Mechatronics最新文献

{"title":"Learning-based modeling of a magnetically steerable soft suction device for endoscopic endonasal interventions","authors":"Majid Roshanfar ,&nbsp;Alex Zhang ,&nbsp;Changyan He ,&nbsp;Amir Hooshiar ,&nbsp;Dale J. Podolsky ,&nbsp;Thomas Looi ,&nbsp;Eric Diller","doi":"10.1016/j.mechatronics.2026.103468","DOIUrl":"10.1016/j.mechatronics.2026.103468","url":null,"abstract":"<div><div>This paper introduces a learning-based modeling framework for a magnetically steerable soft suction device designed for endoscopic endonasal brain tumor resection. The device is miniaturized (4 mm outer diameter, 2 mm inner diameter, 40 mm length), 3D printed using biocompatible SIL 30 material, and integrates embedded Fiber Bragg Grating (FBG) sensors for real-time shape feedback. Shape reconstruction is represented using four Bezier control points, allowing for a compact and smooth representation of the device’s deformation. A data-driven model was trained on 5097 experimental samples to learn the mapping from magnetic field parameters (magnitude: 0–14 mT, frequency: 0.2–1.0 Hz, and vertical tip distances from the surface of the electromagnet coil table: 90–100 mm) to the resulting geometric configuration of the soft robot, represented by four Bezier control points that define its 3D shape. The model was implemented and compared using both Neural Network (NN) and Random Forest (RF) architectures. The RF model outperformed the NN across all metrics, achieving a mean root mean square error of 0.087 mm in control point prediction and a mean shape reconstruction error of 0.064 mm. Feature importance analysis further revealed that magnetic field components predominantly influence distal control points, while frequency and distance affect the base configuration. Unlike prior studies that apply general machine learning methods to soft robotic data, the proposed framework introduces a new modeling paradigm that links magnetic actuation inputs directly to geometric Bezier control points, creating an interpretable and low-dimensional representation of deformation. This conceptual integration of magnetic field characterization, embedded FBG sensing, and Bezier-based learning provides a unified modeling strategy that can be extended to other magnetically actuated continuum robots. This learning-based approach effectively models the complex nonlinear behavior of hyperelastic soft robots under magnetic actuation without relying on simplified physical assumptions. By enabling sub-millimeter shape prediction accuracy and real-time inference, this work establishes an advancement toward the intelligent control of magnetically actuated soft robotic tools in minimally invasive neurosurgery.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"116 ","pages":"Article 103468"},"PeriodicalIF":3.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146116725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Resilient multi-UAV formation control with leader replacement under agent failures","authors":"Takuya Murakami ,&nbsp;Toru Namerikawa","doi":"10.1016/j.mechatronics.2026.103465","DOIUrl":"10.1016/j.mechatronics.2026.103465","url":null,"abstract":"<div><div>This paper develops and validates a resilient formation control for multi-UAV systems under agent failures. Building on our previous work, we extend resilient formation control that tolerates non-compensable faults and enables leader replacement to second-order UAV dynamics, and provide a theoretical stability analysis of the proposed control law. To address leader failures, we integrate a coordinated leader replacement algorithm with the resilient controller. The proposed scheme is validated through numerical simulations and multi-UAV flight experiments on small quadrotors. The results demonstrate that the system preserves formation even under severe leader failures, while the non-faulty agents asymptotically maintain consensus. These theoretical, simulation, and experimental results confirm the effectiveness of the proposed approach and highlight its practical applicability for resilient multi-agent systems.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103465"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146038509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hydraulic actuated leg with passive flexibility and energy efficiency for heavy-duty quadruped robots","authors":"Zisen Hua ,&nbsp;Dongyue Hua ,&nbsp;Xuewen Rong ,&nbsp;Yaru Sun","doi":"10.1016/j.mechatronics.2026.103461","DOIUrl":"10.1016/j.mechatronics.2026.103461","url":null,"abstract":"<div><div>To address the excessive foot-ground impact caused by inherent limitations of hydraulic systems in quadruped robots, while mitigating the over-dependence of traditional passive buffer structures on the touchdown posture of robot legs, this study proposes a novel limb structure design method based on joint passive compliance. The approach involves introducing a transitional link at the knee joint, thereby evolving the traditional three-segment leg topology with three-degree-of-freedom into a four-segment structure with three-degree-of-freedom. Additionally, a fully symmetric four-piston-rod pneumatic support mechanism charactered by tunable high stiffness and minimal deformation is incorporated to constrain the motion of the links at the knee joint, thereby ensuring determinacy in the kinematic relationships among the limb segments. Simultaneously, by optimizing link lengths and hinge point positions, the fixed deformation direction of the elastic element is achieved across almost the entire workspace of the foot-end postures, while joint driving forces during dynamic motion are further reduced. The effectiveness of the proposed leg design, in terms of energy efficiency, motion stability, and impact attenuation, is validated through experiments on a single-leg test platform.