Search Results (21,203)

Legged robots exhibit high mobility on uneven terrain but face challenges in stability, complex control systems, and energy efficiency. This study proposes a leg mechanism that significantly alters its stiffness by inducing jamming in a chainmail structure through only gravity-induced compression. To evaluate the fundamental characteristics of the proposed mechanism, experiments were conducted to identify the jamming point and to assess stiffness in the jammed state. The results confirmed that the force required to trigger jamming increases proportionally with the mass applied from above, which demonstrates properties similar to friction between solid materials. Furthermore, the stiffness in the jammed state is strongly correlated with the contact points within the structure. These results prove the effectiveness of the proposed passive leg mechanism for stiffness switching. In a case study assuming landing on uneven terrain, the mechanism could be fixed in any orientation based on the designed compressive force. Full article

This study explores how integrating digital technologies into STEM-based physics instruction can transform learning outcomes for adult learners in Greek Second Chance Schools, which provide educational opportunities for adults over 18 who have not completed compulsory education. In a comparative design, participants were divided into two groups: the experimental group experienced an innovative STEM approach, combining educational robotics, mobile sensing, and 3D printing within the Biological Sciences Curriculum Study (BSCS) 5E Instructional Model; the control group received enriched lecture-based instruction. Learning gains were measured using a rigorously developed, psychometrically validated multiple-choice physics test administered before and after the intervention. Results reveal that adults exposed to technology-enhanced STEM lessons achieved statistically significant improvements, outperforming their peers in the lecture-based group, who showed no measurable progress. Notably, these gains were consistent across gender and age. The findings highlight the transformative potential of digital technologies and learner-centered STEM pedagogies in alternative education settings, offering new directions for adult education and lifelong learning. Full article

To address the problem of insufficient detection accuracy, high false negative rate of small targets, and large positioning errors of ships in complex marine environments and foggy conditions, an improved DBL-YOLO method based on YOLOv11 is proposed. This method customizes and optimizes modules according to the characteristics of foggy scenes—the C3k2-MDSC module is designed to efficiently extract and fuse multi-scale spatial features, and a dynamic weight allocation mechanism is adopted to balance the contributions of features at different scales in the foggy and blurred environment; a lightweight BiFPN structure is introduced to enhance the efficiency of cross-scale feature transmission and solve the problem of feature attenuation in foggy conditions; a novel fusion of the Deformable-LKA attention mechanism is innovated, which combines a large receptive field and spatial adaptive adjustment capabilities to focus on the key contour features of blurred ships in foggy conditions; an Inner-SIoU regression loss function is proposed, which optimizes the positioning accuracy of dense and small targets through an auxiliary bounding box dynamic scaling strategy. Experimental results show that in foggy scenes, the recall rate is increased by 3.4%, the F1 score is increased by 1%, and mAP@0.5 and mAP@0.5:0.95 are increased by 1.4% and 3.1%, respectively. The final average precision reaches 98.6%, demonstrating excellent detection accuracy and robustness. Full article

Upper limb rehabilitation exoskeletons form a spatial closed kinematic chain with the human arm, where inevitable joint-center and axis misalignment can generate hyperstatic interaction forces and torques. Passive degrees of freedom (DOF) are widely introduced to improve kinematic compatibility, yet different compatible configurations may exhibit distinct wearable performance. This study experimentally compares three compatible four-degree-of-freedom exoskeleton configurations derived from the synthesis of Li et al. using a single reconfigurable rehabilitation robot. The platform is assembled into each configuration through modular passive units and instrumented with two six-axis force–torque sensors at the upper-arm and forearm interfaces. Interaction forces and torques are measured in passive training mode during eating and combing trajectories. For each configuration, tests are performed with passive joints released and with passive joints locked to quantify the effect of passive motion accommodation. Directional and resultant metrics are computed using mean and peak values over movement cycles. Results show that releasing passive joints consistently reduces interaction loading, and Category 2 achieves the lowest forces and torques with the strongest peak suppression, indicating the best practical compatibility. Full article

