15 January 2021 in IEEE Transactions on Cybernetics
In this survey, various concepts and methodologies developed over the past two decades for varying and learning the impedance or admittance of robotic systems that physically interact with humans are explored. For this purpose, the assumptions and mathematical formulations for the online adjustment of impedance models and controllers for physical human–robot interaction (HRI) are categorized and compared. In this systematic review, studies on: 1) variation and 2) learning of appropriate impedance elements are taken into account. These strategies are classified and described in terms of their objectives, points of view (approaches), and signal requirements (including position, HRI force, and electromyography activity). Different methods involving linear/nonlinear analyses (e.g., optimal control design and nonlinear Lyapunov-based stability guarantee) and the Gaussian approximation algorithms (e.g., Gaussian mixture model-based and dynamic movement primitives-based strategies) are reviewed. Current challenges and research trends in physical HRI are finally discussed.
6 April 2022 in IEEE/ASME Transactions on Mechatronics
This article presents a strategy for autonomous locomotion trajectory planning for high-level control of lower limb exoskeletons by defining a novel set of adaptive central pattern generators (ACPGs) to facilitate safe and compliant interaction with the human. A time-varying bounded-gain adaptive disturbance observer is designed for estimating the human–robot interaction (HRI) needed for online central pattern generator (CPG)-based trajectory shaping and low-level nonlinear trajectory tracking control. The proposed ACPG dynamics for each exoskeleton joint updates the motion frequency and amplitude based on the observed HRI torque, which is also coupled with adjacent joints’ CPGs to regulate their phase differences in real time. An integrated Lyapunov analysis is conducted to ensure the closed-loop system’s stability and uniformly ultimately boundedness of both the tracking error and the torque estimation error in the controlled exoskeleton. Experimental studies are performed with an able-bodied human wearer by applying arbitrary torques on the exoskeleton’s joints in order to evaluate the proposed autonomous control strategy in online adjustment and personalization of the locomotion.
20 August 2021 in IEEE Robotics and Automation Letters
In this letter, a new adaptable gait trajectory shaping method is proposed for lower-limb exoskeletons by defining central pattern generators (CPGs). These CPGs are synchronized across different joints and updated online in response to the human users' physical behavior to enhance their safety and comfort. In this CPG structure of a high-level control scheme, an overall locomotion frequency is defined for all joint motions that can be modulated as a function of the human-robot interaction (HRI) energy. The amplitude and equilibrium position of oscillation for each joint can be adjusted in real-time based on the HRI torque. Logarithmic barrier functions are also formulated for these connected CPGs to avoid exceeding safe bounds of the joints' motion. A supervised learning algorithm is employed to identify the exoskeleton-limb dynamics and estimate the active HRI torque on different joints based on an autoregressive network with exogenous inputs (NARX) model. In order to track the reference trajectories generated by CPGs, a proportional derivative (PD) controller with torque compensation is designed. In the experimental evaluation of this intelligent control strategy, an able-bodied person wearing the Indego exoskeleton could amend and personalize the gait features considerably over a short period of time by applying active torques on different joints. In these experiments, the user increased the motion amplitude of the hip and knee joints up to 14% and made more than 200% variation in the gait frequency, which implies a considerable level of flexibility in locomotion planning for further user studies.
26 July 2021 in IEEE Robotics and Automation Letters
In this letter, an integrated control strategy is developed for both locomotion trajectory planning and postural stability, enabling shared autonomy between the human and lower-limb exoskeleton. Divergent component of motion (DCM) analysis was employed previously based on the linear inverted pendulum flywheel (LIPF) model to regulate the position of the center of mass (CoM) for humanoid robots. In this study, a new extended model is investigated for the DCM analysis by replacing the previous LIPF model, which is tailored for multi-degree-of-freedom (DOF) exoskeletons. This new model is designed to be personalized for each specific user's body by relaxing the assumption of having the total CoM at the hip joint in the previous LIPF model. Accordingly, the exoskeleton has the authority to ensure the postural stability and viability of locomotion in this human-robot interaction (HRI) by adjusting the upper body position using a DCM-based hip correction strategy. Integrating adaptive central pattern generators (CPGs), the human has enough authority to modify the gait trajectories in real-time, while the amplitude and frequency of walking are constrained to their feasible ranges. The effectiveness of this intelligent controller for safe and stable locomotion is investigated through experimental studies on a lower-limb exoskeleton.
21 February 2023 in Advanced Intelligent Systems
Healthcare systems are burdened by mobility impairments resulting from aging and neurological conditions. One of the recent advances in robotics is lower limb assistive/rehabilitative devices that can make independent living possible. Nonetheless, some limitations need to be addressed before robotics can be used in home‐based applications. This paper describes the current state of the art in intelligent motion planning and control of lower limb assistive devices, which have addressed some of these challenges. Adaptable central pattern generators and the divergent component of motion are introduced as methods for personalized motion planning based on physical human–robot interaction (pHRI). Uncertainty analysis for neural networks is introduced to increase safety in motion planning based on pHRI. For the case that a user cannot apply physical interaction, a reinforcement‐learning‐based approach is introduced to switch between different modes of walking based on the user's input via a push button embedded in a walker. Moreover, a smart walker is introduced as a device that can be synchronized with the lower limb exoskeleton to assist users with their daily activities. Also, a roadmap for future steps that can make lower limb assistive/rehabilitative devices a good fit for home use is introduced.