人型機器人於電動機車之平衡與轉向控制之應用

Abstract

none

In recently, humanoid robots and autonomous vehicles are two famous and challenging fields. Many people focus on automation and intelligence, and desire to apply their expectations for the future to reality. From the perspective of general-purpose robots, the humanoid robots are capable of naturally operating in any real environment. That includes the humanoid robot operates the vehicles, which is an interesting challenge for the state-of-the-art in both fields.The contribution of this thesis is operating a two-wheeled electric scooter in reality by using large sized humanoid robot. The inverses kinematic for operating the two-wheeled electric scooter in large sized humanoid robot is applied in four different inverse kinematic method and find the optimal solution to operating the steering motion. The different inverse kinematic includes Jacobian pseudo inverse kinematic, PSO inverse kinematic, hybrid inverse kinematic using Jacobian pseudo and PSO, and the Jacobian pseudo inverse kinematic with momentum. By using hybrid inverse kinematic method, a high efficiency motion result can be obtained, and combines the low computational time of iteration method, Jacobian pseudo inverse kinematic with momentum, we successfully have a fast inverse kinematic in an efficient motion planning.To adapt the robot-scoot system, a 3D model for throttle and emergency brake system was developed, and implement a PID controller from physics-based simulation environments in previous work [34] to reality. Therefore, the robot-scoot system is applied fast inverse kinematic and PID controller for the balancing test. The result of steering motion is under the restrictions from our robot servo.