用於平行夾持之定力撓性機構的最佳設計

Optimal Design of a Constant-Force Compliant Mechanism for Parallel Gripping

羅勁暐(Ching-Wei Lo)

指導教授 ： 李志中

國立臺灣大學/工學院/機械工程學研究所/碩士(2022年)

https://doi.org/10.6342/NTU202200234

摘要

定力撓性機構 (Constant-Force Compliant Mechanism, CFCM) 透過被動的力控制防止機械夾具過度施加接觸力。因結構變形所導致的剛性改變，它可以在整個有效輸入位移範圍內，產生近似不變的外部負載。本研究旨在設計與整合一種CFCM，用於平行夾爪上，以夾取未知幾何與機械性質的物件，而不需安裝任何傳感裝置。實際應用上，CFCM 應表現出更大的運動範圍，以及強健的力穩定性。在設計方法上，我們提出了一種拓樸最佳化方法來合成CFCM，分為五個步驟：離散參數化 (discrete parameterization)、環形檢測 (loop detection)、變形分析 (deformation analysis)、目標域枚舉 (objective domain enumeration) 和函數誤差評估 (function error evaluation)。演算法會通過一個帶有條件的目標函數實現誤差最小化，利用一些懲罰函數推斷方案的可行性，過程標記了所有不可行解為違規誤差 (violation error)，而可行解為功能誤差 (function error)。全面性的分析包含關於CFCM最佳解的效能、變形和應力條件，解答的精度已透過有限元素法驗證。拓樸最佳化的收斂品質，是根據迭代數量測試與族群大小測試來審視的；而定力的調整，是藉由撓性拓樸的列舉完成。CFCM的原型和其末端元件的組合，使用了熔融沉積成型 (fused deposition modeling, FDM) 製造。本研究最後進行了實驗來驗證分析結果，並評估實際抓取效率。本研究亦希望為該領域的其他相關研究奠定基礎。

關鍵字

定力撓性機構 ； 拓樸最佳化 ； 目標域枚舉 ； 基因演算法 ； 掃描線演算法 ； 向量式有限元素法

並列摘要

A constant-force compliant mechanism (CFCM) prevents a mechanical gripper from over exertion of contact force via passive force control. In response to changes of mechanical stiffness as induced from structural deformation, it can generate near invariant extrinsic loads throughout the extent of effective input displacement. This research aims to design an embedded CFCM end effector for parallel manipulators to grasp geometrical and mechanical unknown objects without any sensory equipment. The CFCM manifests an extended range of motion, along with an enhanced force stability. In the methodology, we have proposed a topology optimization method for synthesizing CFCMs via discrete parameterization, loop detection, deformation analysis, objective domain enumeration, and function error evaluation. A genetic algorithm practices error minimization via a conditional objective function that deduces on solution feasibility according to several penalty functions, where it denotes every infeasible result as violation error, and feasible result as function error. An extensive analysis has been organized concerning the optimal CFCM’s performance, deformation and stress conditions. Solution accuracy has been verified via a finite element method. Convergent quality has been quantified according to an iteration quantity test, and a population size test. Modulation of CFCMs has been accomplished through enumeration of models of multiple constant-forces without changing the initial setting. A CFCM prototype, and an end effector assembly have been manufactured from fused deposition modeling. Numerous experiments have been prepared to validate the analytical results, and to appraise for actual grasping efficiency. This article hopes to establish the groundworks for other related researches in this field.

並列關鍵字

Constant-Force Compliant Mechanism ； Topology Optimization ； Objective Domain Enumeration ； Genetic Algorithm ； Sweep Line Algorithm ； Vector Form Intrinsic Finite Element Method

參考文獻

[1] Lins, R. G., Givigi, S. N. (2021). Cooperative Robotics and Machine Learning for Smart Manufacturing: Platform Design and Trends Within the Context of Industrial Internet of Things. IEEE Access, vol. 9, pp. 95444-95455.

[2] Kiguchi, K., Fukuda, T. (2000). Position/Force Control of Robot Manipulators for Geometrically Unknown Objects using Fuzzy Neural Networks. IEEE Transactions on Industrial Electronics, vol. 47, no. 3, pp. 641-649.

[3] Kumar, N., Panwar, V., Sukavanam, N., Sharma, S. P., Borm, J. H. (2011). Neural Network based Hybrid Force/Position Control for Robot Manipulators. International Journal of Precision Engineering and Manufacturing, vol. 12, no. 3, pp. 419-426.

[4] Liu, X., Zhang, T., Li, J., Guan, Y., Liu, G. (2018). A Novel End-effector for Robotic Compliant Polishing. In 2018 International Conference on Robotics and Biomimetics (ROBIO), pp. 1858-1863.

[5] Lin, Y., Chen, Z. A. F. T., Yao, B. (2021). Unified Motion/Force/Impedance Control for Manipulators in Unknown Contact Environments based on Robust Model-Reaching Approach. IEEE/ASME Transactions on Mechatronics, pp. 192-199.