Pneumatic artificial muscles (PAMs) have been widely used in actuation of medical devices due to their intrinsic compliance and high power-to-weight ratio features. However, the nonlinearity and time-varying nature of PAMs make it challenging to maintain high-performance tracking control. In this article, a high-order pseudopartial derivative-based model-free adaptive iterative learning controller (HOPPD-MFAILC) is proposed to achieve fast convergence speed. The dynamics of PAM is converted into a dynamic linearization model during iterations; meanwhile, a high-order estimation algorithm is designed to estimate the pseudopartial derivative component of the linearization model by only utilizing the input and output data in previous iterations. The stability and convergence performance of the controller are verified through theoretical analysis. Simulation and experimental results on PAM demonstrate that the proposed HOPPD-MFAILC can track the desired trajectory with improved convergence and tracking performance.

/pdf/high-order-model-free-adaptive-iterative-learning-control-of-yl76au1zb6.pdf

High-Order Model-Free Adaptive Iterative Learning Control of Pneumatic Artificial Muscle With Enhanced Convergence

Gait and waist motions always contain massive personnel information and it is feasible to extract these data via wearable electronics for identification and healthcare based on the Internet of Things (IoT). There also remains a demand to develop a cost-effective human-machine interface to enhance the immersion during the long-term rehabilitation. Meanwhile, triboelectric nanogenerator (TENG) revealing its merits in both wearable electronics and IoT tends to be a possible solution. Herein, the authors present wearable TENG-based devices for gait analysis and waist motion capture to enhance the intelligence and performance of the lower-limb and waist rehabilitation. Four triboelectric sensors are equidistantly sewed onto a fabric belt to recognize the waist motion, enabling the real-time robotic manipulation and virtual game for immersion-enhanced waist training. The insole equipped with two TENG sensors is designed for walking status detection and a 98.4% identification accuracy for five different humans aiming at rehabilitation plan selection is achieved by leveraging machine learning technology to further analyze the signals. Through a lower-limb rehabilitation robot, the authors demonstrate that the sensory system performs well in user recognition, motion monitoring, as well as robot and gaming-aided training, showing its potential in IoT-based smart healthcare applications.

Wearable Triboelectric Sensors Enabled Gait Analysis and Waist Motion Capture for IoT-Based Smart Healthcare Applications

Traditional joint space admittance controller for N-DOF robotic systems is complexity and easily leads to incongruous movement among all joints. Our study introduces a central pattern generator (CPG) network into one-dimension joint space admittance control for the custom-made lower limb robotic exoskeleton with four DOFs, to guarantee the coordinated movement and security of users. The predefined trajectories for four joints are produced by CPG. Unilateral knee joint torque of subjects is detected based on corresponding muscle EMG signals. The torque is transformed into an additional set of state variables for CPG based on the one-dimension admittance controller. CPG harmonically adjusts the predefined trajectories by the additional state variables. Finally, the robotic exoskeleton completes the predefined trajectories with a classical PID controller.

A generalized framework to achieve coordinated admittance control for multi-joint lower limb robotic exoskeleton

Although industrial robots are widely used in production automation, their applications in machining have been limited because of the structural vibrations induced by periodic cutting forces. Since the dynamic characteristics of an industrial robot depends on its configuration, the responses of the robot structure to the cutting forces are affected by how the workpiece is placed within the workspace of the robot. This paper presents a method for workpiece pose optimization for a robotic milling system to improve the quasi-static performance during machining. Since the milling forces are time-varying due to the characteristics of the multi-tooth and discontinuity of milling, these forces can excite vibrations inside the robot structure. To address this issue, a structural dynamics model is established for industrial robots, considering their joint flexibility, and a milling force formulation is incorporated into the robot dynamics model to investigate the forced vibrations of the flexible joints. Then, the quasi-static performance of the robotic machining system is evaluated by the vibration-induced offset of the cutter tool that is mounted on the end-effector. Finally, an optimization approach is given for the workpiece pose to minimize the cutter tool offset under the periodic milling force. A numerical simulation demonstrates that the optimal workpiece pose can significantly reduce the overall tool offset during machining and can lower the variation of the tool offset along the milling path.