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103461"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145941054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reinforcement learning-based assist-as-needed control for shoulder rehabilitation robot based on virtual biomechanical model","authors":"Muhammad Faizan Shah","doi":"10.1016/j.mechatronics.2026.103467","DOIUrl":"10.1016/j.mechatronics.2026.103467","url":null,"abstract":"<div><div>This paper presents an adaptive control strategy for a shoulder rehabilitation robot, leveraging a machine learning-based Assist-as-Needed (AAN) framework to improve transparency and user participation during rehabilitation tasks. The proposed control architecture integrates a conventional impedance controller with a reinforcement learning-based torque modulation layer that actively adapts to patient effort. A biomechanical model of the shoulder joint, developed using a Virtual Biomechanical Shoulder Robot Model (VBSRM), informs the estimation of joint torques and stiffness values. Experimental evaluations involving ten healthy subjects compare the performance of the AAN mode against conventional passive control. Results demonstrate that the AAN strategy reduces joint torque and effort by up to 30%, while preserving accurate trajectory tracking and stable impedance adaptation. Stiffness profiles under AAN control exhibit smoother, user-contingent variations, indicating responsive and compliant assistance. The proposed method provides a robust and energy-efficient rehabilitation interface, paving the way for intelligent, patient-specific therapy protocols in physical human-robot interaction systems.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103467"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of a novel dual-motor driven wheel–foot transformation mechanism for wheel-biped robots","authors":"Jindai Zhang ,&nbsp;Kangcheng Zhang ,&nbsp;Jianlin Zhang ,&nbsp;Xuechao Chen ,&nbsp;Zhangguo Yu ,&nbsp;Qiang Huang","doi":"10.1016/j.mechatronics.2026.103466","DOIUrl":"10.1016/j.mechatronics.2026.103466","url":null,"abstract":"<div><div>Balancing mobility and adaptability in complex environments remains a major challenge for wheeled and biped robots, highlighting multi-modal actuation as a key research focus. This study proposes, designs, and experimentally validates a dual-motor wheel–foot transformation mechanism for wheel-biped robots, enabling both wheeled and legged locomotion. Guided by a “flip-deploy” design concept, the mechanism integrates a screw-nut drive, linkage system, and sliding block-slot mechanism in a coaxial dual-motor layout, achieving reliable bi-directional switching between wheeled and footed modes. The transmission relationships during the mechanism’s mode-switching process are analyzed, and the linkage lengths are optimized based on static force analysis in the footed mode. No-load, model validation and on-robot experiments demonstrate that the mechanism satisfies the design requirements, enabling stable, rapid, and robust posture transitions while exerting negligible influence on the robot’s overall posture stability. Overall, the proposed design makes it possible for wheel-biped robots to achieve multimodal locomotion and adapt to diverse terrains.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103466"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146038508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic modeling and control of an inextensible pneumatically actuated soft continuum manipulator","authors":"Samuel Pilch ,&nbsp;Eva Menrad ,&nbsp;Artem Beger ,&nbsp;Oliver Sawodny","doi":"10.1016/j.mechatronics.2026.103462","DOIUrl":"10.1016/j.mechatronics.2026.103462","url":null,"abstract":"<div><div>Modeling and control of soft continuum manipulators remain challenging due to their infinite degrees of freedom, nonlinear material properties, and demanding sensing requirements. This work presents a dynamic model-based centralized control approach for a spatially moving, inextensible soft continuum manipulator actuated by pneumatic network segments. The equations of motion are formulated using the Euler–Lagrange formalism, with segment stiffness represented by an experimentally identified spring moment model incorporating curvature, orientation, and pressure dependencies. Nonlinear system dynamics are linearized around the desired generalized coordinates, enabling a feedforward controller based on the linearized state representation combined with a PID feedback loop. State feedback is reconstructed from IMU measurements using spherical coordinates expressed in azimuth and zenith angles. The desired pressures are obtained through linear mapping from the target moments and further adapted by introducing a mean pressure, allowing simultaneous pressurization and depressurization of adjacent pneumatic networks for faster actuation. These adapted pressures are realized by an external, model-free pressure controller. The proposed method is experimentally validated, demonstrating accurate control of the continuum manipulator.