The complex interactions between flexible needles and tissues present significant challenges in predicting the needle shape during the puncture procedure. In particular, the accurate prediction of flexible needle shape during insertion into complex multilayer tissues, especially when measurement feedback involves non-Gaussian noise, remains an open problem. In this article, we develop a novel reinforcement learning-based active modeling scheme to predict the deflection of the robotic flexible needle. First, the active modeling scheme is constructed by deriving an extended Kalman filter under the maximum correntropy criterion to enhance insensitivity to non-Gaussian noise. Subsequently, based on this scheme, the reinforcement active modeling (RAM) framework is built by incorporating reinforcement learning to compensate for the modeling residuals. Specifically, the theoretical convergence of the proposed scheme is proved by using the Banach fixed-point theorem, thereby ensuring the reliability of needle shape prediction. Finally, a series of comparative experiments is carried out on a self-built robotic flexible needle. The experimental results demonstrate the superior performance of the proposed deflection predictor. Under non-Gaussian noise conditions, the proposed RAM scheme achieves a generalization prediction error reduction of 46.4% in RMSE and over 76.1% in Var during insertion into unknown multilayer tissue.

Increasing evidence highlights the role of proprio-ceptive deficits in falls, emphasizing the need for targeted rehabilitation in populations with functional movement disorders. Despite advances in rehabilitation robots, movement constraints still hinder active engagement of the lower limb muscles, thereby limiting the effectiveness of proprioceptive training. In this work, We developed a neuro-rehabilitation robotic platform to address this need by physically and virtually simulating multi-terrain scenarios. The robot introduces common perturbations, such as uneven mountain trails, sandy beaches, and bumpy bus rides, to assess user stability and recovery, thereby assisting in the design of individualized training programs. The platform enhances neuromuscular responses across multiple directions and facilitates targeted muscle contraction through motor tasks that combine proprioceptive and visual feedback. Preliminary studies demonstrated that the robot successfully facilitated a complete range of ankle rotational movements. Electromyographic analysis revealed increased activation of specific muscle groups, changes in muscle loading and contraction patterns, suggesting that the system recruits multiple muscle groups while enhancing proprioceptive input to periarticular soft tissues. The proposed robot and control strategies established a feasible solution to enhance proprioception rehabilitation.

Needle puncture is a fundamental technique in minimally invasive surgical procedures. However, the limited flexibility of flexible needles and their complex interactions with tissues make it challenging to avoid critical organs along the puncture path. Preoperative path planning, which generates feasible collision-free trajectories, can effectively reduce repeated punctures and mitigate patient discomfort. To address this challenge, a flexible needle with increased maximum curvature is designed, which introduces more complex kinematic characteristics and poses greater challenges for trajectory planning under kinematic constraints. Then, for the first time, a convex feasible set (CFS)-based flexible needle trajectory planning method is developed to tackle the non-convex optimization problem posed by obstacle avoidance in unstructured surgical environments. Specifically, our method explicitly incorporates kinematic and curvature constraints, enabling direct generation of feasible trajectories without additional post-processing. Finally, comparative experiments on a self-developed robotic-assisted flexible needle system demonstrate the superior performance of the proposed algorithm. In particular, the proposed trajectory generation method allows the flexible needle to effectively avoid obstacles and accurately reach the target.

Hand exoskeletons can recognize user’s intent and provide active resistance training to enhance finger strength in stroke patients. However, achieving fine human-robot interaction (HRI) while maintaining system simplicity for lightweight design remains a key challenge. In this work, we present an improved flexible hand exoskeleton with series elastic actuator (SEA) for hand strength estimation and progressive resistance exercise. The SEA design allows the hand exoskeleton to have backdrivability to improve HRI performance. By combining the flexible linkage with the flex sensor, we propose a novel user interface that is able to sensitively acquire hand motion intent. An Extended Kalman Filter (EKF) based tracking errors estimation is designed to evaluate the finger strength. The results of the finger strength estimation are used to adjust the parameters of the admittance model to provide small or large damping when the user’s finger strength is low or high, achieving active admittance control based progressive resistance exercise. The feasibility has been demonstrated by two sets of experiments, and this work has established a hand exoskeleton solution for finger strength estimation and fine human-robot interaction.

