本論文主要解決並聯式機械手臂在受拘束環境下的運動控制問題，使得機械手臂除了能跟隨所要的輪廓軌跡，並且能保持與接觸面固定的接觸力。基於多重感測的應用，本文將整合機械視覺、力量感測器和各軸編碼器的感測資訊，使得並聯式機械手臂能從事與工件接觸的精密輪廓軌跡跟隨控制。首先，推導並聯式機械手臂的順向/逆向運動學、Jacobian矩陣、奇異點等幾何特性，再透過機械視覺和影像處理技術，規劃機械手臂無奇異點的運動軌跡。為了解決位置和力量混合控制時的耦合問題，將機械手臂端部與加工曲面接觸的工作空間分成兩個次空間，在機械手臂受限制的方向施以力量控制，另外在不受限制的其餘自由度施以軌跡跟隨控制。根據此解耦合的控制架構，本文發展以區間第二型類神經模糊網路系統（Interval type-2 neural fuzzy network, IT2NFN）為基礎的智慧型混合位置/力量控制器，並推導適應性法則，使能線上調整控制系統的所有參數，克服機械手臂所承受的結構性和非結構性的外部干擾，確保整個控制系統的穩定和收斂。最後，利用所建立的Delta並聯式機械手臂實驗平台，透過實驗驗證本文所提方法的有效性。

The aim of this study was to find a solution to the problem of how to control a parallel robot manipulator in a constrained environment. Parallel robot manipulators must be able to follow the contours of an unknown surface while maintaining a specified contact force. We developed a multisensing technique. We installed a camera on the ceiling of the workspace and mounted a force sensor to the end effector of the manipulator and encoders at each joint of the manipulator. Each joint had three degrees of freedom. Kinematics, the Jacobian matrix, and other geometric characteristics of the robot were derived. Machine vision and image processing techniques were used to plan the machining trajectories without creating any singular points. The vector map of the workspace was divided into two subspaces: one for controlling the contact force and one for controlling the surface motion. According to this decoupled control strategy, we developed an intelligent hybrid position/force controller for the counter trajectory tracking of parallel robots. This controller is based on an interval type-2 neural fuzzy network (IT2NFN). Weights were adjusted using the adaptive IT2NFN-based control system, which ensured that the system was stable and converged to the expected result. The effectiveness of the proposed model was validated through experimentation on a practical delta parallel robot.