Accurately predicting the probabilistic stress field in rotating diodes is a critical computational mechanics challenge for reliability design, as it necessitates quantifying both the long-range stress interactions and the inherent prediction uncertainties. Conventional approaches, such as finite element analysis, are often computationally prohibitive, while existing deep learning surrogates typically lack mechanisms for capturing non-local physical dependencies and providing confidence measures. To address these dual gaps, this paper proposes a physics-aware probabilistic computational framework, pd_unet_r2, which integrates a peridynamic-inspired non-local attention mechanism with Monte Carlo Dropout. The non-local module, configured with an optimal radius R=2, explicitly models physically bounded long-range stress redistribution, while Monte Carlo Dropout approximates Bayesian inference to quantify epistemic uncertainty with near-perfect calibration (ECE  <  0.03). Applied to the stress field analysis of rotating diode chips, the framework achieves high-fidelity deterministic predictions and well-calibrated probabilistic outputs; a companion configuration with R=3 demonstrates a 52.0% reduction in normalized root-mean-square error compared to the baseline U-Net, confirming the mechanism’s upper-bound capability. The inference speedup exceeds 276, 000 ×  relative to finite element analysis, enabling near-real-time reliability assessment. This work presents a feasible integration scheme of mechanism-informed deep learning for probabilistic computational mechanics, providing a practical computational framework for rotating machinery design with potential applicability to multiphysics scenarios.

Safe and efficient trajectory planning in unknown, cluttered 3D environments constitutes a critical bottleneck for deploying Unmanned Aerial Vehicles (UAVs) in real-world applications. This challenge is further exacerbated by the limited field-of-view (FOV) and sensing range of onboard sensors. Many existing methods either make simplistic assumptions about unexplored space or rely on conservative heuristics such as speed limits or fixed perception patterns, reducing efficiency and generalizing poorly across different sensor types. In this work, we propose a novel planning framework that directly integrates active perception into trajectory optimization, thereby improving safety while preserving efficiency. The perception constraints are derived from the UAV's dynamic model and formulated in the sensor coordinate frame, which enables precise handling of FOV geometry. The velocity-triggered activation mechanism enables the planner to balance perception and motion efficiency. We introduce an active perception sub-trajectory segment with parametric start-time optimization, mitigating collision risks from late obstacle detection. Our formulation enables active perception during arbitrary 3D maneuvers, extending beyond prior methods designed mainly for horizontal motion. All constraints and penalties are incorporated into a differentiable optimization problem, so the planner requires only a simple front-end global path for guidance, rather than a computationally expensive perception-aware path generator. Extensive simulations and real-world experiments demonstrate robust performance across diverse unknown environments with varying sensor configurations.

Applying large-scale vision-language pre-trained models like CLIP to few-shot action recognition (FSAR) can significantly enhance both performance and efficiency. While several studies have recognized this advantage, most rely on full-parameter fine-tuning to adapt CLIP's visual encoder to FSAR data, which not only incurs high computational costs but also overlooks the potential of the visual encoder to engage in temporal modeling and focus on targeted semantics directly. To tackle these issues, we introduce MA-FSAR, a framework that employs the Parameter-Efficient Fine-Tuning (PEFT) technique to enhance the CLIP visual encoder in terms of action-related temporal and semantic representations. Our solution involves a token-level Fine-grained Multimodal Adaptation mechanism: a Global Temporal Adaptation captures motion cues from video sequences, while a Local Multimodal Adaptation integrates text-guided semantics from the support set to emphasize action-critical features. Additionally, we propose a prototype-level text-guided construction module to further enrich the temporal and semantic characteristics of video prototypes. Extensive experiments demonstrate our superior performance in various tasks using minor trainable parameters.

