雙機械手臂之動態軌跡規劃與控制

Abstract

本文主要是致力於在未知的動態環境下雙機械手臂的安全協作。透過所提出的控制架構，使機械手臂能合作的過程能確保使用者、任務以及本身的安全。本文共分為兩個部分。第一部分：路徑規劃結構，由全局與局部的規劃算法組成。對於全局規劃算法，我們使用RRT-Connect來產生避障的初始軌跡。接下來，局部規劃演算法Timed Elastic Band對於初始軌跡的平滑度與避障進行優化。此外，當障礙物侵入機械手臂的空間並即將造成碰撞時，使用虛擬阻抗控制使機器人平穩地避障。 第二部分：基於阻抗模型的協同控制算法，使機器人可以安全的執行任務，順應未知環境的外力干擾並且在協作的過程中控制雙臂抓取的內力，從而保障用戶、任務和機器人本身的安全，並透過模擬和實驗驗證。

This thesis addresses the safe cooperation of dual manipulators in an unknown dynamic environment. The dual manipulators cooperative process is divided into two parts. First, we propose a path planning structure that consists of both a global and local planning algorithm. For the global planning algorithm, we use the rapidly exploring random tree (RRT)-Connect to obtain an initial trajectory that avoids possible obstacles in the environment. Then, we propose using a local planning algorithm, such as the timed elastic band algorithm, which optimizes the original path by considering its smoothness and the distance from the obstacles and waypoints. Moreover, when the obstacles invade the robots’ critical space, we use virtual impedance control to give the robots a robust, smooth, and consistent reaction for collision avoidance. Second, we propose a cooperative control algorithm based on an impedance model. This allows the robotic manipulators to safely perform tasks, achieve compliance with unknown environmental disturbances, and regulate the internal force in the cooperative process. The proposed safe cooperative multi-robot system can guarantee the safety of users, tasks, and the robots themselves. The system has been justified and demonstrated through simulations and experiments. The results are promising.