Simulating human walking to virtually develop and test new methods for locomotion assistance

Walking is a fundamental human activity that contributes to physical and mental well-being. However, physical impairments and gait disorders can significantly hinder walking ability, impacting an individual’s quality of life and social participation. Developing innovative technologies, such as prostheses, active orthoses, and exoskeletons, to enhance locomotion is crucial. The design of effective assistive control strategies is a critical factor for the proper functioning of these devices. Virtual simulations provide a valuable approach to test control strategies before human experiments, offering time-saving and safer conditions. This thesis aims to address these challenges by replicating and implementing Geyer’s walking model in a widely accessible programming language and the Robotran platform. The thesis follows a three-stage approach: a theoretical understanding of the 2 dimensions human locomotion model, implementing it using Robotran, and validating the results against Geyer’s Simulink model. The findings from the 2D Robotran model demonstrate human walking dynamics, closely resembling Simulink simulation results for an entire left foot gait cycle. The results encompass kinematics, muscle actuation, and neural control, providing conclusive evidence. Limitations and areas for improvement are explored in the context of the contact model. The thesis concludes by highlighting the significant findings and the potential for virtual testing of locomotion assistance methods. The model developed represents a step forward in bio-mechanical research, laying the foundations for future investigations into the bio-mechanics of human walking.