With the increasing demand for automated inspection solutions in complex industrial environments, existing robotic platforms face significant limitations in terms of endurance, payload capacity and obstacle-crossing capabilities. In this paper, we present a novel inspection robot system based on the CubeTrack tracked platform, featuring a large configuration space achieved through the integration of a manipulator and advanced mobility mechanisms. Our system incorporates a quad-slider elliptical trammel mechanism (Qs-ETM) that enables geometry-changing tracks for enhanced terrain adaptability while maintaining track tension stability. To address multi-layer navigation challenges, we propose an efficient trajectory planning algorithm that extracts traversable planes from three-dimensional (3D) point clouds and constructs a lightweight plane graph for path optimisation. Additionally, we develop a flipper control algorithm that uses only low-cost local sensor measurement (time-of-flight [TOF] sensors and inertial measurement unit [IMU]) to enable autonomous stair navigation without pre-mapped environments. The inspection system integrates multiple sensors, including light detection and ranging (LiDAR) sensor, RGB cameras, gas sensors and thermal cameras, providing comprehensive monitoring capabilities for industrial inspection demands. Extensive real-world experiments demonstrate the system's effectiveness in navigating complex environments with stairs, multiple layers and narrow passages, validating both the mechanical design and algorithmic approaches for practical industrial inspection tasks.

Transporting a heavy payload using multiple aerial robots (MARs) is an efficient manner to extend the load capacity of a single aerial robot. However, existing schemes for the multiple aerial robots transportation system (MARTS) still lack the capability to generate a collision-free and dynamically feasible trajectory in real-time and further track an agile trajectory especially when there are no sensors available to measure the states of payload and cable. Therefore, they are limited to low-agility transportation in simple environments. To bridge the gap, we propose complete planning and control schemes for the MARTS, achieving safe and agile aerial transportation (SAAT) of a cable-suspended payload in complex environments. Flatness maps for the aerial robot considering the complete kinematical constraint and the dynamical coupling between each aerial robot and payload are derived. To improve the responsiveness for the generation of the safe, dynamically feasible, and agile trajectory in complex environments, a real-time spatio-temporal trajectory planning scheme is proposed for the MARTS. Besides, we break away from the reliance on the state measurement for both the payload and cable, as well as the closed-loop control for the payload, and propose a fully distributed control scheme to track the agile trajectory that is robust against imprecise payload mass and non-point mass payload. The proposed schemes are extensively validated through benchmark comparisons, ablation studies, and simulations. Finally, extensive real-world experiments are conducted on a MARTS integrated by three aerial robots with onboard computers and sensors. The result validates the efficiency and robustness of our proposed schemes for SAAT in complex environments.

CoNi-MPC [Zhang et al. (2023)] provides an efficient framework for UAV control in air-ground cooperative tasks by relying exclusively on relative states, eliminating the need for global state estimation. However, its lack of environmental information poses significant challenges for obstacle avoidance. To address this issue, we propose a novel obstacle avoidance algorithm, Cooperative Non-inertial frame-based Obstacle Avoidance (CoNi-OA), designed explicitly for UAV-UGV cooperative scenarios without reliance on global state estimation or obstacle prediction. CoNi-OA uniquely utilizes a single frame of raw LiDAR data from the UAV to generate a modulation matrix, which directly adjusts the quadrotor's velocity to achieve obstacle avoidance. This modulation-based method enables real-time generation of collision-free trajectories within the UGV's non-inertial frame, significantly reducing computational demands (less than 5 ms per iteration) while maintaining safety in dynamic and unpredictable environments. The key contributions of this work include: 1) a modulation-based obstacle avoidance algorithm specifically tailored for UAV-UGV cooperation in non-inertial frames without global states; 2) rapid, real-time trajectory generation based solely on single-frame LiDAR data, removing the need for obstacle modeling or prediction; and 3) adaptability to both static and dynamic environments, thus extending applicability to featureless or unknown scenarios.

