Hybrid motion planning with dynamics awareness for aerial-ground robots in industrial inspection and maintenance

ABSTRACT Industry 5.0 emphasises the integration of human intelligence with advanced automation, promoting collaboration between humans and machines in manufacturing environments. This paper presents an Extended Reality (XR)-driven Digital Twin (DT) framework for intelligent robotic laser cleaning, designed to enhance human-robot collaboration through immersive and intuitive interfaces. The system integrates a YOLOv11x-OBB-based defect-detection module, adaptive path-planning algorithms, and gesture-based robot control within Mixed Reality (MR) and Virtual Reality (VR) environments. The DT, implemented in Unity and synchronised with a UR10e collaborative robot via ROS2 (Robot Operating System 2)-RTDE (Real-Time Data Exchange) communication, enables real-time monitoring, trajectory validation, and bidirectional data exchange between physical and virtual systems. Experimental results demonstrate accurate defect localisation (mAP@0.5 = 0.995, mAP@0.5:0.95 = 0.977), responsive hand-gesture interaction (438 ms average latency), and stable DT synchronisation. Furthermore, the proposed defect-targeted path-planning strategy reduced laser-cleaning time by 86% compared with full-surface cleaning, highlighting its potential for industrial productivity and sustainability. The results confirm that combining XR, DT, and AI-based perception establishes an effective human-centric framework for intelligent robotic surface processing, supporting future deployment in Industry 5.0 manufacturing systems.

Autonomous hydraulic excavators are widely used in construction, mining, and material-handling operations, offering improved efficiency and safety. Precise trajectory tracking is essential for such systems. However, inherent nonlinearities and significant time delays in the hydraulic actuators hinder accurate control for autonomous operations. To address these challenges, a nonlinear model predictive control (NMPC) algorithm is proposed. Specifically, a Hammerstein–Wiener structure is employed to model the nonlinear hydraulic system, with parameters identified from experimental data. Based on this model, an NMPC trajectory-tracking algorithm is developed, which accounts for actuator and control input constraints. To mitigate the intrinsic 0.5-second response delay of the hydraulic system, a predictive delay compensation strategy is introduced, whereby predicted joint states over the next 0.5 seconds serve as real-time control references. Simulation results demonstrate that the proposed controller substantially outperforms proportional–integral–derivative (PID) and fuzzy PID methods, maintaining the bucket-end error within 20 cm. Field experiments on an autonomous excavator implemented under the robot operating system (ROS) framework confirm that the maximum trajectory-tracking error remains within 50 cm, validating the effectiveness and robustness of the proposed NMPC approach under real-world operating conditions.

Switchable adhesion underpins emerging technologies in robotics, microelectronics, and biomedical engineering. However, achieving switchable surface adhesion that can adapt to substrates with varying material compositions and surface roughness, while simultaneously enabling real-time and wireless monitoring of adhesion strength, poses a substantial challenge. Here, we present a eutectogel-based system that integrates electrothermally switchable adhesion with wireless sensing capability for in situ monitoring of adhesion forces. The switching mechanism is systematically elucidated through a combination of mechanical analysis and molecular-level characterization. The integration of machine-learning assisted adhesion sensing with dynamic gripping and locomotion enables safer and smarter robotic operation in adhesion joints, smart grippers and climbing robots. Demonstrations in adhesion-aware sensing, robotic grasping, and wall climbing validate the system's practical utility, establishing a pathway toward next-generation intelligent adhesive interfaces that are both adaptive and self-perceptive.

Robotic mitral surgery is one of the most common robotic cardiac procedures being performed, which presents many benefits including smaller incisions, less postoperative pain, quicker recovery, decreased blood transfusion requirements for the patient, better exposure and better visualization of the mitral valve, and increased operative dexterity for the surgeon. As redo robotic mitral operations have increased in number, the use of atrial and ventricular sumps have also increased, especially during cases done under fibrillation. Our practical and easy-to-implement technique that consists of transpericardial atrial and ventricular vent placement during robotic mitral procedures may prevent dislodgement of the vents and may be beneficial to avoid important complications such as air embolisms and left ventricular distension resulting in subendocardial ischaemia.

