臺灣博碩士論文加值系統

本論文提出一種在未知環境中攜帶物體的多台合作型機器人導航的新方法。在導航的過程中，多台合作型機器人同時執行跟隨障礙物邊緣或尋標的行為而到達目標。使用數據驅動的進階多目標連續蟻群優化演算法 (AMO-FCACO) 來執行兩/三台合作型機器人的進化模糊控制。模糊控制器 (FC) 中所有的自由參數皆藉由AMO-FCACO學習，避免了耗時的手動設計任務。AMO-FCACO設計的模糊控制器首先應用於在訓練環境中控制單台機器人的跟隨障礙物邊緣行為，其中在訓練環境中包含了內直角與外直角的地形。然後應用學習方法於解決兩/三台合作型機器人跟隨障礙物邊緣行為的問題，其中AMO-FCACO用來設計其他機器人的互補模糊控制器。AMO-FCACO的效能是透過比較各個基於群體為基礎的優化演算法於合作沿牆走行為來驗證學習的好壞。在三台合作型機器人控制的問題中，提出了使用梯度下降法 (Gradient Decent) 與加強式模糊Q學習法 (Fuzzy Q-learning) 的兩種線上學習方法來更新AMO-FCACO設計的FC的參數讓第三台機器人能追蹤期望的軌跡。在合作尋標行為上，本論文提出了一種用於協調兩/三台機器人的規則。在導航中，論文提出了合作行為的管理者來整合合作跟隨障礙物邊緣行為和合作尋標的行為，其中考慮並避免了死循環的問題。兩/三台合作型機器人成功的在幾個複雜的環境中攜帶物體，驗證了進化模糊控制與導航的方法。

This thesis proposes a new method for the navigation of multiple wheeled mobile robots cooperatively carrying an object in unknown environments. In the navigation course, multiple robots cooperatively perform either an obstacle-boundary-following (OBF) or a target seeking (TS) behavior to reach a target. Evolutionary fuzzy control of two/three robots in executing the cooperative OBF behavior through a data-driven advanced multi-objective continuous ant colony optimization (AMO-FCACO) is used. The AMO-FCACO is used to learn all unconstrained parameters in a fuzzy controller (FC), which avoids the time-consuming manual design task. The AMO-FCACO-designed FC is first applied to control a single robot for the OBF behavior in a training environment, where training environment includes both inner and outer corners with right angles. The learning approach is then applied to solve the cooperative OBF problem of two/three cooperative robots, where AMO-FCACO is used to design complementary FCs for the other robots. Performance of the AMO-FCACO is verified through comparisons with various population-based optimization algorithms in the cooperative OBF behavior learning problem. In the three cooperative robots control problem, two online learning approaches using gradient descent and fuzzy Q-learning are proposed to update the parameters of an AMO-FCACO-designed FC for the third robot to track a desired trajectory. For the cooperative TS behavior, this thesis proposes a rule for coordination of the two/three robots. A cooperative behavior supervisor is proposed to coordinate the learned cooperative OBF behavior and the cooperative TS behavior in navigation, where the problem of dead cycles is considered. Successful navigation of two/three cooperative mobile robots carrying an object in several complex environments verifies the evolutionary fuzzy control and navigation methods.

