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.

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.

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.

Pneumatic artificial muscle (PAM), featuring good flexibility and safety, has been widely used in rehabilitation and bionic robots. However, the complex hysteretic nonlinearities and uncertainties of the PAM cause great difficulties and challenges to the accurate modeling and controller design, especially when confronted with unknown external disturbances in applications. This paper proposes a robust control strategy with disturbance compensation for the hysteresis compensation and trajectory tracking of PAMs. Considering the high hysteretic nonlinearity of the PAM, a modified Prandtl-Ishlinskii model is used as a feedforward hysteresis compensator. For the linearized system, adaptive set-membership filtering (ASMF) is used to estimate the nonlinear terms and external disturbances of the overall system. A sliding mode controller (SMC) with disturbance compensation is designed and cascaded to the feedforward hysteresis compensator in series. The stability of the closed-loop system is theoretically proved. The proposed method guarantees that the tracking error of the PAM system is bounded. Finally, the effectiveness and robustness of the proposed controller are verified via a series of experiments on an in-house built testbench for PAMs. Note to Practitioners—With the increasing demand on human-robot interaction, the safety and compliance of robots have become one key requirement. PAM is a compliant actuator, exhibiting good flexibility, safety, and clean energy. PAM is widely used in rehabilitation robots, whereas its strong hysteresis nonlinearity and sensitivity to external disturbances affect its motion accuracy. This paper proposes an ASMF-based discrete SMC, which uses an inverse hysteresis model to compensate for the strong hysteresis of the PAM and uses ASMF to estimate the lumped disturbance of the system. Compared with the other filters, ASMF is unique in that its estimation error is bounded, which is very useful in the stability proof of the overall system. The effectiveness of the proposed controller is experimentally verified. Experimental results show that PAM’s hysteresis can be efficiently compensated, and the influence of external disturbances can be attenuated by the proposed controller, resulting in improved motion accuracy and robustness. In future work, efforts will be directed towards the modeling and control of PAMs in multi-DOF robots.

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.