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103462"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145979991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lyapunov-based model predictive control for path-following of an autonomous underwater vehicle using line-of-sight guidance","authors":"Guanghao Yang ,&nbsp;Le Li ,&nbsp;Kang Zhang ,&nbsp;Weidong Liu","doi":"10.1016/j.mechatronics.2026.103464","DOIUrl":"10.1016/j.mechatronics.2026.103464","url":null,"abstract":"<div><div>This paper studies the path-following problem of an underactuated autonomous underwater vehicle (AUV) with the ocean current disturbances. A line-of-sight (LOS) guidance law is employed at the outer loop, while a Lyapunov-based model predictive controller (LMPC) is developed at the inner loop to ensure that the AUV can accomplish the path-following task under ocean current disturbances. The proposed LOS-LMPC inherits the stability and robustness of the extended state observer (ESO)-based auxiliary control law and utilizes online optimization to enhance the path-following performance of the AUV system. Simulations and hardware experiments conducted on the “Qilin” AUV demonstrate the effectiveness of the proposed method.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103464"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145979992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tiny learning-based MPC for multirotors: Solver-aware learning for efficient embedded predictive control","authors":"Babak Akbari,&nbsp;Justin Frank,&nbsp;Melissa Greeff","doi":"10.1016/j.mechatronics.2025.103452","DOIUrl":"10.1016/j.mechatronics.2025.103452","url":null,"abstract":"<div><div>Tiny aerial robots hold great promise for applications such as environmental monitoring and search-and-rescue, yet face significant control challenges due to limited onboard computing power and nonlinear dynamics. Model Predictive Control (MPC) enables agile trajectory tracking and constraint handling but depends on an accurate dynamics model. While existing Learning-Based (LB) MPC methods, such as Gaussian Process (GP) MPC, enhance performance by learning residual dynamics, their high computational cost restricts onboard deployment on tiny robots. This paper introduces Tiny LB MPC, a co-designed MPC framework and optimization solver for resource-constrained micro multirotor platforms. The proposed approach achieves 100 Hz control on a Crazyflie 2.1 equipped with a Teensy 4.0 microcontroller, demonstrating a 43% average improvement in tracking performance over existing embedded MPC methods under model uncertainty, and achieving the first onboard implementation of LB MPC on a 53 g multirotor.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103452"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The ForceCAST framework: Methodology and tools for benchmarking force control algorithms","authors":"Eldison Dimo,&nbsp;Matteo Meneghetti,&nbsp;Noè Murr,&nbsp;Andrea Calanca","doi":"10.1016/j.mechatronics.2026.103460","DOIUrl":"10.1016/j.mechatronics.2026.103460","url":null,"abstract":"<div><div>This paper describes the outcomes of the Forecast project, aiming at providing tools and metrics to benchmark force control algorithms for robotics applications. The Forecast project recognizes the importance of considering the interacting environment in order to assess the performance of a force-controlled system. In many papers, force-controlled systems are often evaluated on too specific (and often favorable) environmental conditions, preventing readers from fairly understanding the overall system behavior. Starting from these observations, the Forecast project has developed tools and metrics to ease and standardize the benchmarking process. The objective of this paper is to present such tools and metrics and to foster their diffusion within the robotics community. A case study is proposed to practically showcase the benchmarking process.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103460"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145979993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"On hybrid inverse dynamic modeling for industrial robots","authors":"S. Clavel ,&nbsp;M. Alamir ,&nbsp;J. Faure-Favre","doi":"10.1016/j.mechatronics.2025.103451","DOIUrl":"10.1016/j.mechatronics.2025.103451","url":null,"abstract":"<div><div>Inverse dynamic modeling plays a crucial role in developing optimal feedforward control laws in robotics. A promising approach to improve its accuracy involves using <em>hybrid</em> approaches combining physics-based models with data-driven models. This paper presents an in-depth study of such an approach, based on Gated Recurrent Unit (GRU) neural networks for application on 4-joints and 6-joints <span>Stäubli</span> industrial robots. We start with a comprehensive review of the recent literature. Subsequently, we demonstrate that our model significantly surpasses more traditional black-box models in terms of accuracy and extrapolation. Furthermore, we explore the various factors that influence its accuracy with real data.</div></div>","PeriodicalId":49842,"journal":{"name":"Mechatronics","volume":"115 ","pages":"Article 103451"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145941053","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}