Single-port (SP) robotic operations are the next frontier of minimally invasive surgery (MIS), promising to build upon multi-port robotic and laparoscopic techniques. This study aimed to assess the safety and feasibility of SP robotic cholecystectomy, hiatal hernia repair with fundoplication, and gastrectomy. With IDE and IRB approval, 12 patients underwent cholecystectomy, hiatal hernia repair with fundoplication and gastrectomy using the single-port surgical platform. All operations were undertaken through a 1.2 cm incision at the umbilicus and the patients who underwent a gastrectomy had an additional 1.2 cm incision for the stapler. Patient perioperative data were collected prospectively. Data is reported as mean ± SD. Five patients underwent SP cholecystectomy for cholelithiasis or biliary dyskinesia. Console time was 172 ± 116.7 min, EBL was minimal. There were no conversions or intraoperative complications. There were no postoperative complications or mortalities. All patients were dischargedthesame day. Patient #5 was readmitted within 30 days for anaberrantside branch bile duct leak, treated with a percutaneous drain and a stent. Five patients underwent SP hiatal hernia repair with Nissen or Toupet fundoplication for GERD.Console time was 246 ± 16.0 min, EBL was minimal, there were no conversions or intraoperative complications. There were no postoperative complications, mortalities, or readmissions. All patients were discharged the following day. In the second phase, two patients underwent SP gastrectomy, one partial gastrectomy for GIST and one near total gastrectomy with D2 lymphadenectomy for gastric adenocarcinoma,without conversion or perioperative complications. Both had negative marginswithadequate lymph node harvest and were discharged on postoperative days 2 and 4 respectively without readmissions. SP cholecystectomy, gastrectomy and hiatal hernia repair with fundoplication are safe and efficacious. Implementation of SP robotic surgery using the da Vinci SP platform lays the groundwork for expanding single-port robotic techniques into advanced foregut and hepatopancreatobiliary (HPB) operations.

Physics-informed ensemble learning for hierarchical fault diagnosis in quadruped robots

Unified T-S fuzzy-relaxed control barrier function framework for unicycle Mobile robot safe trajectory tracking.

CBAM-ST-GCN: An enhanced DRL-based end-to-end visual navigation framework for mobile robot.

Impact of preventive measures on welding fume exposure among welders, robot operators, and bystanders: Air and biomonitoring.

Background: The growing demand for versatile laboratory automation is exemplified in the context of liquid biopsy, where multi-analyte approaches are increasingly recognised for their potential to enhance diagnostic sensitivity in oncology. However, current practice often necessitates the use of dedicated instruments and workflows for the extraction of each analyte, posing financial and logistical barriers for automated multi-analyte liquid biopsy. Methods: Here, we present Robotic Centrifugal Microfluidics (RoCM), an all-in-one platform that combines the versatility of centrifugal microfluidics and operational flexibility of robotic liquid handling. This combination enables the automation of complex micro- and macrofluidic protocols, realised through the use of (1). exchangeable microfluidic cartridges and (2). programmable robotic operations such as in-rotation liquid supply, magnetic bead manipulation, or microfluidic valving. In-rotation robotic liquid manipulation maintains fluid control under centrifugal forces and reduces the cartridge footprint associated with pre-loaded liquid reservoirs. Platform applicability was demonstrated using two exemplary liquid biopsy workflows: the extraction of cell-free DNA (cfDNA) from blood plasma using RoCM-cfDNA slices and the extraction of extracellular vesicles (EVs) from blood plasma using RoCM-EV slices. Results: In a pilot study with patient samples from different cancer entities, the RoCM-cfDNA slices yielded comparable variant allele frequencies to a commercial bead-based instrument, while the RoCM-EV slices achieved a recovery of a greater diversity of EV subpopulations than semi-automated size-exclusion chromatography. Conclusions: By simply exchanging cartridges, RoCM enables the extraction of diverse analytes within a single automated system. Its application can be extended to further analytes, such as circulating tumour cells (CTCs), or to applications beyond liquid biopsies, where versatile micro- and macrofluidic protocols benefit from implementation in a single automation instrument.

