Please use this identifier to cite or link to this item:
標題: 建置田間自主載具人員追隨與自動返航性能之研究
A Study on Implementation of Performances of Human Following and Trace Back for Autonomous Field Vehicle
作者: 陳姿樺
Tzu-Hua Chen
關鍵字: 粒子群演算法
Particle Swarm Optimization
Fuzzy Logic Controller
Human Following
引用: [1] 楊智凱、施瑩艷、楊舒涵(民105)。以智慧科技邁向臺灣農業4.0時代。檢自 (Oct. 15, 2017)。 [2] 蔡伯萱。2016。機器人基於模糊理論與基因演算法於目標追隨之研究。碩士論文。台中:國立中興大學生物產業機電工程學系。 [3] Limitless IQ. com智慧機器人網。影/法國研發除草機器人Dino  翻土、播種、除草樣樣通。 檢自 (Aug. 15, 2018)。 [4] 林信成(民101)。模糊集合理論。檢自 (Jan. 12, 2018)。 [5] 羅茂榮。2014。模糊理論應用於自主性機器人避障路徑導引之研究,碩士論文。台中:國立中興大學生物產業機電工程學系。 [6] 華人百科。粒子群演算法。檢自 (Fab. 8, 2018)。 [7] J. G. Lee, M. S. Kim, T. M. Hwang and S. J. Kang, 'A mobile robot which can follow and lead human by detecting user Location and behavior with wearable devices,' IEEE International Conference on Consumer Electronics (ICCE), pp. 209-210, 2016. [8] Y. Yoo, W. Chung, 'Detection and following of human legs using the SVDD (support vector data description) scheme for a mobile robot with a single laser range finder,' International Conference on Electrical, Control and Computer Engineering 2011 (InECCE), pp. 97-102, 2011. [9] C. Jianzhao and T. Matsumaru 'Robot human-following limited speed control,' 2013 IEEE Ro-Man, pp. 81-86, 2013. [10] C. S. Chiu, T. S. Chiang and Y. T. Ye, 'Fuzzy obstacle avoidance control of a two-wheeled mobile robot,' International Automatic Control Conference (CACS), pp. 1-6, 2015. [11] J. Kennedy and R. Eberart, 'Particle swarm optimization,' Proceedings of ICNN'95 - International Conference on Neural Networks, pp. 1942-1948, 1995. [12] R. Bibi, B. S. Chowdhry and R. A. Shah, 'PSO based localization of multiple mobile robots employing LEGO EV3,' 2018 International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), pp. 1-5, 2018. [13] J. J. Yu, Y. F. Sun, X. G. Ruan and Y. K. Zhang, 'Research on path planning for robots based on PSO optimization for fuzzy controller,' Proceeding of the 11th World Congress on Intelligent Control and Automation, pp. 5293 – 5298, Shenyang, China, 2014. [14] Hokuyo Automatic co., LTD. Scanning laser range finder URG-04LX-UG01(Simple-URG). Internet:, Aug. 2009 [Nov 30, 2017]. [15] STMicroelectronics. L298N dual full-bridge driver. Internet:, Jan. 2000 [Nov. 30, 2017].
摘要: 隨著人工智慧的蓬勃發展,機械化與自動化的基調早已根深於各項產業之中,智能機器人的應用與輔助人的概念推陳出新,世界爭相開發新的產品以提高工作效率。在農業生產中,則有自動施藥、溫室環控自動化......等研究持續進行中。本論文則著力於開發出擁有跟隨能力與自動返航功能之無人載具系統,協助農民採收時搬運作業之勞力需求,緩解農業勞動力不足問題。而此系統實現於自行改良打造並命名為L1M2-III型載具機器人,L1M2-III型機器人採用單顆雷射測距儀進行跟隨目標偵測及障礙物位置判斷,當機器人追隨操作者至滿載狀態時,將自行啟動返航路徑自主巡航回歸至集貨或起始位置。系統之前進跟隨控制策略採用模糊粒子群演算法,其模糊規則經機器學習後,將調整歸屬函數以符合使用者習慣並提升巡航之穩定性及準確性。自主巡航返回控制策略則採逆轉倒序馬達激磁步數控制,沿原路徑返航,返航過程中若遇障礙物或偏移時,模糊邏輯控制器將主導避障及方向修正完成任務。本論文以跟隨距離和角度的平均值及變異係數做為跟隨效能評比標準,返程則使用返回時間差及返回原點位移做為檢驗標準。5種不同粒子點設定之驗證結果顯示,通過配置最佳化後粒子群演算法之控制器於各項評比皆有極佳之性能表現。
With the flourishing of artificial intelligence, the concept of mechanization and automation is deeply rooted in industry and agriculture. For improving efficiency, the application of intelligent robots and artificial assistants are renewed rapidly. In agricultural application, the autonomous spraying pesticides system and the intelligent controller of greenhouse are typical achievements. Therefore, for solving the problem of the lack of agricultural labor, this thesis proposes an autonomous vehicle with features of human-following and trace back control system to help transport operation during harvesting, and verify by the reformed L1M2-III robot. This system employed a single laser to detect the trajectory of the gait and determine the obstacle location. As the following robot with full load, the process of trace back along the original path started. In control of following process, the technique of machine learning to conform the followed habits combined to particle swarm optimization conducted the adjustment of the membership function of the fuzzy laws to achieve the best control strategy. In implementation of trace back, the controller operated in a reverse sequence and the fuzzy logic controller managed the obstacle avoidance. For assessment, the mean of the distance and angle between human and vehicle and the coefficient of variation of the distance were chosen as the parameters of effectiveness of the following performance. The difference of round trip time and the distinct displacement between the start and return points were evaluated in the trace back voyage. Five different particle field checks validate the excellent performance of PSO fuzzy controller in human following and trace back.
文章公開時間: 2021-08-27
Appears in Collections:生物產業機電工程學系



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.