Applying the Insights of SLIP Model on Musculoskeletal Bipedal Robot to Interpret Human Walking
Musculoskeletal structure is demonstrated to play an essential role in the versatile adaptivity of human locomotion. In order to figure out the underlying principles of muscle activation during human locomotion, we present a musculoskeletal bipedal robot that aims to replicate human walking based on SLIP model. 1)The mechanism is just designed to meet the passive behaviors of SLIP model. 2)Joint compliance is tuned with the utilization of McKibben pneumatic artificial muscles. 3)A ROS-based control system that provides low-level control for air valves and reads sensor values is adopted. Both the mechanical features and the control concepts dedicate to the realization of desired locomotion.
Development of a Feedback System for Musculoskeletal Robots Inspired by Muscle Reflexes
Some biomechanists implied that human legged system has several principles of legged kinematics which simplify the complex task of locomotion control.
A neuromechanical simulation generated the human walking gait using only force feedback.
We propose a length feedback system to apply the simulation model to musculoskeletal robots driven by pneumatic artificial muscles(PAMs).
This system can be the new approach of feedback system for musculoskeletal robots.
2DOF link mechanism mimicking cheetah's spine and leg movement
A spine of a quadruped animal has compliantly connected several segments. This structure enables the animal to realize bending/stretching motion during running along with the motion of the legs. In this research, we develope a link mechanism for mimicking such bending/stretching motion of the spine and the leg with few degrees of freedom. By designing a linkage mechanism, we try to realize a similar motion of the spine and the hind leg as that of a cheetah.