The demographic shift towards an aging population necessitates innovative solutions for care and mobility support. While commercial quadruped robots like the Unitree Go1 offer dynamic stability, their native following modes often lack the safety margins and predictability required, and they do not consistently follow the user, at times deviating and navigating independently. This paper presents a robust, vision-based, person-following algorithm designed to address these limitations. Utilizing a ZED 2 stereo camera and Robot Operating System (ROS), the system employs a finite state machine to ensure deterministic target tracking. A velocity control strategy partitions the robot’s motion into distinct stability, proportional, and braking zones based on depth data to ensure fluid interaction. The framework was validated on a Unitree Go1 quadruped platform in an outdoor environment involving 90-degree turns to evaluate tracking robustness. By operating in a headless mode, the system achieved a mean processing latency of ms. Experimental results demonstrated consistent operational stability, 0.0% intrusion into the intimate safety zone, and effective velocity synchronization between and m/s. While this study establishes a robust technical baseline using healthy subjects, it serves as a preliminary development platform; further iterative testing with elderly users in clinical settings is required to move toward deployment. Beyond the evaluated trials, the framework maintained reliable functional performance across various care facility workshops, successfully following the target in all deployment scenarios. These findings establish a stable technical foundation for the future development of robotic walking partners.

Emulating human skin’s ability to perceive temperature and identify material through thermotactile perception is critical for human-machine interaction, robotics operation, and prosthetic sensory feedback systems. However, conventional artificial thermal sensors are largely limited to temperature measurement and cannot replicate the thermotactile-mediated material recognition capabilities inherent in biological systems. Here, we present a bioinspired ionic thermoreceptor with anisotropic thermal response characteristics, enabling high-fidelity material recognition and accurate temperature monitoring. The device incorporates spatially specialized sensing elements that encode thermal signals into modality-specific temporal response patterns, allowing the extraction of the thermal contact coefficient for material discrimination and the absolute temperature for precise thermal monitoring. It achieves high-accuracy material recognition (98.9%) and substantial temperature resolution (0.81 millikelvins). Furthermore, robust covalent bonding between functional layers ensures mechanical durability, signal stability, and environmental resilience. This work establishes a physically grounded and energy-autonomous thermotactile sensing platform for safe, intuitive, and intelligent human-machine interaction.

Accurate stockpile volume estimation is crucial for material quantification and inventory management in civil engineering, directly affecting cost assessment and on-site decision-making. Traditional manual methods suffer from subjective bias and limitations in handling irregular geometries, resulting in reduced accuracy and efficiency. This study presents a Light Detection and Ranging (LiDAR)-based workflow integrated with Robot Operating System (ROS) for point cloud processing, enabling accurate volume estimation of irregular stockpiles. The core innovation lies in the integration of multi-station scanning, point cloud registration, boundary extraction, layered slicing, and numerical integration using the trapezoidal rule, thereby enabling geometrically precise volume estimation of irregular stockpiles. The proposed system was validated through three experimental scenarios: (1) controlled experiments, showing strong agreement with theoretical volumes; (2) verification experiments, demonstrating high stability and consistency; and (3) field experiments, yielding a volume of 124.93 m3 compared to 130–135 m3 obtained by manual measurement. The results indicate that the proposed approach reduces processing time by over 80% while significantly decreasing labor requirements and improving operational safety. Overall, the proposed method provides a reliable and efficient solution for volume estimation in practical engineering applications.

Navigating through everyday environments, like walking down a sidewalk, which many people often take for granted, is a difficult task for millions of people with vision impairments since it involves sophisticated object detection, depth perception, and situational awareness, all working seamlessly to guide a person through complex surroundings. Many current assistive devices for vision-impaired people are either expensive, information-overabundant, or missing critical information. This paper details our Vision Alarming System (VAS), which can improve the safety for blind and vision-impaired people by providing awareness of both positions and nature of nearby obstacles; thus, assisting users to make decisions to avoid collisions, reduce accidents and casualties, while enhance their experience, independence, and confidence when participating in traffic. VAS is an Artificial Intelligence/Internet-of-Things (AI/IoT)—powered system developed utilizing the cutting-edge Raspberry Pi 5, a Light Detection and Ranging (LiDAR) sensor, and an AI depth camera, operating as different containers in a Docker architecture, and leveraging a Robotic Operating System 2 (ROS 2) backbone. VAS communicates the obstacle detections to users via Bluetooth interface, using the neural Text-To-Speech (TTS) system, namely, Piper, and the Sound eXchange (SoX) technologies. Our proof-of-concept system proves that VAS can be a standalone, open-source, extremely low cost, low power consumption assistive device which can synergistically utilize the cutting-edge AI/IoT technologies to provide blind and vision-impaired users with an appropriate amount of critical information about their surrounding environments.