The use of machine learning in fluid dynamics is becoming more common to expedite the computation when solving forward and inverse problems of partial differential equations. Yet, a notable challenge with existing convolutional neural network (CNN)-based methods for data fidelity enhancement is their reliance on specific low-fidelity data patterns and distributions during the training phase. In addition, the CNN-based method essentially treats the flow reconstruction task as a computer vision task that prioritizes the element-wise precision which lacks a physical and mathematical explanation. This dependence can dramatically affect the models' effectiveness in real-world scenarios, especially when the low-fidelity input deviates from the training data or contains noise not accounted for during training. The introduction of diffusion models in this context shows promise for improving performance and generalizability. Unlike direct mapping from a specific low-fidelity to a high-fidelity distribution, diffusion models learn to transition from any low-fidelity distribution towards a high-fidelity one. Our proposed model - Physics-informed Residual Diffusion, demonstrates the capability to elevate the quality of data from both standard low-fidelity inputs, to low-fidelity inputs with injected Gaussian noise, and randomly collected samples. By integrating physics-based insights into the objective function, it further refines the accuracy and the fidelity of the inferred high-quality data. Experimental results have shown that our approach can effectively reconstruct high-quality outcomes for two-dimensional turbulent flows from a range of low-fidelity input conditions without requiring retraining.

Accurate physical simulation is fundamental to science and engineering, yet conventional numerical solvers incur high costs when handling complex geometries, varying boundary and initial conditions, and diverse physical parameters. Recent deep-learning-based methods offer faster solutions, while limited flexibility and generalization on irregular meshes still hinder their practical deployment. Here we show an efficient graph-transformer operator, named PhysGTO, for learning physical dynamics through explicit manifold embeddings in both physical and latent spaces. The method aligns heterogeneous node-level conditions, constructs sparse structure-preserving connections, and integrates lightweight local message passing with global attention to capture multiscale physical dependencies. Its design scales linearly with the number of mesh points, reducing model size and computational cost while enabling efficient inference. On a benchmark of 11 datasets covering irregular meshes, time-dependent flows, and large three-dimensional geometries, PhysGTO achieves state-of-the-art accuracy with substantially lower computational cost, showing strong flexibility, scalability, and generalization across diverse physical systems.

Imitation learning on human demonstration mechanisms and exhibits excellent skill acquisition performance. To explore the efficiency-enhancing mechanisms of human demonstration in the joint space and the robustness of multi-task imitation, we propose a novel approach called an intuitive multi-task manipulation policy for imitation learning in simulated and real robots (MIRO). First, we design a constrained joint alignment teleoperation. By incorporating kinematics, smoothness, collision, and contact constraints into joint control, the robot can intuitively imitate human motion while maintaining good data acquisition efficiency. Second, we design a multi-task context-fusion imitation policy. By integrating language-guided encoding, multi-scale context enhancement, and gated feature fusion mechanisms, we improve multi-modal trajectory learning and enhance its adaptability to diverse tasks in diverse scenarios. Finally, we validate MIRO’s effectiveness by applying it to simulated and real robotic manipulation tasks and present a set of evaluation results, demonstrating its potential in various applications.

This study introduces a control framework designed to improve the disturbance rejection capability of uncrewed aerial vehicles (UAVs) by integrating learning-augmented disturbance estimation into the conventional composite hierarchical antidisturbance control architecture. A sensorless freestream velocity estimation scheme is developed to enable real-time freestream velocity estimation without requiring additional hardware. Leveraging this estimation, Gaussian process regression is employed to accurately model and integrate with model predictive control to compensate for wind-induced external forces. Furthermore, an SO(3)-based backstepping attitude controller is designed, incorporating a nonlinear disturbance observer to actively compensate for external moment disturbances. Rigorous Lyapunov analysis guarantees the bounded convergence of both attitude and estimation errors, ensuring robust performance in the presence of external disturbances. The effectiveness of the proposed method is validated through high-fidelity simulations and real-world flight experiments conducted on a medium-sized tail-sitter UAV. The experimental results demonstrate significant improvements in robustness under complex operating conditions.

