The telerehabilitation robotic system has been envisioned as an alternative to conventional hospital-centered therapy because of convenient training and offering equal opportunity to access medical resources for patients in different areas. However, due to the internet communication latency, how to realize safe, stable, and biomechanics-perceptible remote therapist–patient interaction (RTPI) remains a significant challenge for the therapist-in-the-loop telerehabilitation (TILT) system. To address this issue, a novel position-position/stiffness (P-PK) telerehabilitation architecture was proposed in this article, which exchanges the position information of the therapist and patients and feeds back the reference stiffness of the patient's affected limb to the therapist's side. Furthermore, a novel state-switching RTPI control (S2-RTPIC) scheme is first presented for this P-PK architecture to induce the active participation of the patients during the online TILT training by the variable stiffness voluntary control and their biomechanical states could be synchronously perceived by the therapists over distances for teleassessments. The stability and transparency criteria of the S2-RTPIC scheme under asymmetric time delay conditions were comprehensively analyzed and theoretically proved. Experimental results showed the proposed S2-RTPIC scheme can provide safe RTPI training with effective biomechanical perceptions and participation-inducing training assistance to facilitate teleassessment and telerehabilitation.

This paper explores a hydraulically powered double-joint soft robotic fish called HyperTuna and a set of locomotion optimization methods. HyperTuna has an innovative, highly efficient actuation structure that includes a four-cylinder piston pump and a double-joint soft actuator with self-sensing. We conducted deformation analysis on the actuator and established a finite element model to predict its performance. A closed-loop strategy combining a central pattern generator controller and a proportional–integral– derivative controller was developed to control the swimming posture accurately. Next, a dynamic model for the robotic fish was established considering the soft actuator, and the model parameters were identified via data-driven methods. Then, a particle swarm optimization algorithm was adopted to optimize the control parameters and improve the locomotion performance. Experimental results showed that the maximum speed increased by 3.6% and the cost of transport (COT) decreased by up to 13.9% at 0.4 m/s after optimization. The proposed robotic fish achieved a maximum speed of 1.12 BL/s and a minimumCOTof 12.1 J/(kg·m), which are outstanding relative to those of similar soft robotic fish. Lastly, HyperTuna completed turning and diving–floating movements and long-distance continuous swimming in open water, which confirmed its potential for practical application

This paper provides a comprehensive analysis of the existing landscape of conventional and highly biomimetic robotic arms, highlighting a prevalent trade-off between size, range of motion, and load capacity in current highly biomimetic designs. To overcome the limitations, this paper undertakes an in-depth exploration of the human shoulder, focusing on the identification of mechanical intelligence within the biological glenohumeral joint such as the incomplete ball-and-socket structure, coupling stability of humeroradial and glenohumeral joints, and the self-locking mechanism of the glenohumeral joint. These intelligent features potentially enhance both the stability and mobility of robotic joints, all the while preserving their compactness. To validate these potential benefits, this paper introduces a novel, highly biomimetic robotic glenohumeral joint that meticulously replicates human musculoskeletal structures, from bones and ligaments to cartilage, muscles, and tendons. This novel design incorporates the mechanical intelligence found in the biological joint. Through rigorous simulations and empirical studies, this paper demonstrates that the aforementioned mechanical intelligences significantly enhance the flexibility and load capacity of the robot's glenohumeral joint. Furthermore, extensive manipulation experiments confirm the robustness and viability of the proposed highly biomimetic robotic arm. Remarkably, the presented robotic arm executed 46.25% glenohumeral flexion/extension, 105.43% adduction/abduction and 99.23% rotation, and can sustain a payload of 4 kg, and open the door which requires a torque of over 1.5 Nm to twist the handle. Hence, this paper not only validates the intrinsic mechanical intelligence identified in the deconstruction of the human shoulder joint, but also contributes a pioneering design of a new, highly biomimetic robotic arm, significantly pushing boundaries of current the robotic technology.