摘 要 i
Abstract ii
Contents iii
List of Figures v
List of Tables viii
Chapter 1 Introduction 1
1.1 Literature Review 1
1.1.1 Multi-Objective Evolutionary Fuzzy Systems 1
1.1.2 Multiple Cooperative Wheeled Mobile Robot Navigation 2
1.2 Thesis Organization 3
Chapter 2 Multi-Objective Continuous Ant Colony Optimization for Fuzzy System Design 5
2.1. Fuzzy Controller 5
2.2. Multi-objective Optimization Structure 6
2.3. AMO-FCACO 8
2.3.1 Learning Configuration 8
2.3.2 New Solution Generation 9
2.3.3 Solution Global Updated 11
Chapter 3 Evolutionary Fuzzy Control of Two Cooperative Robots for Obstacle Boundary Following 13
3.1 Evolutionary Fuzzy Control of a Single Robot for OBF 14
3.2 Evolutionary Fuzzy Control of Two Cooperative Mobile Robots for OBF 17
Chapter 4 Evolutionary Fuzzy Control of Three Cooperative Robots for Obstacle Boundary Following 23
4.1 Evolutionary Fuzzy Control of Three Cooperative Mobile Robots for OBF 23
4.2 Online Controller Learning Using Gradient Descent 28
4.3 Online Controller Learning Using Fuzzy Q-Learning 32
4.3.1 Q-learning 32
4.3.2 Fuzzy Q-learning 33
Chapter 5 Navigation of Multiple Cooperative Robots In Unknown Environments 37
5.1 Navigation of Two Mobile Robots in Unknown Environments 37
5.1.1 Cooperative OBF 37
5.1.2 Cooperative TS 38
5.1.3 Cooperative Behavior Supervisor 39
5.1.3.1 Cooperative TS to cooperative OBF behavior 39
5.1.3.2 Cooperative OBF to cooperative TS behavior 42
5.2 Navigation of Three Mobile Robots in Unknown Environments 43
Chapter 6 Simulations 44
6.1 Single Robot Control 44
6.2 Two Cooperative Robots Control 47
6.3 Three Cooperative Robots Control 55
Chapter 7 Experiments 60
Chapter 8 Conclusion 63
Reference 64

[1]S. Nolfi and D. Floreano, Evolutionary Robotics – The Boloogy, Intelligence, and Technology of Self-Organizing Machines, MIT Press, London, England, 2001.
[2]C. F. Juang, “Temporal problems solved by dynamic fuzzy network based on genetic algorithm with variable-length chromosomes,” Fuzzy Sets and Systems, vol. 142, no. 2, pp. 199-219, March 2004.
[3]A. Ratnaweera, S. K. Halgamuge, and H. C. Watson, “Self-organizing hierarchical particle swarm optimizer with time-varying acceleration coefficients,” IEEE Trans. Evol. Comput, vol. 8, no. 3, pp. 240-255, Jun. 2004.
[4]C. F. Juang, “A hybrid of genetic algorithm and particle swarm optimization for recurrent network design,” IEEE Trans. Syst, Man, Cybern. B, Cybern., vol. 34, no. 2, pp. 997-1006, Apr. 2004.
[5]F. J. Lin, L. T. Teng, J. W. Lin, and S. Y. Chen, “Recurrent functional-link-based fuzzy-neural-network-controlled induction-generator system using improved particle swarm optimization,” IEEE Trans. Ind. Electron, vol. 56, no. 5, pp. 1557-1577, May 2009.
[6]W. Jatmiko, K. Sekiyama, and T. Fukuda, “A pso-based mobile robot for odor source localization in dynamic advection-diffusion with obstacles environment: theory, simulation and measurement,” IEEE Computational Intelligence Magazine, vol. 2, no. 2, pp. 37-51, 2007.
[7]C. F. Juang and Y. C. Chang, “Evolutionary group-based particle swarm-optimized fuzzy controller with application to mobile robot navigation in unknown environments,” IEEE Trans. Fuzzy Systems, vol. 19, no. 2, pp. 379-392, April 2011.
[8]D. W. Gong, Y. Zhang, and C. L. Qi, “Localising odour source using multi-robot and anemotaxis-based particle swarm optimization,” IET Control Theory & Applications, vol. 6, no. 11, pp. 1661-1670, 2012.
[9]C. F. Juang, Y. C. Chang, and C. M. Hsiao, “Evolving gaits of a hexapod robot by recurrent neural networks with symbiotic species-based particle swarm optimization,” IEEE Trans. Industrial Electronics, vol. 58, no.7, pp. 3110-3119, July 2011.
[10]A. Ghosh, A. Ghosh, A. Konar, and R. Janarthanan, “Multi-robot cooperative box-pushing problem using multi-objective Particle Swarm Optimization technique,” Proc. 2012 World Cong. Information and Communication Technologies, pp. 272-277, 2012.
[11]T. W. Manikas, K. Ashenayi, and L. Wainwright, “Genetic algorithms for autonomous robot navigation,” IEEE Instrumentation & Measurement Magazine, vol. 10, no. 6, pp. 26-31, 2007.