Hysteresis is a typical nonlinear characteristic that exists in mechanical systems, which brings significant challenges to the robust tracking control of twisted tendon-sheath systems. In this paper, an active data-driven model is proposed to describe the hysteresis phenomenon of a twisted tendon-sheath system based on the Koopman operator, and a robust controller is designed to cancel the effect of the model error and deal with the physical constraints in practical applications. First, by utilizing the Koopman theory, an active data-driven model is built to describe the twisted tendon-sheath hysteresis system in a straightforward linear form. Then, an active model is proposed based on a modified set-membership filter to estimate the finite-dimensional approximation error. Furthermore, a robust controller is developed by taking advantage of both the magnitude and bound of the model error (obtained by the active model) to enhance the control performance while considering security constraints. To the best of our knowledge, the rule-based constraint term is first considered in the data-driven model-based control scheme to prevent potential instabilities for the twisted tendon-sheath system. The theoretical stability of the closed-loop system is proven by using the barrier Lyapunov theory to ensure the security boundary. Extensive experiments are also carried out on a self-built robotic ureteroscopy prototype to demonstrate the superior tracking performance and robustness of the proposed method. Note to Practitioners—This paper is motivated by the accurate transmission problems of twisted tendon-sheath hysteresis systems, which aims to provide a precise active modeling method and a robust controller for the robotic-assisted instrument (e.g., endoscope, catheter, etc.) twisting in the sheath/orifice. Most existing studies on tendon-sheath hysteresis systems realize trajectory tracking controllers by using parametric-model-based compensation, which still lacks a practical data-driven modeling approach to characterize the hysteresis phenomenon in the linear form, and ignore the security constraints of tendon outputs. Based on the set-membership filter, this paper builds an active Koopman-based model, which is a practical method to follow for systems characterized by complex dynamics. Subsequently, by employing the constructed active model and a rule-based term to handle output constraints, a robust controller is elaborately designed to realize accurate tracking control for twisted tendon-sheath hysteresis systems. In particular, no priori knowledge of the complex dynamics is required in the implementation and gains selection of the proposed controller, which holds theoretically and practically significance for various tendon-sheath hysteresis systems. A series of comparative hardware experiments further validate the effectiveness and robustness of the suggested control scheme. In future work, we will aim to extend the applicability of the proposed active modeling and control scheme to interventional procedures of endoscopic operation robots for complex steerings with varying sheath configurations.

Hysteresis has posed significant challenges to the modeling and control of flexible endoscopic robots, which impedes the advancement of automated endoscopic operation. Despite numerous hysteresis modeling approaches aimed at improving accuracy, there are still several unresolved issues, such as inappropriate model selection and non-ideal assumption of noise. Focusing on these challenges, a novel reinforcement active modeling (RAM) scheme is proposed in this paper. By incorporating reinforcement learning, this method augments an Extended Kalman Filter (EKF)-based active modeling strategy, which improves the insensitivity and generalization ability to non-Gaussian noise that is not introduced in training. Finally, a series of comparative experiments are conducted on the self-built flexible endoscopic robot to validate the improvement achieved by the proposed scheme. Compared with some widely-applied methods, the proposed scheme achieved at least 63.8% improvement in the root mean square error (RMSE) in modeling accuracy under Gaussian noise conditions, and at least 36.5% improvement in RMSE under Poisson noise conditions.

