應用類免疫網路於行動機器人巡航

MOBILE ROBOT NAVIGATION USING ARTIFICIAL IMMUNE NETWORK

劉偉文(Wei-Wen Liu)

指導教授 ： 陸冠群

大同大學/工程學院/機械工程學系所/博士(2008年)

摘要

行動機器人之相關應用領域極為廣泛：工業、醫院、辦公室甚至軍隊，為了讓這些機器人在未知的複雜環境中做出適當的行為，如何演算出目標的追蹤與障礙物的閃避行為是非常重要的工程。
在本篇論文中，首先，提出了反射式類免疫網路(Reactive Immune Network)應用在智能機器人於未知環境中的巡航策略，除了建立一套詳細的類免疫系統數學模型，還探索研究類免疫系統中的自我組織能力、適應學習能力以及免疫的回饋方法，藉由融合改良過的虛擬目標方法來解決行動機器人陷入局部空間解的問題，這些環境都是由先前學者們設計出來的許多陷阱型態的環境。經由模擬與實作的結果驗證得知，行動機器人都可以有效的追蹤目標並且防止碰撞上障礙物。
在靜態障礙物的研究之後，繼續提出了力場免疫網路(Potential Field Immune Network)應用在行動機器人的追蹤靜止或移動中的目標與閃躲行動障礙物，利用障礙物速度的方法(Velocity Obstacle)計算行動障礙物和行動機器人是否會發生碰撞，力場免疫網路每一時刻會先讓行動機器人閃躲最危險最有可能優先碰撞到的行動障礙物，整個免疫網路的反應還包含了模糊邏輯和基因演算法。根據模擬的結果可以得知此網路的應用可以讓行動機器人有效的適應於動態的環境當中。

關鍵字

行動機器人 ； 類免疫系統 ； 防碰撞.

並列摘要

Autonomous mobile robots have a wide range of applications in industries, hospitals, offices, and even the military, due to their superior mobility. In order to adapt the robot's behavior to any complex, varying and unknown environment, autonomous robot must be able to maneuver effectively in its environment and to achieve its goals while avoiding collisions with static and moving obstacles.
In this dissertation, first, a Reactive Immune Network (RIN) is proposed and employed to intelligent mobile robot navigation strategies within unknown environments. Rather than building a detailed mathematical model of immune systems, we try to explore the principle in immune network focusing on its self-organization, adaptive learning capability and immune feedback. A modified virtual target method is integrated to solve the local minima problem. Several trap situations designed by early researchers are adopted to evaluate the performance of the proposed immunized architecture. Simulation and experiments results show that the mobile robot is capable of avoiding obstacles, escaping traps, and reaching goal efficiently and effectively.
After static obstacles’ research, continue proposing a potential field immune network (PFIN) for dynamic motion planning of mobile robots in an unknown environment with moving obstacles and fixed/moving goal. The Velocity Obstacle method is utilized to determine the imminent collision obstacle. Subsequently, PFIN is implemented to guide the robot avoid collision with the most danger object at every time instant. The overall response of the immune network is calculated by the aid of fuzzy logic and genetic algorithms. Simulation results are presented to verify the effectiveness of the proposed architecture in dynamic environments with single and multiple fix/moving obstacles.

並列關鍵字

Mobile robot ； Artificial immune system ； Obstacle avoidance.

參考文獻

Borenstein, J. and Koren, Y. (1989). Real-time obstacle avoidance for fast mobile robots. IEEE Transactions on Systems, Man, and Cybernetics, Vol. 19, No. 5, pp.1179–1187.

Canham, R., Jackson, A.H. and Tyrrell, A. (2003). Robot error detection using an artificial immune system, 2003. Proceedings. NASA/DoD Conference on Evolvable Hardware, (9-11, July 2003), 199 – 207.

Carneiro, J.; Coutinho, A.; Faro, J. and Stewart, J. (1996). A model of the immune network with B-T cell co-operation I-prototypical structures and dynamics, Journal of theoretical Biological, Vol.182, No.4, 1996, pp. 513-529.

Chakravarthy, A. and Ghose, D. (1998). Obstacle Avoidance in a Dynamic Environment: A Collision Cone Approach, IEEE Transactions on Systems, Man, and Cybernetics—Part A: Systems and Humans, Vol.25, No.5, 1998, pp. 562-574.

Chang, H. (1996). A new technique to handle local minima for imperfect potential field based motion planning, Proceedings of the IEEE International Conference on Robotics and Automation, pp. 108-112, Minneapolis, Minnesota, April, 1996.