Wearable robots can improve human walking economy; however, their effectiveness depends on user adaptation to assistance. This study introduces a framework for real-time estimation of user adaptation that relies only on wearable sensor data during operation. Metabolic measurements were used solely to establish the ground truth adaptation curves for model training and validation but are not required for real-time inference. Five healthy adults completed six days of treadmill walking while wearing a soft hip exosuit that provided hip extension assistance. Thigh-mounted inertial measurement units recorded step timing and hip-angle trajectories, from which three variability-based features (step-frequency variability, maximum hip-flexion variability, and maximum hip-extension variability) were extracted. A Long Short-Term Memory (LSTM) model used these gait-variability inputs to estimate each user’s adaptation level relative to a metabolic cost benchmark obtained from respiratory gas analysis. Across sessions, the metabolic cost decreased by 9.0 ± 5.6% from Day 1 to Day 6 (p < 0.01) with a mean time constant of 202 ± 78 min, In contrast, the variability in step frequency, maximum hip flexion, and maximum hip extension decreased by 66.4 ± 6.8%, 37.9 ± 24.2%, and 42.8 ± 10.6%, respectively, indicating that these reductions were users’ progressive adaptation to the exosuit’s assistance. Under leave-one-subject-out (LOSO) evaluation across five participants, 59.2% of the model predictions fell within ±10 percentage points of the metabolic cost–based adaptation curve. These results suggest that simple kinematic variability measured with wearable sensors can track user adaptation and support practical approaches to real-time monitoring. Such capability can facilitate adaptive control and training protocols that personalize exosuit assistance. Full article

Underwater structural surface defect detection—such as identifying cavities and spalling—faces significant challenges due to complex environments, scarce annotated data, and the reliance of modern detectors on large-scale datasets. While current approaches often combine large-data training with fine-tuning or image enhancement, they still require extensive underwater samples and are typically too computationally heavy for resource-constrained robotic platforms. To address these issues, we introduce a defect detection model based on feature distillation, which achieves high detection accuracy with limited samples. We tackle three key challenges: enhancing sample diversity under data scarcity, selecting and training a baseline model that balances accuracy and efficiency, and improving lightweight model performance using augmented samples under computational constraints. By integrating a feature distillation mechanism with a sample augmentation strategy, we develop a compact detection strategy and framework that delivers notable performance gains in limited data, offering a practical and efficient solution for real-world underwater inspection. Full article

Due to the ability to perceive fine-grained 3D scenes and recognize objects of arbitrary shapes, 3D occupancy prediction plays a crucial role in vision-centric autonomous driving and robotics. However, most existing methods rely on voxel-based methods, which inevitably demand a large amount of memory and computing resources. To address this challenge and facilitate more efficient 3D occupancy prediction, we propose HBEVOcc, a Bird’s-Eye-View based method for 3D scene representation with a novel height-aware deformable attention module, which can effectively leverage latent height information within BEV framework to compensate for lack of height dimension, significantly reducing computing resource consumption while enhancing the performance. Specifically, our method first extracts multi-camera image features and lifts these 2D features into 3D BEV occupancy features via explicit and implicit view transformations. The BEV features are then further processed by a BEV feature extraction network and height-aware deformable attention module, with the final 3D occupancy prediction results obtained through a prediction head. To further enhance voxel supervision along the height axis, we introduce a height-aware voxel loss with adaptive vertical weighting. Extensive experiments on the Occ3D-nuScenes and OpenOcc dataset demonstrate that HBEVOcc can achieve state-of-the-art results in terms of both mIoU and RayIoU metrics with less training memory (even when trained on 2080Ti). Full article

This paper presents the development and experimental evaluation of the Athena parallel robot, a novel system designed for robot-assisted pancreatic surgery. The development of the experimental model based on the kinematic scheme, including the command and control system (hardware and software), the calibration procedure, and the performance measurements of the experimental model based on finite element analyses of the 3D model, are also detailed in this paper. Based on these finite element analyses, a region of the robot that introduces clearance during the operation of the experimental model is found. The paper also presents the methodology used for mapping the robot’s workspace with an optical system, which enabled improvements to ensure coverage of the entire pancreas area. The results obtained before and after the mechanical improvements are presented, demonstrating a reduction in clearance by up to 4.1 times following part replacement, as well as a workspace extension that enables the active instrument to reach the entire pancreatic region. Full article