The robotic flexible endoscope is developed rapidly in the field of surgery robots due to its high flexibility and safety. However, some inherent features, e.g., high nonlinearity, material creep, complex dynamic hysteresis behaviors, and the unknown coupling effects between bending and twisting motions, can lead to the significant degradation on three-dimensional (3-D) positioning performance of the endoscope. Aiming at these challenges, this paper built a practical multi-motion hysteresis phenomenon model for the bending and twisting motions of the robotic flexible endoscope with consideration of the coupling effects. Then, the time-optimal synchronous terminal motion planner is first proposed for the 3-D motions of the robotic endoscope to decouple the coupling effects in an intuitive separate control scheme. Finally, a series of hardware experiments are conducted on a robotic flexible ureteroscope platform. The accuracy of the proposed model and the trajectory-planning-based decoupling strategy is comprehensively validated. Particularly, the experimental results with the proposed trajectory planner show the satisfactory performance of vibration suppression and over-shoot suppression.

Functional near-infrared spectroscopy (fNIRS) classification of mental states is of important significance in many neuroscience and clinical applications. Existing classification algorithms use all signal-collected brain regions as a whole, and brain sub-region contributions have not been well investigated. This paper proposes a functional region decomposition (FRD) method to incorporate brain sub-region contributions and enhance fNIRS classification of mental states. Specifically, the method iteratively decomposes the brain region into multiple sub-regions to maximize their contributions with respect to the validation accuracy and coverage of brain sub-regions. Then for the fNIRS data in brain sub-regions, features are extracted and classified to output the predictions. The final predictions are determined by fusing predictions from multiple brain sub-regions with stacking. Experiments on a publicly available fNIRS dataset showed that the proposed functional region decomposition method led to 9.01% and 10.58% increase of classification accuracy for the methods related to slope-based features and mean concentration change features, respectively. Therefore, the proposed method can decompose the brain region into sub-regions with respect to their functional contributions and fundamentally enhance the performance of mental state classification.

Variable impedance control is advantageous for physical human-robot interaction to improve safety, adaptability and many other aspects. This paper presents a gain-scheduled variable stiffness control approach under strict frequency-domain constraints. Firstly, to reduce conservativeness, we characterize and constrain the impedance rendering, actuator saturation, disturbance/noise rejection and passivity requirements into their specific frequency bands. This relaxation makes sense because of the restricted frequency properties of the interactive robots. Secondly, a gain-scheduled method is taken to regulate the controller gains with respect to the desired stiffness. Thirdly, the scheduling function is parameterized via a nonsmooth optimization method. Finally, the proposed approach is validated by simulations, experiments and comparisons with a gain-fixed passivity-based PID method.

Series elastic actuators (SEAs) are growingly important in physical human-robot interaction (HRI) due to their inherent safety and compliance. Cable-driven SEAs also allow flexible installation and remote torque transmission, etc. However, there are still challenges for the impedance control of cable-driven SEAs, such as the reduced bandwidth caused by the elastic component, and the performance balance between reference tracking and robustness. In this paper, a velocity sourced cable-driven SEA has been set up. Then, a stabilizing 2 degrees of freedom (2-DOF) control approach was designed to separately pursue the goals of robustness and torque tracking. Further, the impedance control structure for human-robot interaction was designed and implemented with a torque compensator. Both simulation and practical experiments have validated the efficacy of the 2-DOF method for the control of cable-driven SEAs.

Autonomous as well as teleoperated robots find wide applications in various environments. Their capability to accomplish complex and dynamic operations can be significantly improved by fusing human intelligence with autonomous algorithms. In this paper, we propose a haptic shared control algorithm to provide flexible human assistance for semi-autonomous mobile robots. Through the admittance and impedance models, the haptic shared controller smoothly puts together human operator inputs with robot autonomy. Further, the level of autonomy is fully determined by the operator with the grasp motion. A decomposed design has been taken for the autonomous controller of the mobile robot. The algorithm was implemented on the haptic interface omega.7 together with a QBot mobile robot, and its feasibility and efficacy have been validated by experiments.