IEEE ... International Conference on Rehabilitation Robotics : [proceedings]最新文献

Most commercial ankle-foot orthoses (AFOs) are passive structures that cannot modulate stiffness to assist with a diverse range of activities, such as stairs and ramps. It is sometimes possible to change the stiffness of passive AFOs through reassembly or benchtop adjustment, but they cannot change stiffness during use. Passive AFOs are also limited in their ankle mechanics and cannot replicate a biomimetic, nonlinear torque-angle relationship. Many research labs have developed ankle exoskeletons that show promise as viable alternatives to passive AFOs, but they face challenges with reliability, mass, and cost. Consequently, commercial translation has largely failed to date. Here we introduce the Variable Stiffness Orthosis (VSO), a quasi-passive variable stiffness ankle-foot orthosis that strikes a balance between powered and passive systems, in terms of mass, complexity, and onboard intelligence. The VSO has customizable torque-angle relationships via a cam transmission, and can make step-to-step stiffness adjustments via motorized reconfiguration of a spring support along a lead-screw. In this work, we introduce two versions: a nominal and a stiff prototype, which differ primarily in their mass and available stiffness levels. The available torque-angle relationships are measured on a custom dynamometer and closely match model predictions. The experimental results showed that the prototypes are capable of producing ankle stiffness coefficients between 9 - 330 Nm/rad.

Evaluating trunk control ability is significant in guiding patients towards proper functional training. Most existing devices have only a singular assessment function, resulting in prolonged and asynchronous assessments. Devices with multi-dimensional assessment capabilities may address these limitations. This study utilizes a robotic brace, RoboBDsys-II, to assess the trunk ability of individuals with spinal disorders and to validate its effectiveness. The device can simultaneously collect kinematic, kinetic, and center of pressure data, reducing the assessment time and enabling the simultaneous assessment. The force platform is designed to measure the center of pressure and the force control of the parallel module is developed for the coronal movement assessment. Four patients with spinal cord injury participated in the study to assess their trunk range of motion and muscle strength. Results demonstrate that the trunk range of motion determines the center of pressure metrics in lateral bending experiments. Furthermore, RoboBDsys-II exhibits excellent test-retest reliability in lateral bending experiments and can reveal the muscle strength differences in different directions. The system has potential advantage in the trunk ability assessment.

Computer systems based on motion assessment are promising solutions to support stroke survivors' autonomous rehabilitation exercises. In this regard, researchers keep trying to achieve engaging and low-cost solutions suitable mainly for home use. Aiming to achieve a system with a minimal technical setup, we compare Microsoft Kinect, OpenPose, and MediaPipe skeleton tracking approaches for upper extremity quality of movement assessment after stroke. We determine if classification models assess accurately exercise performance with OpenPose and MediaPipe data against Kinect, using a dataset of 15 stroke survivors. We compute Root Mean Squared Error to determine the alignment of trajectories and kinematic variables. MediaPipe World Landmarks revealed high alignment with Kinect, revealing to be a potential alternative method.

Functional electrical stimulation (FES) has been increasingly integrated with other rehabilitation devices, including rehabilitation robots. FES cycling is one of the common FES applications in rehabilitation, which is performed by stimulating leg muscles in a certain pattern. The appropriate pattern varies across individuals and requires manual tuning which can be time-consuming and challenging for the individual user. Here, we present an AI-based method for finding the patterns, which requires no extra hardware or sensors. Our method starts with finding model-based patterns using reinforcement learning (RL) and customised cycling models. Next, our method fine-tunes the pattern using real cycling data and offline RL. We test our method both in simulation and experimentally on a stationary tricycle. Our method can robustly deliver model-based patterns for different cycling configurations. In the experimental evaluation, the model-based pattern can induce higher cycling speed than an EMG-based pattern. And by using just 100 seconds of cycling data, our method can deliver a fine-tuned pattern with better cycling performance. Beyond FES cycling, this work is a case study, displaying the feasibility and potential of human-in-the-loop AI in real-world rehabilitation.

Wheelchair users are often perceived as someone ill and who will be limited in performing daily activities. This paradigm can be changed if instead to focus on limits, we start to think about the new possibilities that could be explored from their current mobility and technology. We present a novel dancing wheelchair with augmented mobility named Volting. Our novel wheelchair was designed to tilt the seat laterally up to 14°. This inclination is performed proportionally to the inclination of the user by a mechanism based on passive suspensions. Our system was analyzed as a double inverted pendulum and a mathematical model was developed using Euler-Lagrange equations. This analysis was used to calculate the ideal stiffness. Thus, we performed experiments with three distinct stiffness values and varying the weight of participants to analyze the behavior of our mechanism. Our results show that lateral inclinations in our wheelchair can be unstable, low sensitivity or linear tendency. The latter behavior, which is the most appropriate, was obtained using the suspension whose stiffness was close to the ideal value, thus validating our mathematical approach. Moreover, this behavior was maintained even if the user weight varies up to 10kg above the estimated value, ensuring a good performance for varying morphologies. Finally, our device was tested by a professional wheelchair dancer who shows the new possibilities of Volting in terms of mobility.