To address trajectory tracking of cable-driven continuum robots (CDCRs) in the presence of obstacles, this paper proposes an integrated control framework that combines online trajectory replanning, obstacle avoidance, and tracking control. The control system consists of two modules. The first is a trajectory replanning controller developed on an improved model predictive control (IMPC) framework. The second is a trajectory-tracking controller that integrates an adaptive disturbance observer with a fast non-singular terminal sliding mode control (ADO-FNTSMC) strategy. The IMPC trajectory replanning controller updates the trajectory of the CDCRs to avoid collisions with obstacles. In the ADO-FNTSMC strategy, the adaptive disturbance observer (ADO) compensates for uncertain dynamic factors, including parametric uncertainties, unmodeled dynamics, and external disturbances, thereby enhancing the system’s robustness and improving trajectory tracking accuracy. Meanwhile, the fast non-singular terminal sliding mode control (FNTSMC) guarantees fast, stable, and accurate trajectory tracking. The average tracking errors for IMPC-ADO-FNTSMC, MPC-FNTSMC, and MPC-SMC are 1.185 cm, 1.540 cm, and 1.855 cm, with corresponding standard deviations of 0.035 cm, 0.057 cm, and 0.078 cm in the experimental results. Compared with MPC-FNTSMC and MPC-SMC, the IMPC-ADO-FNTSMC controller reduces average tracking errors by 29.96% and 56.54%. Simulation and experimental results demonstrate that the designed two-module controller (IMPC-ADO-FNTSMC) achieves fast, stable, and accurate trajectory tracking in the presence of obstacles and uncertain dynamic conditions. Full article

Advances in artificial intelligence (AI), soft robotics, and miniaturized imaging technologies have accelerated the development of endorobotic platforms that aim to enhance detection accuracy and improve patient experience. In this narrative review, we synthesize evidence on AI-assisted detection and characterization systems (CADe/CADx), robotic locomotion mechanisms, adhesion strategies, imaging modalities, and material and power constraints relating to next-generation CRC screening technologies. Reported performance metrics are interpreted within their original methodological contexts, acknowledging the heterogeneity of datasets, limited representation of diverse populations, underreporting of negative findings, and scarcity of large, real-world comparative trials. We introduce a conceptual translational framework that links engineering design principles with validation needs across in silico, in vitro, preclinical, and clinical stages, and we outline safety considerations, workflow integration challenges, and sterility requirements that influence real-world deployability. Regulatory alignment is discussed using the U.S. FDA Total Product Life Cycle (TPLC) and Good Machine Learning Practice (GMLP) frameworks to highlight expectations for data quality, model robustness, device–software interoperability, and post-market monitoring. Collectively, the evidence demonstrates promising technological innovation but also highlights substantial gaps that must be addressed before AI-enabled endorobotic systems can be safely and effectively integrated into routine CRC screening. Continued interdisciplinary work, supported by rigorous validation and transparent reporting, will be essential to advance these technologies toward meaningful clinical impact. Full article

With the rapid development of multi-modal large models, the Visual–Language–Action (VLA) model has gradually become a new paradigm for autonomous robot operations. The VLA model encodes experimental images and text instructions separately using an image encoder and a text encoder. The encoded multi-modal vector information is then fed into a large language model (LLM) to generate the next action. While they inherit the generalization capabilities of large language models, VLA models often struggle to ensure accuracy and reliability in complex scenes. Some studies have attempted to improve VLA performance by enhancing the fine-tuning process or introducing staged operations; however, these improvements often overlook the stable extraction of important visual features, which are crucial for VLA models. In typical VLA tasks, the instruction text inherently contains semantic information related to image elements. Research has shown that leveraging text supervision for visual feature extraction can enhance feature quality. In this paper, we propose a semantically supervised visual encoder called SeDINO (Semantically Supervised DINO), which efficiently fuses DINO’s element localization capabilities with CLIP’s semantic information. We further employ an MLP (Multi-Layer Perceptron) network to align the semantic vectors output by the CLIP text encoder with the image feature vectors derived from DINO, fully leveraging DINO’s element localization and CLIP’s semantic interaction capabilities. We validate SeDINO on six mainstream image datasets, and it demonstrates superior segmentation performance compared to current leading models. Additionally, we incorporate the proposed SeDINO into the VLA framework, using OpenVLA-7B and DINOv2-base as backbone models, and evaluate it on the LIBERO dataset and real-world scenarios. Full article

This paper presents the design and implementation of a novel autonomous underwater robot with fully vectored propulsion based on model predictive control (MPC) to rapidly respond to the position and attitude required for autonomous operation. Specifically, the mechatronic design of the eight vector-distributed thruster layout for the robot’s fully vectored propulsion is detailed, and the software architecture based on the robot operating system (ROS) is constructed. Then, the corresponding dynamics model is established by adopting the Fossen approach for the prediction and optimization of the control process. To achieve autonomous control, an MPC-based controller is designed and implemented to calculate the control input for the specified control objective. Finally, way-point tracking and trajectory-tracking experiments are carried out in an indoor tank equipped with a motion-capture system to validate the feasibility and effectiveness of the robot’s design and control framework. In addition, the robustness of the robot system is verified by artificially perturbing the robot in the hovering state. Full article

