기하학적 알고리듬과 확률론적 로드맵을 이용한 여유 자유도 로봇의 경로 계획

Abstract

In this thesis we present the path planning of the redundant robot to minimize a moving distance using geometric algorithm and probabilistic roadmap methods. We use techniques and notation from the theory of Lie group, Lie algebra for equation of motion and Probabilistic Roadmaps. Our primary focus is on the path planning considering collision-free paths for a manipulator with seven revolute joints and dynamic analysis of open kinematic robotic systems. We also derive equations of motion using formulation and application of an optimal algorithm for the redundant robot motion taking into account the dynamics and collision avoidance. We first derive the kinematics and dynamics modeling of open kinematic chain systems. We then introduce the pseudoinverse method for inverse kinematics of seven degree of freedom open chain system. We also derive the seven degree of freedom robot model. The generalization of the dynamic modeling method with Lie group and algebra tool is also summarized. The resulting dynamic equations can be expressed in an explicit fashion for applications that require high-level manipulation of the equations of motion, or can be expressed recursively for applications which require computationally efficient algorithms. We finally demonstrate the simulation of seven degree of freedom system which is obstacle avoidance retain many of the desirable features of the open chain formulation.