Design, development, and validation of a wrist exoskeleton and related wearable technologies

This thesis aims to provide a contribution in the field of wearable technologies for workers assistance to prevent injuries among industrial workers while performing repetitive manual handling activities. In particular, it focuses on design strategies that lead to the development of a soft wrist exoskeleton (FleXo Wrist) and a novel actuation system (Xomino) that can be integrated into new generation of quasi-passive wearable robot and exoskeletons to absorb energy and relief human body from bio-mechanical overloads on muscles and skeletal apparatus. As presented in Chapter 2, the research starts by identifying key design principles for the development of exoskeleton systems, highlighting the complexities of replicating human wrist anatomy, and offering practical insights. An in-depth analysis of wearable wrist exoskeleton devices and related current technologies that find application in both industrial and health care settings, either as research prototypes or commercialized products, was conducted, and a classification was provided by leveraging the state-of-the-art. The essence of this thesis is embodied in Chapters 3, 4 and 5, starting with the evaluation of ergonomics in the workplace with a use case scenario, to the development of a wrist exoskeleton that may satisfy workers' needs, and ending with the design of a new actuation system that can be embodied in quasi-passive soft exoskeletons. Since an assessment of the working environment and activity is necessary to design these devices effectively, guaranteeing reliability, compliance, accurate assistance, and safety, and justify the introduction of an assistive device, Chapter 3 describes the most commonly used methods for validating physical and mental fatigue, and proposes a case study to show how to apply these tools for ergonomic risk assessment in a working environment. The results obtained correlate worker-perceived fatigue and electromyography sensor measurements using questionnaires and scales such as NASA Task Load Index, Quick Exposure Check, Visual Hand Pain Mapping and Hand Activity Level (HAL/ACGIH TLV). Results reported that subjects tend to perceive greater fatigue than is actually objectively detected at the level of muscle activity. This difference can be explained by the fact that in addition to physical exertion, also perceive medium to high mental demand, temporal demand, and frustration, which cannot be measured by processing muscle activation signals. Overall, the muscle fatigue to which workers are subjected is medium to high, and they highlight discomfort in their hands and wrists at critical points such as the carpal bones of the wrist, ulna and radius joints, in the thumb, the index and middle fingers. Building on all the requirements discussed in the previous chapters, Chapter 4 goes into the details of designing a wrist exoskeleton that can be effective for occupational use. This chapter discusses two different wrist exoskeletons one rigid and one soft. A preliminary first rigid version of the device (XoWrist) has been designed. It consists in a 6-DoF rigid kinematic chain designed over a 3D scanned hand-forearm to be as comfortable as possible. It would allow the execution of the so-called Dart-Throwing-Motion, as it is considered the most natural wrist movement during any manual handling task. However, since the acceptance rate of this device among workers is very low because of its stiffness (FleXoWrist), we decided to design a soft wrist exoskeleton that could better meet users' needs. FleXoWrist is based on a glove. Tendon mechanisms, actuated remotely via DC motors, will be the first to be tested as they offer more reliable, robust, lightweight, safe, and cost-effective solutions. For sensing, position and force sensors will be used to detect wrist movements and applied forces/torques. In terms of control strategies, advanced machine learning algorithms for the Assist-As-Needed will be implemented. These algorithms detect motion intentions and provide assistance based on the wrist's position and overload. The soft exoskeleton have been assessed on a test-bench which consist of a rigid plastic mannequin with a 3D printed custom-made soft hand and a flexible passive wrist made with a soft rubber coupling. The device allows main relevant movements of the wrist/hand: flexion/extension (range ), radial/ulnar deviation (range) and grasping. Experimental results show that the device is effective in safely satisfying the RoM requirements of the workers. As an alternative to passive or active device, Chapter 5 presents Xomino, a one-way self-locking clutched actuator specifically designed for soft textile-based materials. The mechanism consists of a series of pivoted plates with rectangular apertures that grip and release a webbing strap using a push-pull effect of a miniature linear motor (Actuonix PQ-12) and controlled via an Arduino-based system using PWM signals. Xomino is fast (engages/releases in less than 1 s), lightweight (70 g), compact, low power (12 V), low-cost, and easy to fabricate (3D-printed ABS components). Tensile tests were conducted to evaluate performance: a static testing demonstrates that the clutch holds up to 250 N without failure; a cyclic testing shows repeatability and stability over 15 loading/unloading cycles up to 200 N, repeated twice. Moreover, once engaged, the clutch's gripping force increases with external load, ensuring secure holding up to 350 N before slipping occurs. Unlike many passive clutches, Xomino can be disengaged under tension (up to 75N), which is a crucial safety feature. Looking at its characteristics, Xomino is well positioned in the landscape of existing clutched actuators. Further works include improvements of both FleXoWrist and Xomino. In particular, the wrist exoskeleton will be tested with real subjects by using a closed-loop control strategy based on force exerted and EMG signals amplitude, and an optimization will be performed in terms of design and aesthetics. Moreover advancements will include a detailed evaluation of the dynamics and structural analysis of Xomino, along with an estimation of elastic bands stiffness, system energy, the integration into XoSoft lower limb exoskeleton. Furthermore, new materials and manufacturing methods will be investigated to enhance clutch load capacity.