Parallel trajectory optimization via the Alternating Direction Method of Multipliers (ADMM) has emerged as a scalable approach to long-horizon motion planning. However, existing frameworks typically decompose the problem into parallel subproblems based on a predefined fixed structure. Such structural rigidity often causes optimization stagnation in highly constrained regions, where a few lagging subproblems delay global convergence. A natural remedy is to adaptively re-split these stagnating segments online. Yet, deciding when, where, and how to split exceeds the capability of rule-based heuristics. To this end, we propose ATRS, a novel framework that embeds a shared Deep Reinforcement Learning policy into the parallel ADMM loop. We formulate this adaptive adjustment as a Multi-Agent Shared-Policy Markov Decision Process, where all trajectory segments act as homogeneous agents and share a unified neural policy network. This parameter-sharing architecture endows the system with size invariance, enabling it to handle dynamically changing segment counts during re-splitting and generalize to arbitrary trajectory lengths. Furthermore, our formulation inherently supports zero-shot generalization to unseen environments, as our network relies solely on the internal states of the numerical solver rather than on the geometric features of the environment. To ensure solver stability, a Confidence-Based Election mechanism selects only the most stagnating segment for re-splitting at each step. Extensive simulations demonstrate that ATRS accelerates convergence, reducing the number of iterations by up to 26.0

CFD-DEM integrated with advanced particle shape models has emerged as a powerful approach for resolving non-spherical particle-fluid flows, but its practical application still suffers from degeneration in accuracy and stability when dealing with wide grid-to-particle size ratios (GSR). To tackle this issue, we developed a dual support domain-based CFD-DEM (DSD-CFD-DEM) framework for non-spherical particle-fluid flows. In this framework, two sets of support domains, one centered on non-spherical particles and the other centered on fluid cells, were established to perform information reconstruction local to particles and quantity smoothing within cells, thereby enabling robust two-way coupling and accurate prediction of particle dynamics under wide GSR. DSD-CFD-DEM was validated through comprehensive benchmark studies: (i) sedimentation of six non-spherical particles under various mesh and Reynolds number conditions, showing agreement with particle-resolved direct numerical simulations; (ii) non-spherical gas-solid flows in a spouted bed, where predicted hydrodynamics and pressure drops match experimental data. The computational cost of DSD-CFD-DEM reaches 1.46 times that of conventional CFD-DEM simulations across the tested range of grid numbers.

Particle image-based fluid measurement techniques are widely used to study complex flows in nature and industrial processes. Despite that particle tracking velocimetry (PTV) has shown potential in various experimental applications for quantitatively capturing unsteady flow characteristics, estimating fluid motion with long displacement and high particle density remains challenging. We propose an artificial-intelligence-enhanced PTV framework to track particle trajectories from consecutive images. The proposed framework, called GOTrack+ (a learning framework with graph optimal transport for particle tracking velocimetry), contains three components: a convolutional neural network-based particle detector for particle recognition and sub-pixel coordinate localization; a graph neural network-based initial displacement predictor for fluid motion estimation; and a graph-based optimal transport particle tracker for continuous particle trajectory linking. Each component of GOTrack+ can be extracted and used independently, not only to enhance classical PTV algorithms but also as a simple, fast, accurate, and robust alternative to traditional PTV programs. Comprehensive evaluations, including numerical simulations and real-world experiments, have shown that GOTrack+ achieves state-of-the-art performance compared to recent PTV approaches. All the codes are available at https://github.com/wuwuwuas/GOTrack.git.

