Simulating a Flexible Robotic System based on Musculoskeletal Modeling

Humanoid robotics offers a unique research tool for understanding the human brain and body. The synthesis of human motion is a complex procedure that involves accurate reconstruction of movement sequences, modeling of musculoskeletal kinematics, dynamics and actuation, and characterization of reliable performance criteria. Many of these processes have much in common with the problems found in robotics research, with the recent advent of complex humanoid systems. This work presents the design and development of a new-generation bipedal robot. Its modeling and simulation has been realized by using an open-source software to create and analyze dynamic simulation of movement: OpenSim. Starting from a study by Fuben He, our model aims to be used as an innovative approach to the study of a such type of robot in which there are series elastic actuators represented by active and passive spring components in series with motors. It has provided of monoarticular and biarticular joint in a very similar manner to human musculoskeletal model. This thesis is only the starting point of a wide range of other possible future works: from the control structure completion and whole-body control application, to imitation learning and reinforcement learning for human locomotion, from motion test on at ground to motion test on rough ground, and obviously the transition from simulation to practice with a real elastic bipedal robot biologically-inspired that can move like a human being