Workpiece Pose Optimization for Milling with Flexible-Joint Robots to Improve Quasi-Static Performance

This article propounds addressing the design of a sliding mode controller with adaptive gains for trajectory tracking of unicycle mobile robots The dynamics of this class of robots are strong, non

https://journals.sagepub.com/doi/pdf/10.1177/1729881420987082

Design of a robust tracking controller for a nonholonomic mobile robot based on sliding mode with adaptive gain

This paper discusses the problem of squatting training of stroke patients. The main idea is to correct the patient's training trajectory through an iterative learning control (ILC) method. To obtain better rehabilitation effect, a patient will typically be required to practice a reference posture for many times, while most of active training methods can hardly keep the patients training with correct posture. Instead of the conventional ILC strategy, an impedance-based iterative learning method is proposed to regulate the impedance value dynamically and smartly which will help patients correct their posture gradually and perform better. To facilitate impedance-based ILC, we propose two objectives. The first objective is to find the suitable values of impedance based on the ILC scheme. The second objective is to search the moderate learning convergence speed and robustness in the iterative domain. The simulation and experimental results demonstrate that the performance of trajectory tracking will be improved greatly via the proposed algorithm.

/pdf/iterative-learning-impedance-for-lower-limb-rehabilitation-2a4ftyepn0.pdf

Iterative Learning Impedance for Lower Limb Rehabilitation Robot

The force analysis of a pelvic support walking robot with joint compliance is discussed in this paper. During gait training, pelvic motions of hemiplegic patients may be excessively large or out of control; however, restriction of pelvic motions is not likely to facilitate successful rehabilitation. A robot-assisted pelvic balance trainer (RAPBT) is proposed to help patients control the range of motion via force field, and force analysis is necessary for the control of the compliant joints. Thus, kinematic model and static model are developed to derive the Jacobian and the relation between the interaction forces and the pelvic movements, respectively. Since the joint compliance is realized through a nontorsional spring, a conventional (linear) Jacobian method and a piecewise linear method are derived to relate the interaction forces with the pelvis movements. Three preliminary experiments are carried out to evaluate the effectiveness of the proposed methods and the feasibility of the RAPBT. The experiment results indicate that the piecewise linear method is effective in the calculation of the interaction forces. Gait with pelvic brace strongly resembles free overground walking and partly decreases motion range via force field. The findings of this research demonstrate that the pelvic brace with joint compliance may provide effective interventions.

/pdf/force-analysis-and-evaluation-of-a-pelvic-support-walking-53u5rk0ev8.pdf

Force Analysis and Evaluation of a Pelvic Support Walking Robot with Joint Compliance.

After analyzing the rehabilitation needs of stroke patients and the previous studies on lower limb rehabilitation robot, our lower limb rehabilitation robot is designed for stroke patients' gait and balance training. The robot consists of the mobile chassis, the support column and the pelvis mechanism and it is described in detail. As the pelvis mechanism allows most of the patient's motion degrees of freedom (DOFs), the kinematics model of the mechanism is set up, and kinematics simulation is carried out to study the motion characteristics of the mechanism. After analyzing the calculation and simulation results, the pelvis mechanism is proven to measure up to the movement needs of the paralytic's waist and pelvis in walking rehabilitation process.

Lower limb rehabilitation robot for gait training

Process Parameter Optimization of a Mobile Robotic Percussive Riveting System With Flexible Joints

Nerve injury after stroke leads to disorders of locomotion and a declining balance function, which increases the risk of falling. Restriction of pelvic motions can hinder successful rehabilitation, hence a robotic-assisted rehabilitation trainer (RART) is proposed to assist patients in controlling the pelvic motions via force field. The mechanical design, kinetic framework, and the intention-based controller of the RART are introduced in this paper. Based on an intention-based control strategy, the robot generates force field that affects the pelvic motions (vertical and lateral motion, rotation, and obliquity) via a compliant pelvic brace while the patient is walking on ground. A control experiment with 16 hemiplegic patients is carried out to examine the effects on recovery of the balance function. Clinical evaluations and gait analysis are performed before and after the treatment. The experimental results show that the proposed control method is effective in motion recognition, and significant improvements of the balance function, gait speed, stride, stride frequency, Fugl-Meyer assessment, peak knee, and hip flexion angle during swing phase for their affected side are observed in the comparisons within the RART group. These preliminary results demonstrate that the proposed robot with force field and visual feedback may be effective in improving the balance function.

Shuai Guo

Papers

Iterative Learning Impedance for Lower Limb Rehabilitation Robot

Force Analysis and Evaluation of a Pelvic Support Walking Robot with Joint Compliance.

Lower limb rehabilitation robot for gait training

Process Parameter Optimization of a Mobile Robotic Percussive Riveting System With Flexible Joints

Robotic-Assisted Rehabilitation Trainer Improves Balance Function in Stroke Survivors