Soft Robots for Healthcare Applications: Design, modelling, and control
2: Department of Mechanical Engineering, University of Auckland, Auckland, New Zealand
3: School of Information Engineering, Wuhan University of Technology, Wuhan, China
Robot-assisted healthcare offers benefits for repetitive, intensive and task specific training compared to traditional manual manipulation performed by physiotherapists. However, a majority of existing rehabilitation devices use rigid actuators such as electric motors or hydraulic cylinders which cannot guarantee the safety of patients; novel soft robots combining soft and compliant actuators with stiff skeletons offer a better alternative. This book focuses on the development of these soft robotics for rehabilitation purposes. Topics covered include an introduction to soft robots and the state of the art of their use in healthcare; concept and modelling of a soft rehabilitation actuator - the Peano muscle; design of the reactive Peano muscle; soft wrist rehabilitation robot; development and control of a soft ankle rehabilitation robot (SARR); design, modelling and control strategies of a gait rehabilitation exoskeleton (GAREX); and conclusions and future work. This book presents novel applications of mechatronics to provide better clinical rehabilitation services and new insights into emerging technologies utilized in soft robots for healthcare, and is essential reading for researchers and students working in these and related fields.
Inspec keywords: patient rehabilitation; health care; muscle; medical robotics; gait analysis
Other keywords: soft wrist rehabilitation robot; gait rehabilitation exoskeleton; soft ankle rehabilitation robot; reactive Peano muscle; soft rehabilitation actuator; healthcare applications
Subjects: Textbooks; Robotics; Biomechanics, biorheology, biological fluid dynamics; General and management topics; Patient care and treatment; Biomedical engineering; Biological and medical control systems; General electrical engineering topics
- Book DOI: 10.1049/PBHE014E
- Chapter DOI: 10.1049/PBHE014E
- ISBN: 9781785613111
- e-ISBN: 9781785613128
- Page count: 239
- Format: PDF
-
Front Matter
- + Show details - Hide details
-
p.
(1)
-
1 Introduction
- + Show details - Hide details
-
p.
1
–21
(21)
This chapter introduces background information on healthcare requirements and existing soft healthcare robots, and also identifies the critical issues in developing such robots. The main incentives for the usage of soft robotic devices for healthcare applications can be summarised as: their ability to reduce physical workload of physiotherapists, intrinsic compliance for safe patient-robot interaction, light weight, objective assessment and rehabilitation from built-in sensors, and great potential in making wearable devices. Some typical examples of existing soft robots used in the field of healthcare applications are presented to give readers an intuitive understanding. The key challenges in the design and control of soft healthcare robots are also discussed in the chapter.
-
2 State of the art
- + Show details - Hide details
-
p.
23
–48
(26)
A comprehensive literature review on rehabilitation robots and soft healthcare robots is carried out to identify the key issues. It begins with a survey of existing robotassisted rehabilitation techniques proposed for human assistance and treatment. Three typical types of rehabilitation devices are upper-limb rehabilitation robots, gait rehabilitation exoskeletons and ankle rehabilitation robots. Following it, a variety of existing soft robots are presented, especially for healthcare applications.
-
3 Concept and modelling of a soft rehabilitation actuator: the Peano muscle
- + Show details - Hide details
-
p.
49
–66
(18)
The large, heavy and rigid form of traditional electric, hydraulic and pneumatic actuators is ill-suited to actuating rehabilitation devices worn on the body. Many alternative actuators have a more promising form, but do not share the reliability and availability of their counterparts. We propose the Peano muscle as solution. It is a soft, sheet-like fluid-powered actuator invented in 1959 by Mettam [1] that potentially shares the form, fabrication and materials of everyday clothing. Hence, it can give clothing the ability to move the body, whether for assistance or rehabilitation. As a relatively unexplored actuator, the Peano muscle lacks models that account for its material and geometry properties. We present and validate the MECHanical Approximation Lumped Parameter (MECHALP) model for accurately predicting the static force generation of Peano muscles. Its accuracy and physics-based nature makes it a useful tool that is foundational for the design and control of Peano muscle actuated rehabilitation wearables. This chapter explains the limitations of current actuators compared to the relevant requirements of rehabilitation devices; presents the concept of the Peano muscle as a promising rehabilitation actuator; and describes the MECHALP static model and its validation.
-
4 Design of the reactive Peano muscle
- + Show details - Hide details
-
p.
67
–86
(20)
In this chapter we first explore how sensors are embedded into biological muscles and how researchers have mimicked this principle in various contractile fluid powered muscles. Next, we propose a method for sensing the physical state of the Peano muscle and describe how to fabricate these sensors and embed them into a Peano muscle. Finally, we characterise the resulting Peano muscle's actuation and sensing performance and demonstrate that embedded DE sensors do not compromise its capabilities as a soft actuator for rehabilitation.
