Download Free Towards Personalized Robot Assisted Gait Training Book in PDF and EPUB Free Download. You can read online Towards Personalized Robot Assisted Gait Training and write the review.

Robot-Assisted Training (RAT) is a growing body of research in Human-Robot Interaction (HRI) that studies how robots can assist humans during a physical or cognitive training task. Robot-Assisted Training systems have a wide range of applications,varying from physical and/or social assistance in post-stroke rehabilitation to intervention and therapy for children with Autism Spectrum Disorders. The main goal of such systems is to provide a personalized and tailored session that matches user abilities and needs, by adjusting task-related parameters (e.g., task difficulty, robot behavior), in order to enhance the effects of the training session. Moreover, such systems need to adapt their training strategy based on user's affective and cognitive states. Considering the sequential nature of human-robot interactions, Reinforcement Learning (RL) is an appropriate machine learning paradigm for solving sequential decision making problems with the potential to develop adaptive robots that adjust their behavior based on human abilities, preferences and needs. This research is motivated by the challenges that arise when different types of users are considered for real-time personalization using Reinforcement Learning, in a Robot-Assisted Training scenario. To this end, we present an Interactive Learning and Adaptation Framework for Personalized Robot-Assisted Training. This framework utilizes Interactive RL (IRL)methods to facilitate the adaptation of the robot to each individual, monitoring both behavioral (task performance) and physiological data (task engagement). We discuss how task engagement can be integrated to the personalization mechanism, through Learning from Feedback. Moreover, we show how Human-in-the-Loop approaches can be used to utilize human expertise using informative control interfaces, towards a safe and tailored interaction. We illustrate this framework with a Socially Assistive Robotic (SAR) system that instructs and monitors a cognitive training task and adjusts task diculty and robot behavior, in order to provide a personalized training session. We present our data-driven approach (data collection, data analysis, user modeling and simulation), as well as a user study to evaluate our real-time SAR-based prototype system for personalized cognitive training. We discuss the limitations and challenges of our approach, as well as possible future directions, considering the different modules of the proposed system (RL-based personalization, user modeling,EEG analysis, Human-in-the-Loop). The long-term goal of this research is to develop personalized and co-adaptive human-robot interactive systems, where both agents(human, robot) adapt and learn from each other, in order to establish an efficient interaction.
Neurologic injuries, such as stroke and spinal cord injuries (SCI), cause damage to neural systems and motor function, which results in lower limb impairment and gait disorders. Subjects with gait disorders require specific training to regain functional mobility. Traditionally, manual physical therapy is used for the gait training of neurologically impaired subjects which has limitations, such as the excessive workload and fatigue of physical therapists. The rehabilitation engineering community is working towards the development of robotic devices and control schemes that can assist during the gait training. The initial prototypes of these robotic gait training orthoses use conventional, industrial actuators that are either extremely heavy or have high endpoint impedance (stiffness). Neurologically impaired subjects often suffer from severe spasms. These stiff actuators may produce forces in response to the undesirable motions, often causing pain or discomfort to patients. The control schemes used by the initial prototypes of robotic gait training orthoses also have a limited ability to provide seamless, adaptive, and customized robotic assistance. This requires new design and control methods to be developed to increase the compliance and adaptability of these automated gait training devices. This research introduces the development of a new robotic gait training orthosis that is intrinsically compliant. Novel, assist-as-needed (AAN) control strategies are proposed to provide adaptive and customized robotic assistance to subjects with different levels of neurologic impairments. The new robotic gait training orthosis has six degrees of freedom (DOFs), which is powered by pneumatic muscle actuators (PMA). The device provides naturalistic gait pattern and safe interaction with subjects during gait training. New robust feedback control schemes are proposed to improve the trajectory tracking performance of PMAs. A dynamic model of the device and a human lower limb musculoskeletal model are established to study the dynamic interaction between the device and subjects. In order to provide adaptive, customized robot assisted gait training and to enhance the subject's voluntary participation in the gait training process, two new control schemes are proposed in this research. The first control scheme is based on the impedance control law. The impedance control law modifies the robotic assistance based on the human subject's active joint torque contributions. The levels of robot compliance can be selected by the physical therapist during the impedance control scheme according to the disability level and stage of rehabilitation of neurologically impaired subjects. The second control scheme is proposed to overcome the shortcomings of impedance control scheme and to provide seamless adaptive, AAN gait training. The adaptive, AAN gait training scheme is based on the estimation of the disability level of neurologically impaired subjects based on the kinematic error and adapts the robotic assistance accordingly. All the control schemes have been evaluated on neurologically intact subjects and the results show that these control schemes can deliver their intended effects. Rigorous clinical trials with neurologically impaired subjects are required to prove the therapeutic efficacy of the proposed robotic orthosis and the adaptive gait training schemes. The concept of intrinsically compliant robotic gait training orthosis, together with the trajectory tracking and impedance control of robotic gait training orthosis are the important contributions of this research. The algorithms and models developed in this research are applicable to the development of other robotic devices for rehabilitation and assistive purposes. The major contribution of the research lies in the development of a seamless, adaptive AAN gait training strategy. The research will help in evolving the field of compliant actuation of rehabilitation robots along with the development of new control schemes for providing seamless, adaptive AAN gait training.
