Improvement and Validation of a Spring-Based Pelvis Module for the Lokomat Exoskeleton: Control, Sensors, and Human-Robot Interaction.

The main objective of this thesis was to improve the interaction between the exoskeleton and the human.   Initially, the Lokomat system did not provide a rigid connection at the pelvis, as it was only connected to the user's legs. This resulted in a lack of synchronization between the exoskeleton user's pelvic movements and the exoskeleton's pelvis plate movements, leading to delays and incongruence.  To address this issue, the focus of the study was to enhance the connection between the user's pelvis and the Lokomat system, enabling rigid and synchronized interaction. The integration of a new design aimed to enhance the system's ability to accurately track and replicate the user's movements without any noticeable delays.  This required a careful assessment of the existing system and the identification of areas for improvement.  The initial phase of my research involved designing and implementing the new connection mechanism, which focused on optimizing the coupling between the robot and the user's body.  The new design incorporated elements such as the C-shaped structure, which provided enhanced stability and improved alignment between the user and the Lokomat. To ensure the effectiveness of the design, testing, and validation of the integrated sensors were conducted. This validation process aimed to verify the accuracy and reliability of the collected data and the overall functionality of the system.  In addition to the design integration, a pelvic controller was also studied and implemented in conjunction with the new C-shaped design.   The experimental phase involved testing the effectiveness of both the pelvic controller and the new design in improving the transparency of the exoskeleton and its capability to track better the user's movements.  A series of experiments were conducted with human participants using the Lokomat system to validate this hypothesis. The participants performed walking tasks under various conditions, including the C-shaped design with the pelvic controller activated and deactivated.  The participants were instructed to perform normal walking movements while their interactions with the robot were monitored and recorded. The primary focus was on assessing the system's ability to track the participants' movements accurately in real-time, additionally, the participants were asked to provide subjective feedback on their experience, including factors such as comfort, ease of use, and the perceived responsiveness of the system.  The analysis aimed to assess the effectiveness of the integrated design in achieving the desired objectives of improved tracking accuracy and enhanced user-robot interaction. By establishing a more robust and efficient connection between the robot and the user, the Lokomat system is expected to deliver enhanced rehabilitation outcomes and improve the overall experience for individuals undergoing robotic-assisted therapy.