Design and clinical testing of new control strategies for an hip exoskeleton for individuals with moderate gait impairments.

Gait impairments due to aging and pathological conditions, such as neurological disorders or lower-limb amputation, are among the major causes of restricted life and loss of personal independence. Early rehabilitation provided along a continuum of care from hospital to community is essential to improve recovery outcomes and the quality of life of individuals with movement disabilities. In this scenario, lower-limb exoskeletons are seen as a cutting-edge technology that can support impaired people in activities of daily living and improve rehabilitation programs in ecological conditions. The robotics research community is fostering the growth of the wearable robotics industry, by promoting technology transfer actions and translating research outcomes into market products. Several full lower-limb exoskeletons have already reached the commercial market internationally. However, for individuals who retain walking capacity, such as the majority of stroke survivors (up to 80%), and lower-limb amputees (up to 60%), individuals at the early stage of multiple sclerosis or with incomplete spinal cord injury, these devices may not be the best approach to promote a more functional walking pattern. Lightweight lower-limb exoskeletons which selectively assist some of the lower-limb joints are emerging as promising alternative robotic tools. Despite recent advances, yet, major challenges in designing effective exoskeletal robots relate to realizing natural and synergistic cooperation between the human user and the robot. To accomplish this objective, several aspects of the system must be considered. On one hand, the physical human-robot interface has to guarantee safe and comfortable fitting of the device; on the other hand, the control interface for the human-robot interaction should ensure intuitive movement intention decoding, as well as adaptive and reliable assistance in different locomotion tasks. Within this framework, the aim of this thesis was the design and clinical testing of innovative control strategies for a hip exoskeleton to assist individuals with mild-to-moderate gait impairments. To pursue this goal, the research activity followed two lines of action. Firstly, a novel method for real-time gait phase estimate was developed. Secondly, novel assistive strategies were explored to investigate the effects of hip assistance in specific end-users, i.e., individuals with mild-to-moderate neurological gait disorders and transfemoral amputees. This thesis proposed a novel wavelet-based gait phase estimation method capable of: (i) continuously tracking the user gait phase and (ii) identifying in real time relevant biomechanical gait events (i.e., initial and final foot contact) of physiological and pathological gait patterns. The algorithm uses only hip joint angle signals (measured by encoders onboard the hip exoskeleton), thus avoiding the use of additional sensors that could challenge the overall system usability/acceptability and dependability in out-the-lab applications. The proposed method (patented method WO2022053934) proves that distal events, related to foot–ground interaction can be reliably detected by using hip joint angles. Experimental tests with healthy subjects indicated that the method can robustly adapt to different walking speeds. Experimental tests carried out with transfemoral amputees and post-stroke subjects showed that the proposed method is reliable and accurate also with pathological gait patterns. Furthermore, in this thesis, clinical investigations were designed to explore the effectiveness of hip assistance for two different objectives: (i) to improve the walking economy in transfemoral amputees and (ii) to increase the walking speed and induce the recovery of a physiological gait pattern in individuals with neurological disorders. For the first application, although achieved results showed that tailored energy injection at the residual and intact limbs of transfemoral amputees could be promising, further research is needed for a better understanding of the real benefits in daily-life scenarios. For the second application, the use of the hip exoskeleton was investigated both as a gait trainer and as a mobility extender. In addition, a method for tailoring the hip assistance based on subject-specific gait impairments was developed (patented method WO2022137031). Experimental tests carried out with eighteen post-stroke survivors suggested that the proposed approach is promising to improve their walking performance, paving the way for future clinical investigations.