One of the limitations in the development of really soft robotic devices is the development of soft actuators. In recent years, our research group has developed a new flexible shape memory alloy actuator that provides more freedom of movements and a better integration in wearable robots, especially in soft wearable robots. Shape memory alloy wires present characteristics such as force/weight ratio, low weight, and noiseless actuation, which make them an ideal choice in these types of applications. However, the control strategy must take into account its complex dynamics due to thermal phase transformation. Different control approaches based on complex non-linear models and other model-free control methods have been tested on real systems. Some exoskeleton prototypes have been developed, which demonstrate the utility of this actuator and the advantages offered by these flexible actuators to improve the comfort and adaptability of exoskeletons.

The elbow joint is a complex articulation composed of the humeroulnar and humeroradial joints (for flexion-extension movement) and the proximal radioulnar articulation (for pronation-supination movement). During the flexion-extension movement of the elbow joint, the rotation center changes and this articulation cannot be truly represented as a simple hinge joint. The main goal of this project is to design and assemble a medical rehabilitation exoskeleton for the elbow with one degree of freedom for flexion-extension, using the rotation center for proper patient elbow joint articulation. Compared with the current solutions, which align the exoskeleton axis with the elbow axis, this offers an ergonomic physical human-robot interface with a comfortable interaction. The exoskeleton is actuated with shape memory alloy wire-based actuators having minimum rigid parts, for guiding the actuators. Thanks to this unusual actuation system, the proposed exoskeleton is lightweight and has low noise in operation with a simple design 3D-printed structure. Using this exoskeleton, these advantages will improve the medical rehabilitation process of patients that suffered stroke and will influence how their lifestyle will change to recover from these diseases and improve their ability with activities of daily living, thanks to brain plasticity. The exoskeleton can also be used to evaluate the real status of a patient, with stroke and even spinal cord injury, thanks to an elbow movement analysis.

New robotic applications, among others, in medical and related fields, have in recent years boosted research in the development of new actuators in the search for solutions that are lighter and more flexible than conventional actuators. Shape-Memory Alloy (SMA)-based actuators present characteristics that make them an excellent alternative in a wide variety of applications. This paper presents the design, tests (with the control description) and analysis of various configurations of actuators based on SMA wires: flexible SMA actuators, different mechanical design to multiply the displacement and different configurations for actuators with multiple SMA wires. The performance of the actuators has been analyzed using wires of different activation temperatures. The influence of the Bowden sheath of the flexible actuator has been tested, as has the thermal behavior of actuators with several wires. This work has allowed determination of the most effective configuration for the development of a flexible actuator based on SMA, from the point of view of dimensions, efficiency, and work frequency. This type of actuator has been applied in the development of soft robots and light robotic exoskeletons.

A high-level control algorithm capable of generating position and torque references from surface electromyography signals (sEMG) was designed. It was applied to a shape memory alloy (SMA)-actuated exoskeleton used in active rehabilitation therapies for elbow joints. The sEMG signals are filtered and normalized according to data collected online during the first seconds of a therapy session. The control algorithm uses the sEMG signals to promote active participation of patients during the therapy session. In order to generate the reference position pattern with good precision, the sEMG normalized signal is compared with a pressure sensor signal to detect the intention of each movement. The algorithm was tested in simulations and with healthy people for control of an elbow exoskeleton in flexion–extension movements. The results indicate that sEMG signals from elbow muscles, in combination with pressure sensors that measure arm–exoskeleton interaction, can be used as inputs for the control algorithm, which adapts the reference for exoskeleton movements according to a patient’s intention.