This paper delineates the formulation and verification of an innovative robotic elbow-and-forearm system design, mirroring the intricate biomechanics of human musculoskeletal systems. Conventional robotic models often undervalue the substantial function of soft tissues which provides a compromise between compactness, safety, stability, and range of motion. In contrast, this study proposes a holistic replication of biological joints, encompassing bones, cartilage, ligaments, and tendons, culminating in a biomimetic robot. The research underscores a compact and stable structure of the human elbow and forearm, attributable to a tri-bone framework and diverse soft tissues. The methodology involves exhaustive examinations of human anatomy, succeeded by a theoretical exploration of the contribution of soft tissues to the stability of a prototype robotic elbow-and-forearm system. Evaluation results unveil remarkable parallels in the range of motion between the robotic joints and their human counterparts. The robotic elbow emulates 98.8% of the biological elbow's range of motion, with high torque capacities of 11.25 Nm (extension) and 24 Nm (flexion). Similarly, the robotic forearm achieves 58.6% of the human forearm's rotational range, generating substantial output torques of 14 Nm (pronation) and 7.8 Nm (supination). Moreover, the prototype exhibits significant load-bearing abilities, resisting a 5 kg dumbbell load without substantial displacement. It demonstrates a payload capacity exceeding 4 kg and rapid action capabilities, such as lifting a 2 kg dumbbell at a speed of 0.74 Hz and striking a ping-pong ball at an end-effector speed of 3.2 m/s. This research underscores that a detailed biomechanics study can address existing robotic design obstacles, optimize performance and anthropomorphic resemblance, and reaffirm traditional anatomical principles.

Exploring human hand fundamental biomechanical features and exploiting them to robotic hands have been proven to be an effective approach to enhancing artificial hands' performance, especially when interacting with various objects in dynamic unstructured environments. In this article, a bioinspired anthropomorphic robotic finger is first proposed, which embeds human finger musculoskeletal features in the design. Based on this design, three human-finger-like biomechanical advantages are systematically investigated and embodied in the bioinspired robotic finger. This article for the first time derives, presents, and experimentally verifies the mathematical models for the variable stiffness of finger ligamentous joints and self-adaptive morphing mechanism of finger flexible tendon sheaths, and validates and compares the influence of the reticular and linear extensor morphologies on fingertip feasible forces in three-dimensional (3-D) space. In this Part I of the article, two of the biomechanical properties, i.e., joint stiffness generated by the ligamentous joint of the finger, and fingertip feasible force space influenced by the reticular extensor mechanism are systematically investigated through theoretical modeling and experimental verification. Correspondingly, two biomechanical advantages were found, i.e., the ligamentous joint of the finger could provide anisotropic variable joint stiffness, enhancing the adaptivity, dexterity, and stability of fingers; and a reticular extensor mechanism could enlarge the fingertip feasible force space in 3-D space by 30.9% theoretically and 146.4% experimentally on average compared with the linear extensor, contributing to enrich force conditions during interactions. The third biomechanical advantage, i.e., fingertip force–velocity workspace can be augmented through the flexible tendon sheath, and grasping tests for a robotic hand designed with the aforementioned advantages are presented in Part II of this article.

The human hand has a fantastic ability to interact with various objects in the dynamic unstructured environment of our daily activities. We believe that this outstanding performance benefits a lot from the unique biological features of the hand musculoskeletal system. In Part I of this article, a bio-inspired anthropomorphic robotic finger was developed, based on which two human-finger-like biomechanical advantages were elaborately investigated, including the anisotropic variable stiffness associated with the ligamentous joints and the enlarged feasible force space associated with the reticular extensor mechanisms. In Part II, the fingertip force-velocity characteristics resulting from the flexible tendon sheath are studied. It indicates that the fingertip force–velocity workspace can be greatly augmented owing to the self-adaptive morphing of the flexible tendon sheaths, showing the average improvement of 41.2% theoretically and 117.5% experimentally compared with the results of 2 mm, 4 mm, and 6 mm size rigid tendon sheaths. Grasping tests and comparisons are then conducted with four three-fingered robotic hands (one with the robotic finger proposed in Part I, one with hinge joints, one with linear extensors, and one with rigid tendon sheaths) and the human hands of six subjects to handle various objects on flat, rough, and soft surfaces. The results show that the novel bio-inspired design in this research could improve the grasping success rates of the robotic hand. Compared with the grasping test results from the robotic hand with the bio-inspired robotic finger proposed in Part I, the overall grasping performance of a robotic hand with hinge joints, linear extensors, and rigid tendon sheaths decreases by 10%, 6%, and 17%, respectively. The results have also shown that with the embedded biomechanical advantages, even without complex control and sensory systems, the robotic fingers can achieve very comparable performance to human fingers in the grasping demonstrations presented, indicating average 94% of the success rate achieved by the human fingers. Successfully demonstrating 14 of 16 grasp types in the Cutkoskey taxonomy further shows the human-finger-like grasping capability of the proposed robotic fingers.