Implementation of motion capture system for trajectory planning of leg swing simulator

摘要

Marker-based motion capture systems have increased interest in various fields such as animation, automation, robotics, and simulation. In this paper, we present the motion capture system for trajectory planning of the leg swing simulator. First, the camera captured the identifier that mounted on the human leg. Identifiers as markers are placed to be tracked as joints position. Camera is tracked for the position movement in coordinates, and then stored in the database. The coordinate data of joint position would form a series of movements, for instance the movement of human leg motion. The movement of marker is used to characterize the walking cycle. To plan a cycle of leg swing trajectory planning, the authors propose it by using the joint space trajectory planning. The trajectory method is applied to leg swing simulator model. It is a robot leg with a pivotal position covering the three joints: hip, knee, and ankle. The results present the regression process of system to track the desired trajectories of leg joints in one walking cycle that consists of 15 steps by using leg swing simulator actuators. The largest angle performed on the hip joint is 41°. At the knee joint, the movements from step 1 to 9 of a leg produce the smallest angular value during the heel rise from the floor. The angle value of the ankle is controlled by a trajectory of joint position that approaches 90°. From the desired trajectory comparison between the joint angle and the controlled joint, the results indicate that trajectory error is caused by the simulator actuator joint position.