing the mobile abilities includes optimizing the configuration of each actuator and integrating them with the structure, so as to achieve a large range of movement and also reduce the mass and inertia of the legs. Secondly, the environmental adaptability can be ensured with a magnetorheological (MR) fluid-based damper and direct-drive wheels. Thirdly, improving the reliability of hydraulic system involves using the selective laser melting (SLM) technology to integrate hydraulic system and reducing the number of exposed tubes. The maneuverability of the WLR is demonstrated with a series of experiments. At present, the WLR can perform the following operations, including moving on the flat ground, squatting, and picking up a heavy load.

Prediction of interaction force between human and contacting environment has always been an important issue for such as robot control and clinical treatment. Although force/torque sensors can provide direct and precise measurement results, in many cases it is inconvenient to attach these sensors on the environment surface or on human beings. The purpose of this paper is to propose a prediction method for interaction force between human and contacting environment, using only electromyographic (EMG) signals. The motions are touch motion and push motion of upper extremity. Seven muscles of the upper limb were selected to record EMG signals. The predict function were derived from two muscle-skeleton models implemented for touch motion and push motion, respectively. The Bayesian linear regression (BLR) algorithm was implemented for parameters calibration. In order to avoid complex model for dynamic movement, a neural network classifier was used to recognize the force exerting motions. The proposed method was applied in a remote interaction force evaluation experiment. The “Phantom Premium” haptic device is used to represent the predicted force in the remote site. The experimental results show that the proposed method can provide acceptable prediction results with root-mean-square (RMS) error below 2.20 N.