Background: Artificial intelligence (AI) has emerged as a transformative tool in medicine, leveraging rapid analysis of large datasets to accelerate diagnosis, enhance clinical decision-making, and improve clinical workflows. This is highly relevant in stroke care given the time-sensitive nature of the disease process. This review evaluates the current landscape of evidence-based medicine utilizing AI in stroke, with emphasis on its use in phases of clinical care across the stroke continuum, including pre-hospital, acute, and recovery phases. This offers a comprehensive understanding of the current state of AI in both practice and literature. Methods: A review of major databases was conducted, identifying peer-reviewed literature evaluating the use of AI and its level of evidence across the stroke continuum. Given the heterogeneity of study designs, interventions, and outcome metrics spanning multiple disciplines, findings were synthesized narratively. Results: Across all phases of care, there remain no randomized controlled trials (RCTs) evaluating patient-level outcome data using AI (Level A). In the pre-hospital phase of care, AI has been used to identify stroke symptoms and assist EMS routing/training but presently remains limited to research. AI is most studied in the acute phase of care, representing the only phase to achieve commercial application in imaging detection and telestroke assistance, supported by non-randomized evidence (Level B-NR). In the recovery phase, AI may enhance wearable technologies, tele-rehabilitation, and robotics/brain–computer interfaces, with early RCTs (Level B-R) supporting the latter two, representing the strongest evidence for AI in stroke care to date. Conclusions: Despite the potential for AI to transform all phases of care across the stroke continuum, major challenges remain, including transparency, generalizability, equity, and the need for externally validated clinical studies. Full article

Patient engagement and mental workload (MWL) are often overlooked when optimising robotic-assisted rehabilitation, despite their potential impact on its effectiveness. This study aims to propose a multimodal approach to assess MWL and engagement, using electrophysiological signals and questionnaires, to explore their modulation across different assistance modalities and engaging strategies. Thirty healthy subjects were enrolled and performed repetitive upper-limb movements with a robotic device under three assistance modalities (active, passive, semi-assisted) while listening to a 1 Hz auditory stimulus (metronome or music). Electroencephalography, Electrocardiogram, the NASA Task Load Index, and the Short Stress State Questionnaire were used to assess objective and perceived MWL and engagement. Engagement increased significantly in the music condition, whereas MWL showed no significant change. The passive modality was perceived as significantly less demanding and less engaging compared to active and semi-assisted conditions. Although EEG objective indicators did not vary across modalities, the ECG objective metric was modulated significantly in agreement with the subjective measures. Overall, the auditory stimulus significantly influenced engagement, and assistance levels affected both perceived mental demand and engagement. The proposed multimodal approach is sensitive to both engagement and MWL constructs, highlighting the potential for adaptive rehabilitation systems designed to maintain engagement, prevent overload or monotony, and ultimately support better functional outcomes over the long term of robotic training. Full article

Designing provably safe controllers for high-dimensional nonlinear systems with formal guarantees represents a fundamental challenge in control theory. While control barrier functions (CBFs) provide safety certificates through forward invariance, manually crafting these barriers for complex systems becomes intractable. Neural network approximation offers expressiveness but traditionally lacks formal guarantees on approximation error and Lipschitz continuity essential for safety-critical applications. This work establishes rigorous theoretical foundations for learned barrier functions through explicit probabilistic bounds relating neural approximation error to safety failure probability. The framework integrates Lipschitz-constrained neural networks trained via PAC learning within multi-objective model predictive control. Three principal results emerge: a probabilistic forward invariance theorem establishing $\mathbb{P}\left(\mathrm{violation}\right)\le T{\delta }_{\mathrm{local}}+exp(-h_{min}^2/\left(2L^{2}T{\sigma }^2\right))$, explicitly connecting network parameters to failure probability; sample complexity analysis proving $\mathcal{O}\left(N^{-1/4}\right)$ safe set expansion; and computational complexity bounds of $\mathcal{O}\left(H^3{m}^3\right)$ enabling 50 Hz real-time control. An experimental validation across 648,000 time steps demonstrates a 99.8% success rate with zero violations, a measured approximation error of $\sigma =0.047$ m, a matching theoretical bound of $\sigma \le 0.05$ m, and a 16.2 ms average solution time. The framework achieves a 52% conservatism reduction compared to manual barriers and a 21% improvement in multi-objective Pareto hypervolume while maintaining formal safety guarantees. Full article