Robotic-assisted gait training has been cited as a potential rehabilitation intervention for the recovery of ambulation in hemiplegic stroke patients. Previous reviews on robotic gait training have elicited a need for more randomized, clinical studies on robotic gait training for the development of evidence-based intervention protocols and elimination of research bias. The purpose of this scoping review was to provide an updated examination of the clinical research on robotic assisted gait training and address any research gaps to be considered for future research. This review included randomized, clinical trials on stroke patients with a robotic-assisted treatment (RAGT) and conventional gait training (CGT) group. This review included studies observing the following outcome measures: Functional Ambulatory Category, Gait Speed, 6-minute Walking Test, Berg Balance Scale, Fugl-Meyer Assessment, and Modified Ashworth Scale. Screening on Web of Science and PubMed revealed 11 studies for review from an initial pool of 231 unique titles. Studies focused on subacute stroke patients, non-ambulatory subjects, and longer intervention protocols were more likely to observe clinically meaningful improvements in outcome. RAGT groups were more likely to observe greater improvements in outcome than CGT groups. However, individual studies observed relatively few significant differences between treatment groups. Clinical trials on robotic-assisted gait training for stroke patients were at different stages in research, based on the assistive device in question. Future clinical trials should focus on cost-effectiveness of RAGT, proper intervention dosage, and personalization of RAGT based on stroke subject characteristics.
Nowadays there is an increasing percentage of elderly people. It is expected that this percentage will continue increasing, carrying huge organizational costs in rehabilitation services. Recent robotic devices for gait training are more and more regarded as alternatives to solve cost-efficiency issues and provide innovative approaches for training. However, some control strategies implemented in current robotic devices should be improved and validated. There is a need to address how to target muscular activation and kinematic patterns for optimal recovery after a neurological damage. The main objective of this work was to understand the underlying principals that the human central nervous system employs to synchronize muscular activity during walking assisted by a robotic gait trainer. Our tested hypothesis was that a basic low-dimensional locomotor program can explain the synergistic activation of muscles during assisted gait. As a main contribution, we generated a detailed description of the electromyographic and biomechanical response to variations in robotic assistance in intact humans, which can be used for future control strategies to be implemented in motor rehabilitation.
The first finding, that providing an intermediate amount of freedom to the pelvis can interfere with gait training, suggests that future devices should have very high amounts of freedom or very restricted pelvic motions. The final finding, that damping fields can be used to induce gait adaptations using a much lower force, can drastically change exoskeleton design and how robotic therapy is provided. Exoskeletons can be made lighter as a result of the force being highly reduced so that lighter weight components can be used, and the dissipative nature of the force reduces dependence on heavy power sources because regenerative breaking can be used to power the device. These factors also make it possible to for devices to be used overground, which may make training more transferable to the real world.
Stroke is one of the leading causes of physical disability around the world, resulting in significant motor deficits and ultimately reduced gait performance. In order to regain or improve motor function, stroke patients will undergo rehabilitation. Current rehabilitation techniques are labour-intensive and time-consuming for both the therapist and the patient. Therefore, not all patients receive the recommended quantity and quality of rehabilitation. To improve the gait rehabilitation process, robot-assisted gait rehabilitation has been receiving increased interest over the past few years. The research in this thesis aims to demonstrate feasibility of robotic gait rehabilitation devices which are designed to be easy to use, simple and compact. The Linkage-based Gait Trainer (LGT) was the first device developed and is an overground gait rehabilitation device consisting of a four-bar linkage end-effector mechanism. The literature review suggested the design could be an effective gait rehabilitation device while minimising the components required. The design of the LGT was validated with a healthy human participant, and kinematic data showed that the device successfully constrained the user’s foot motion to a pre-defined gait trajectory. The imposed trajectory of the LGT is set by the linkage design and cannot be changed. To allow for modification to the trajectory a subsequent device (robotic Re-Link Trainer [rRLT]) was developed. The rRLT also uses a four-bar linkage, however the spatiotemporal parameters of gait, such as walking speed, stride length and cadence, can be varied to accommodate a wider range of users with different limb lengths and desired walking speeds. A trial with healthy participants was carried out to validate the rRLT’s design and its function. The results confirmed the device’s ability to vary the spatiotemporal parameter of the gait pattern the device imposes on the user. One of the limitations of both the LGT and rRLT is that the constrained gait path is not variable. To make the device available for more users with different gait trajectories, a new gait rehabilitation device with a five-bar linkage called PRO-GaiT was developed. A trial with healthy participants confirmed the ability to change and constrain different gait trajectories during walking exercises. The research in this thesis confirmed the ability of a linkage-based overground gait rehabilitation device to constrain a user’s footpath to a desired trajectory. Such devices can vary the footpath and spatiotemporal parameters of gait to accommodate many different users. The devices developed in this research could improve the effectiveness of the gait rehabilitation by potentially increasing the time and intensity of each session while reducing the number of staff required. A rehabilitation device based on this research could be compact, easy to use and low cost thus can be widely adopted by many hospitals and clinics.