Collective behavior arises when locally interacting units produce coordinated global organization, from synchronization in dynamical systems to task-relevant information flow on graphs. The central challenge is not only to explain how collective behavior emerges, but to design local interaction rules that can produce desired global organization and generalize across graphs, dynamics and tasks.To address this challenge, we introduce the Swarm-Inspired Emergent Synchronizer (SIES), a graph-dynamical framework that learns generalizable local-interaction laws for controllable collective organization. Each node is an agent-like dynamical unit with a state and task cue, and signed source-target-conditioned attention acts as an adaptive coupling term inside an explicit evolution model. Therefore, SIES combines an explicit dynamical engine with local agent intelligence, similar to biological swarms. For synchronization control, SIES learns a generalizable coupling operator that produces prescribed synchronization patterns for CDSs across untrained network scales, target phase relations, and intrinsic node dynamics without retraining. The learned operator also reaches gait-related modes faster than three oscillator baselines and generalizes synchronization-driven locomotion to simulated multi-legged robots of different scales and a physical hexapod after leg disablement. For graph representation learning, SIES applies the same signed interaction principle to message passing and achieves the highest performance among the compared methods on heterophilous node-classification benchmarks. Together, these results position SIES as a generalizable and learnable graph-dynamical interaction framework with promise for synchronization control, adaptive robot coordination, and heterophilous graph representation learning.

Accurate relative localization is critical for multi-robot cooperation. In robot groups, measurements from different robots arrive asynchronously and with clock time-offsets. Although Continuous-Time (CT) formulations have proved effective for handling asynchronous measurements in single-robot SLAM and calibration, extending CT methods to multi-robot settings faces great challenges in achieving high-accuracy, low-latency, and high-frequency performance. In particular, existing CT methods suffer from the inherent query-time delay of unclamped B-splines and high optimization latency. This paper proposes CT-RIO, a novel Continuous-Time Relative-Inertial Odometry framework. We adopt Clamped Non-Uniform B-splines (C-NUBS) to represent states, eliminating the query-time delay. We further augment C-NUBS with closed-form extension and shrinkage operations that preserve the spline shape, making it suitable for online estimation and enabling flexible knot management. This flexibility leads to the concept of a knot-keyknot strategy, which supports spline extension at high frequency while retaining sparse keyknots for adaptive relative motion modeling. We then formulate a reference-centric sliding-window relative localization problem that operates purely on relative kinematics and inter-robot constraints. To enable low-latency and high-frequency estimation, we decompose the tightly coupled optimization into robot-wise subproblems and solve them in parallel using asynchronous block coordinate descent. Extensive experiments show that CT-RIO converges from time-offsets as large as 264 ms to sub-millisecond within 3 s, and achieves RMSEs of 0.046 m and 1.8 degree. It consistently outperforms evaluated published methods, with improvements of up to 60

Autonomous flight in unknown cluttered environments is hindered by the computation-quality-memory trilemma of onboard trajectory generation. In this paper, we propose an efficient end-to-end local planner via imitation learning. A lightweight offline-primitive-based dataset collection framework is designed to produce safe and high-quality trajectory primitives in non-convex environments. A compact neural network directly maps sensory inputs to polynomial coefficients that inherently encode higher-order dynamical information. The learned policy generates smooth, empirically collision-free and dynamically feasible trajectories in real time without back-end solving. It achieves ultra-fast computation (below 1ms on a standard desktop and average 3.68ms during onboard flight), while maintaining low onboard memory requirements (less than 1.5MiB). Extensive simulation benchmarks demonstrate superiority in both planning latency and target-reaching progress quality. Zero-shot deployment in real-world experiments further validates the robust sim-to-real transfer capability of the proposed method.

Due to the limited online computational resources and the inherent probability of hardware and software failures of real-world robots, large-scale formation planning faces two common challenges: computational intractability and agent failures. Based on the theory of sparse graphs and the maximum clique, {we achieve a resilient and efficient formation planning (mathbf{RE}-mathbf{Formation}) to address these issues.} To improve the computational efficiency of trajectory planning while ensuring flexible formation maneuvers, we introduce sparse graphs to describe connection relationships and present a sparse graph construction method with closed-form solutions. The sparse graphs ensure the {}{underline{G}}lobal {}{underline{R}}igidity for uniquely corresponding to a geometric shape and {}{underline{P}}reserve the main {}{underline{F}}eatures of complete graphs, {}{denoted as the mathbf{GRPF} sparse graph}. To prevent the impact of abnormal agents, the problem of eliminating abnormal agents is transformed into an outlier rejection problem that can be solved by computing the maximum clique. {}{We approximate the maximum clique by periodically triggering the calculation of the maximum k-core to meet the real-time computational demands of large-scale swarms.} We validate the performance through real-world experiments and implement formation planning with 100 drones in simulation. Benchmark comparisons and ablation experiments demonstrate the effectiveness of