In this paper, we propose a task-generic learning-based model for the control of a powered ankle exoskeleton. In contrast to the traditional state machine-based control approaches that hard codes the transition heuristics for the different states and motion conditions during gait, we propose to learn the finer constraints of gait from multiple demonstrations of human gait. We validate our proposed approach on a dataset of ten subjects walking on various inclines and at multiple speeds. We deploy our model on an ankle exoskeleton, and conduct user studies on able-bodied subjects who perform gait scenarios across varying speeds and inclines. We conduct multiple online experiments to validate our learning-based approach for different motion conditions, e.g., normal walking, walking at different speeds and inclines, turns, cross-overs with variable speed and cadence, walking on a treadmill as well as on level ground. We find that our proposed learning-based model has the capability to extrapolate its learned decision rules to support untrained gait conditions, for, e.g., walking at higher speeds and inclines not seen during training. The subjects were able to adapt to the different gait scenarios comfortably without loss of stability.

In this work, we present the implementation of a momentum-based balance controller in a lower-limb exoskeleton that can successfully reject perturbations and self-balance without any external aid. This controller is able to withstand pushes in the order of 30 N in forward and sideways directions with little sway. Additionally, with this controller, the system can perform balanced weight-shifting motions without the need for an explicit joint reference trajectory. There is potential, with fine parameter tuning, for a more robust balance performance that can reject stronger pushes during the presented tasks. Backward pushes were not rejected due to practical limitations (the mass of the device is concentrated in the back) rather than due to the control method itself. This controller is a preliminary result that brings paraplegic patients closer to crutch-free balance in a lower-limb exoskeleton.

The Assistive Robotic Arm Extender (ARAE) is an upper limb assistive and rehabilitation robot that belongs to the end-effector type, enabling it to assist patients with upper limb movement disorders in three-dimensional space. However, the problem of gravity compensation for the human upper limb with this type of robot is crucial, which directly affects the deployment of the robot in the assistive or rehabilitation field. This paper presents an adaptive gravity compensation framework that calculates the compensated force based on the estimated human posture in 3D space. First, we estimated the human arm joint angles in real-time without any wearable sensors, such as inertial measurement unit (IMU) or magnetic sensors, only through the kinematic data of the robot and established human model. The performance of the estimation method was evaluated through a motion capture system, which validated the accuracy of joint angle estimation. Second, the estimated human joint angles were input to the rigid link model to demonstrate the support force profile generated by the robot. The force profile showed that the support force provided by the developed ARAE robot could adaptively change with human arm postures in 3D space. The adaptive gravity compensation framework can improve the usability and feasibility of the 3D end-effector rehabilitation or assistive robot.

Small obstacles on the ground often lead to a fall when caught with commercial prosthetic feet. Despite some recently developed feet can actively control the ankle angle, for instance over slopes, their flat and rigid sole remains a cause of instability on uneven grounds. Soft robotic feet were recently proposed to tackle that issue; however, they lack consistent experimental validation. Therefore, this paper describes the experimental setup realized to test soft and rigid prosthetic feet with lower-limb prosthetic users. It includes a wooden walkway and differently shaped obstacles. It was preliminary validated with an able-bodied subject, the same subject walking on commercial prostheses through modified walking boots, and with a prosthetic user. They performed walking firstly on even ground, and secondly on even ground stepping on one of the obstacles. Results in terms of vertical ground reaction force and knee moments in both the sagittal and frontal planes show how the poor performance of commonly used prostheses is exacerbated in case of obstacles. The prosthetic user, indeed, noticeably relies on the sound leg to compensate for the stiff and unstable interaction of the prosthetic limb with the obstacle. Therefore, since the limitations of non-adaptive prosthetic feet in obstacle-dealing emerge from the experiments, as expected, this study justifies the use of the setup for investigating the performance of soft feet on uneven grounds and obstacle negotiation.

This work describes a three-degrees-of-freedom rehabilitation exoskeleton robot for wrist articulation movement: the Biomech-Wrist. The proposed development includes the design requirements based on the biomechanics and anthropometric features of the upper limb, the mechanical design, electronic instrumentation, software design, manufacturing, control algorithm implementation, and the experimental setup to validate the functionality of the system. The design requirements were set to achieve human wrist-like movements: ulnar-radial deviation, flexion-extension, and pronation-supination. Then, the mechanical design considers the human range of motion with proper torques, velocities, and geometry. The manufacturing consists of 3D-printed elements and tubular aluminum sections resulting in lightweight components with modifiable distances. The central aspect of the instrumentation is the actuation system consisting of three brushless motors and a microcontroller for the control implementation. The proposed device was evaluated by considering two control schemes to regulate the trajectory tracking on each joint. The first scheme was the conventional proportional-derivative controller, whereas the second was proposed as a first-order sliding mode. The results show that the Biomech-Wrist exoskeleton can perform trajectory tracking with high precision ( RMSE_max = 0.0556 rad) when implementing the sliding mode controller.