Development of robotic platform for biomechanical simulation of lower limb support under reduced gravity: design and validation

Abstract

As humanity prepares for long-duration missions to the Moon, Mars, and beyond, the need to understand how altered gravitational environments affect the human body has never been more urgent. One of the primary physiological systems impacted by these conditions is the musculoskeletal system, which undergoes substantial adaptation in reduced gravity, often leading to muscle atrophy, bone density loss, and changes in motor control. These challenges not only affect astronaut performance in space but also pose significant rehabilitation demands upon return to Earth. Consequently, there is a growing demand for advanced experimental platforms that can simulate partial gravity on Earth for the purpose of studying locomotion, muscle activation, and biomechanical adaptation. This thesis addresses that need by presenting the design, development, implementation, and evaluation of a novel dual Stewart platform robotic system specifically engineered to simulate reduced gravity environments for biomechanical experimentation.At the core of the research is a dual parallel manipulator configuration, known as a dual Stewart platform, which provides high-fidelity control over six degrees of freedom for each of its two stages. This setup enables the simulation of complex support and perturbation forces typically experienced during gait in altered gravity. The upper platform serves as the primary interface for subject interaction, capable of supporting a test subject’s lower limbs, while the lower platform is used to simulate ground reaction forces with precise control.A significant contribution of this work is the development of a robust embedded control architec- ture designed to manage the dynamic interaction between the user and the robotic system. The control framework employs a super-twisting sliding-mode control (ST-SMC) algorithm with state-dependent gain adaptation. This approach ensures robust and precise trajectory tracking of the platform’s end- effector, even in external disturbances such as user motion or force feedback. The controller was rig- orously validated through both simulation and experimental trials, demonstrating its superior stability and performance over conventional PID or linear feedback controllers, particularly in highly nonlinear operating conditions. To complement the mechanical and control systems, a multi-channel surface electromyography (sEMG) system was developed and integrated into the platform. This circuit was custom-designed to capture real-time muscle activation signals from multiple lower-limb muscle groups, providing syn- chronized neuromuscular data during locomotion trials. The sEMG system enables high-resolution monitoring of muscle recruitment patterns, allowing researchers to study how gravitational changes af- fect neuromuscular coordination and effort during walking or balance tasks. The integration of sEMG data with motion control feedback creates a powerful experimental tool that bridges the gap between kinematic performance and physiological response.Experimental validation of the complete system was conducted using simulated gait patterns. Preliminary results demonstrated the system’s capacity to reproduce biomechanically plausible motion trajectories and consistent sEMG activation profiles corresponding to expected changes in muscle load and coordination. These findings validate the platform’s functionality as a reliable testbed for studying locomotion under variable gravity conditions.Overall, this thesis presents a novel, multi-disciplinary approach that merges robotics, control theory, biomechanics, and neurophysiology into a single integrated system. The platform has potential applications in astronaut training, rehabilitation engineering, and human performance research.