Multi-object detection and tracking for cross-domain operations of underwater robots

Robotic Versus Laparoscopic Cholecystectomy: Operative Time and Conversion in Urgent Cases.

Robust gait phase recognition is essential for gait analysis, biomechanical monitoring, and human-centered robotics. A significant gap persists between computational methods and field deployment spanning generalizability, computational budget, hardware optimization, and integration with robotic frameworks. This study presents a reproducible wearable system that bridges this gap by combining robust, real-time recognition with minimal modalities. The proposed probabilistic heuristic recognition algorithm for sequential events (PHRASE) leverages statistical modeling and artificial neural networks (ANN) to achieve robust low-latency detection. The system integrates two inertial measurement units (IMUs) and a portable microcomputer within a Robot Operating System (ROS) framework, enabling seamless integration with external systems. The method was validated against benchmarks on 31 participants with diverse biometrics, sensor models, and physical conditions during multi-scenario level-ground walking, outperforming 5 state-of-the-art deep learning benchmarks. PHRASE demonstrates strong generalization, achieving an accuracy of specifically across diverse unseen subjects, conditions, and experimental setups combined. The accuracy improvement over the best-performing benchmark has a 95% confidence interval of . The wearable interface maintains a stable average inference latency of 11.6 ms. Overall, the proposed interface addresses key limitations in terms of resilience to unseen subjects, unseen sensor configurations, varying walking speeds, and latency. The source code and implementation details are available at: https://github.com/SamMans/PHRASE/tree/main . • Real-time Bayesian gait phase recognition using only two wearable IMUs. • Portable ROS-based interface for assistive and robotic gait applications. • Combination of heuristics, artificial neural networks and Bayesian inference. • More robust and generalizable than deep learning benchmarks in cross-setup testing without retraining. • Faster than deep learning benchmarks in processing latency.

Internet of Robotics Things (IoRT) shifts from traditional robotic operations to a paradigm shift that integrates resource-constrained robots as part of the Computing Continuum, providing additional resources and more advanced application capabilities. Edge Artificial Intelligence (AI) plays a pivotal role in this transition, facilitating real-time decision-making and timely adaptation in response to changes in the robot’s environment. In this paper, we propose a framework for the implementation of Edge AI in the IoRT, compatible with various IoRT scenarios. Robot resources are integrated into a unified orchestration platform, facilitating seamless and on-demand deployment of application nodes at the edge of the network in an event-triggered manner. An AI-assisted task offloading mechanism is proposed to cater to the needs of real-time adaptation of an inference pipeline based on the robot’s state. Experimental results indicate that the framework outperforms baseline techniques in terms of accuracy of the tracking algorithm and a significant reduction in energy consumption of the edge resources due to the n-demand deployment.

This paper presents a digital twin system that facilitates robot operation and combined prediction of state and dynamics of a biped robot with imperfections such as parasitic compliance. The key contribution lies in utilising proprioceptive sensor feedback together with a contact-aware forward-dynamics (shadow) simulation, augmented by a contact-implicit inverse-dynamics controller—assuming MuJoCo's rigid-body contact formulation—to improve the replication of the robot's posture and dynamics. The speed and accuracy of the developed method is demonstrated through physical experiments on a small biped robot standing on flat, steps or inclined ground. The predicted contact forces are validated using the feedback of a force-sensing robot foot via the concept of zero-tilting moment point. For the first time, a systematic design of a robot foot using force-sensing resistors is presented to achieve repeatability, linearity, and sensitivity to small loads.