A Bioinspired Approach to Bipedal Running Control: Incorporating Appendage Inertia and Lower Leg Stiffness

In this research, a bio-inspired control framework based on the biomechanics behavior of human running is proposed. The proposed control framework integrates three key components: vertical oscillation control , landing posture adjustment , and take-off state adjustment . Each of these components incorporates a task-specialized controller, namely: Norm Regulation Control (NRC) for vertical oscillation control, Null-Space Avoidance Control (NSAC) for landing posture adjustment, and angular momentum control for take-off state adjustment. This comprehensive approach enables human- like robust running behavior by effectively managing the transition between the flight and stance phases while maintaining the forward moving motion by regulating the forward velocity. A significant contribution of this study is the replication of human- like bipedal running locomotion that explicitly reveals crucial insights into the relationship between swing leg motion and overall system stability from the perspective of reducing the system angular momentum. Addressing the substantial force requirements in running, the calves are designed as linear parallel elastic actuators. This facilitates the management of rapid and substantial vertical momentum fluctuations within each stride, thereby reducing the actuator's load. The validity of this approach was verified by a simple bipedal running robot and evaluated through simulations of a 5-link bipedal model, with the resultant running speed reaching a Froude number of 1.23.