This study develops a $\boldsymbol{q}$ -axis current loop-free positioning technique for permanent magnet synchronous motors (PMSMs), integrating a specially structured proportional-integral (PI) controller, a model-independent cascade observer, and a fourth-order disturbance observer (FO-DOB). The method effectively handles PMSM parameter and load uncertainties and remains practical for industrial use. The core contribution lies in applying pole-zero cancellation (PZC) to both estimation and control loops: 1) the tailored gains of the observers induce PZC, yielding first-order estimation error dynamics independent of the PMSM model; 2) the filtered position output by the observer drives a FO-DOB to achieve triply-damped disturbance estimation; and 3) the combined observer and FO-DOB construct the PI-type controller, whose gain structure assigns a critically damped closed-loop response through PZC. Experimental validation is conducted on a 1-kW PMSM dynamometer.

Transient simulations offer the advantage of capturing time-dependent flow behavior, making them more suitable than steady simulations for modeling complex phenomena such as turbulence, vibration, cavitation, and noise. While traditional CFD methods are more suitable for handling steady simulations, they are less effective for transient simulations due to limited parallel processing capabilities, leading to high computational costs. As a result, the lattice Boltzmann method (LBM) is employed in this study, which is a more efficient approach for transient simulation owing to its efficient handling of complex geometries, programming simplicity, and strong parallel scalability. In order to enhance the stability of LBM in the numerical simulation of high Reynolds number flow fields, the multiple relaxation time (MRT) collision model and the Smagorinsky-Lilly large eddy simulation (LES) turbulence model are utilized. To address the high dissipation near the wall in the Lilly model, the van Driest damping function is incorporated, improving the accuracy of the LES model in boundary regions. Additionally, to minimize memory consumption and reduce computation time without sacrificing accuracy, wall functions and local grid refinement techniques are applied, reducing the overall number of computational grids required. An experimental platform was established to measure the flow characteristics of control valves, and the accuracy of the proposed method was validated by comparing the flow coefficient Cv at various valve openings. Finally, the effects of local grid refinement and wall functions on simulation accuracy were compared, demonstrating that these techniques significantly improve the precision of transient simulations.

Triphibious robots capable of multi-domain motion and cross-domain transitions are promising to handle complex tasks across diverse environments. However, existing designs primarily focus on dual-mode platforms, and some designs suffer from high mechanical complexity or low propulsion efficiency, which limits their application. In this paper, we propose a novel triphibious robot capable of aerial, terrestrial, and aquatic motion, by a minimalist design combining a quadcopter structure with two passive wheels, without extra actuators. To address inefficiency of ground-support motion (moving on land/seabed) for quadcopter based designs, we introduce an eccentric Center of Gravity (CoG) design that inherently aligns thrust with motion, enhancing efficiency without specialized mechanical transformation designs. Furthermore, to address the drastic differences in motion control caused by different fluids (air and water), we develop a unified propulsion system based on Field-Oriented Control (FOC). This method resolves torque matching issues and enables precise, rapid bidirectional thrust across different mediums. Grounded in the perspective of living condition and ground support, we analyse the robot's dynamics and propose a Hybrid Nonlinear Model Predictive Control (HNMPC)-PID control system to ensure stable multi-domain motion and seamless transitions. Experimental results validate the robot's multi-domain motion and cross-mode transition capability, along with the efficiency and adaptability of the proposed propulsion system.