-
5 Soft wrist rehabilitation robot
- + Show details - Hide details
-
p.
87
–106
(20)
Ageing society in many countries has led to an increasing number of stroke and cerebral palsy patients who require rehabilitation therapy. Affected wrist joints often show an increased spasticity and stiffness, caused by impairments of the surrounding muscles and tendons. However, the medical devices for wrist joint assessment and rehabilitation are lacking. The goal of this chapter is to develop and control a robotic orthosis to assist the patient's wrist to perform rehabilitation exercise in a compliant way. A one-degree of freedom (DOF) robotic device with parallel mechanism was designed for the wrist joint by utilising pneumatic artificial muscles (PAMs) that are compliant and lightweight. Mechanical design of the wrist orthosis and the corresponding development of pneumatic control system were presented. A model-based pressure close-loop control strategy was implemented for the PAMs in order to track the trajectory in high-performance. Experiments on the orthosis demonstrated that the robot could assist the hand to move along a torque-sensitive trajectory with small errors and the differential forces were also kept stable.
-
6 Development of a soft ankle rehabilitation robot
- + Show details - Hide details
-
p.
107
–132
(26)
This chapter presents the development of an intrinsically compliant soft ankle rehabilitation robot (SARR). Specifically, it involves the robot kinematics and dynamics, the Festo Fluidic Muscles (FFM) modelling, the force distribution that distributes a given robot torque of the task space into individual FFM force of the joint space, the calculation of ankle forces and torques, and the construction of the SARR. These provide the basis for the implementation of advanced interactive training schemes. In general, the SARR devised from two literature reviews has advantages over other rehabilitation devices, including intrinsically compliant actuation system, compatible robot structure with aligned rotation centre between the robot and the human ankle, three degrees of freedom (DOFs) for three-dimensional rehabilitation exercises, and the integration of real-time ankle assessment. By using the SARR, robot-assisted ankle rehabilitation can be safer, more comprehensive and effective.
-
7 Control of a soft ankle rehabilitation robot
- + Show details - Hide details
-
p.
133
–151
(19)
Interaction control is crucial to the soft ankle rehabilitation robot (SARR). This is usually implemented on the basis of a position controller by which accurate trajectory tracking can be obtained to guarantee safe passive training and active training control to promote the patient's participation. The first half of this chapter is dedicated to the development of a cascade position controller with position loop and force loop, which aims to implement smooth and safe robotic passive training on the SARR. The second half of this chapter proposes the active training strategies of generating an adaptive predefined trajectory according to the movement intention of the patient. The proposed adaptive interaction training can be then implemented by integrating it into a virtual reality based video game to make the robotic training process more attractive to the patients.
-
8 Design of a GAit Rehabilitation Exoskeleton
- + Show details - Hide details
-
p.
153
–165
(13)
Pneumatic artificial muscle (PAM) actuated rehabilitation robots have been widely researched, because of PAM's intrinsic compliance and high power-to-weight ratio. Task-specific gait rehabilitation training imposes strict torque, range of motion (ROM) and bandwidth requirements to the robotic exoskeleton design. However, the PAM's nonlinear and hysteresis behaviour, slow pressure dynamics and negative correlation between its force output and contracting length make the development even more challenging. To address such challenges, a new robotic GAit Rehabilitation EXoskeleton (GAREX) has been developed in order to facilitate task-specific gait rehabilitation with controlled intrinsic compliance. GAREX was designed for the experiments with human subjects. Several implementations ensure the safety of the subject, which is of paramount importance. GAREX has modular design to accommodate anthropometrics of most of the population.
-
9 Modelling and control strategies development of GAREX
- + Show details - Hide details
-
p.
167
–206
(40)
This chapter focuses on a multi-input-multi-output (MIMO) sliding mode (SM) controller which is aimed to simultaneously control the angular trajectory and compliance of the GAit Rehabilitation Exoskeleton (GAREX). The MIMO controller is developed based on the complete model of GAREX system which consists of four sub-system models. They are flow dynamics of the valves, pressure and force dynamics of the antagonistic pneumatic artificial muscles, as well as the load dynamics of the exoskeleton. Experiments with and without healthy subjects are conducted a GAREX to validate the developed SM controllers. The experimental results indicate good multivariable tracking performance of this controller.
-
10 Conclusion and future work
- + Show details - Hide details
-
p.
207
–214
(8)
This chapter seeks to summarise the main outcomes and conclusion of the research presented within this book. Various aspects including the design, modelling and control of some soft and compliant rehabilitation robots have been presented and discussed in previous chapters, as well as the modelling and development of the Peano muscle. This chapter also provides a discussion of future work that can be further explored to extend and advance the research presented in this book.
-
Back Matter
- + Show details - Hide details
-
p.
(1)