[12]G. S. Tewolde, and W. Sheng “Robot path integration in manufacturing processes: Genetic algorithm versus ant colony optimization,” IEEE Trans. Syst., Man and Cyber., Part A: Syst. and Humans, vol. 38, no. 2, pp. 278-287, 2008.
[13]J. H. Kim, Y. H. Kim, S. H. Choi, and I. W. Park, “Evolutionary multi-objective optimization in robot soccer system for education,” IEEE Computational Intelligence Magazine, vol. 4, no. 1, pp. 31-41, 2009.
[14]M. Mucientes and J. Casillas, “Quick design of fuzzy controllers with good interpretability in mobile robotics,” IEEE Trans. Fuzzy Systems, vol. 15, no. 4, pp. 636-651, Aug. 2007.
[15]C. F. Juang and C. H. Hsu, “Reinforcement ant optimized fuzzy controller for mobile-robot wall-following control,” IEEE Trans. Industrial Electronics, vol. 56, no. 10, pp. 3931-3940, Oct. 2009.
[16]C. H. Hsu and C. F. Juang, “Evolutionary robot wall-following control using type-2 fuzzy controller with species-DE activated continuous ACO,” IEEE Trans. Fuzzy Systems, vol. 21, no. 1, pp. 100-112, Feb. 2013.
[17]F. Cupertino, V. Giordano, D. Naso and L. Delfine, “Fuzzy control of a mobile robot,” IEEE Robot Autom. Mag., vol. 13, no. 4, pp. 74-81, Dec. 2006.
[18]C. F. Juang; Y. H. Jhan; Y. M. Chen; C. M. Hsu, "Evolutionary wall-Following hexapod robot using advanced multi-objective continuous ant colony optimized fuzzy controller," in IEEE Transactions on Cognitive and Developmental Systems , vol.PP, no.99, pp.1-1, March. 2017.
[19]M. Udomkun and P. Tangamchit, "Cooperative overhead transportation of a box by decentralized mobile robots," Proc. IEEE Int. Conf. Robotics Automation and Mechatronics, no. 21-24, pp. 1161-1161, Sept. 2008.
[20]C. F. Juang, M. G. Lai, and W. T. Zeng, “Evolutionary fuzzy control and navigation for two wheeled robots cooperatively carrying an object in unknown environments,” IEEE Trans. Cybernetics, vol. 45, no. 9, pp. 1731-1743, Sep. 2015.
[21]D. T. Pham and M. H. Awadalla, “Neuro-fuzzy based adaptive co-operative mobile robots,” Proc. IEEE Annual Conf. Industrial Electronics Society, vol. 4, no. 5-8, pp. 2962 - 2967, Nov. 2002.
[22]A. Yamashita, T. Arai, J. Ota, and H. Asama, “Motion planning of multiple mobile robots for cooperative manipulation and transportation,” IEEE Trans. Robotics and Automation, vol. 19, no. 2, pp. 223- 237, Apr. 2003.
[23]Y. Tohyama and H. Igarashi, “Cooperative transportation by multi-robots with selecting leader,” IEEE Conf. Industrial Electronics, no. 3-5, pp. 4179- 4184, Nov. 2009.
[24]Q. Lv, T. Hu, S. Qiao, Y. Sun, J. Huangfu, and L. Ran, "Non-contact detection of Doppler bio-signals based on gradient decent and extended DACM algorithms," 2013 IEEE MTT-S International Microwave Workshop Series on RF and Wireless Technologies for Biomedical and Healthcare Applications (IMWS-BIO), Singapore, 2013, pp. 1-3, 2013.
[25]C. F. Juang and C. M. Lu, “Ant colony optimization incorporated with fuzzy Q-learning for reinforcement fuzzy control,” IEEE Trans. Syst., Man, and Cyber., Part A: Systems and Humans, vol. 39, no. 3, pp. 597-608, May 2009.
[26]C. C. Phiri, Z. Ju, N. Kubota, and H. Liu, "Enhanced robot learning using fuzzy Q-Learning & context-aware middleware," 2016 International Symposium on Micro-Nano Mechatronics and Human Science (MHS), Nagoya, 2016, pp. 1-8, 2016.