本論文主要闡述在車載系統當中，建立多代理人（Multi-Agents）間之合作互動模式，將多方面的訊息傳遞利用Multi-Agents來進行資料交換，並透過所建構的中介軟體（Middleware）來進行管控，使其運用在車聯網(Internet of Vehicles，IOV）之中，並且整合各車輛之車載裝置彼此間的溝通。在整體架構上，同時可利用中介軟體來有效處理(Vehicle to Vehicle ,V2V)及(vehicle to infrastructure ,V2I)，對各介面間以及車載系統上的資源取用的管控與限制，並將所有的訊息與服務整合於一個車輛上之中央控制平台(On-Board Unit，OBU)上，同時可利用車上所收集到的各種資訊，整合且即時地提供給駕駛人，來增加且更充分的判斷行車安全與駕駛資訊參考。

In this paper, a collaborative interaction mechanism is designed for multiple agents in an on-board unit so that all of the agents can be leveraged to transmit messages from various sources and exchange information; additionally, middleware is developed to manage this mechanism’s application in an Internet of Vehicles (IOV) and to integrate communication among on-board units in vehicles. In the overall architecture, the middleware is responsible for effective vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) interaction, resource access management and control between interfaces and on-board units, integration of all of the information and services in a vehicle’s on-board unit (OBU), and synthesis of the various pieces of information collected by the vehicle for prompt presentation to the driver. In this way, the driver can obtain comprehensive information for assessing driving safety and for reference.

第一章	緒論	1
1.1	 研究動機與目的	1
1.2	 論文架構	3
1.3	 相關研究	4
第二章	相關文獻探討	7
2.1	車聯網與物聯網	7
2.2	中介軟體	10
2.3	多代理人系統	11
2.4	智慧型代理人系統	12
2.5	智慧代理人的分類	13
2.6	Agent 重要交談類型語言KQML	18
2.7	全球衛星定位系統	26
第三章	本研究所提之系統	29
3.1	多代理人合作機制	29
3.2	宣告代理人	32
3.3	狀態代理人	36
3.4	監聽代理人	39
3.5	接受代理	42
3.6	回報代理	46
3.7	資源管理 (RM)	48
第四章	多代理人系統環境中之溝通機制	50
4.1	交互式中介軟體管理	50
4.2	車輛通訊管理（VCM）	51
4.3	車輛訊息管理（VIM）	53
4.4	廣播管理（ANM）	54
4.5	車輛導航管理 (VNM)	56
4.6	車聯網多代理人合作機制	57
4.7	V2I交通燈號互動通知狀態	58
4.8	V2I當地路況回報互動預警功能	62
4.9	V2V 協同合作以及車載預警功能	65
第五章	結論與未來研究	69
參考文獻	71
附錄-英文論文	77

圖目錄
圖 1   OBU上之MIDDLEWARE架構示意圖	5
圖 2  物聯網的應用領域	8
圖 3  是說明中介軟體上下層架構圖	10
圖 4  KQML語言的關係	20
圖 5  MULTI AGENT系統架構關聯表	29
圖 6  MULTI AGENT SYSTEM狀態概念圖。	30
圖 7  DECLARE AGENT溝通示意圖	32
圖 8  由DA宣告啟動所有AGENT溝通連動狀態	33
圖 9  STATUS AGENT狀態代人理概念圖	36
圖 10  MONITOR AGENT以及STATUS AGENT循環互動示意圖	39
圖 11  MONITOR AGENT監督OBU MIDDLEWARE	40
圖 12  NETWORK收到V2V AND V2I概念圖	42
圖 13  各AGENT透過KQML宣告等相互溝通運作等對應位置。	43
圖 14  回報代理程式結構的概念	46
圖 15 SENSOR感應器位置概念圖。	48
圖 16 通訊連接互動概念圖	51
圖 17 VIM對其他模組溝通傳遞概念圖	53
圖 18 廣播訊息通告V2V、V2I方式廣播概念圖	54
圖 19 車輛導航管理搭配V2V、V2I整體通訊概念圖	56
圖 20 結合三項主要概念功能狀態圖	57
圖 21  V2I等相互溝通狀態圖	58
圖 22  V2I以及V2V車輛預警狀態圖	62
圖 23   車輛預警狀態圖	63
圖 24   V2V協同合作智慧型車輛預警圖	65
圖 25   V2V車輛預警狀態溝通圖	66

表目錄
表格 1  KQML語言的基本訊息架構	21
表格 2  詳細介紹KQML語法訊息結構等說明	22
表格 3 FIPA ACL 基本的訊息結構	23
表格 4 FIPA ACL 溝通行為類型的描述	24
表格 5 針對DECLARE AGENT 啟動透過KQML宣告片段狀態。	34
表格6 STATUS AGENT透過KQML跟其他AGENT溝通等宣告片段。	37
表格 7 MONITOR AGENT以及STATUS AGENT循環資料傳送片段。	41
表格 8 各AGENT內所運作KQML相關互動程式碼	43
表格 9  對RESOURCE POOL所提取的資料格式	49
表格 10 V2I等相互溝通狀態部分程式等片段	59

[1]	S. Kraus, "Negotiation and cooperation in multi-agent environments," Artificial intelligence, vol. 94, pp. 79-97, 1997.
[2]	S. K. C. Lo, "A collaborative multi-agent message transmission mechanism in intelligent transportation system – A smart freeway example," Information Sciences, vol. 184, pp. 246-265, 2/1/ 2012.
[3]	A. Byrski, R. Dreżewski, L. Siwik, and M. Kisiel-Dorohinicki, "Evolutionary multi-agent systems," The Knowledge Engineering Review, vol. 30, pp. 171-186, 2015.
[4]	J. Hu, A. Saleem, S. You, L. Nordström, M. Lind, and J. Østergaard, "A multi-agent system for distribution grid congestion management with electric vehicles," Engineering Applications of Artificial Intelligence, vol. 38, pp. 45-58, 2// 2015.
[5]	T. Bui and J. Lee, "An agent-based framework for building decision support systems," Decision Support Systems, vol. 25, pp. 225-237, 1999.
[6]	W. Qing, F. Pingyi, and K. B. Letaief, "On the Joint V2I and V2V Scheduling for Cooperative VANETs With Network Coding," Vehicular Technology, IEEE Transactions on, vol. 61, pp. 62-73, 2012.
[7]	I. Sen and D. W. Matolak, "V2V channels and performance of multiuser spread spectrum modulation," Vehicular Technology Magazine, IEEE, vol. 2, pp. 19-25, 2007.
[8]	M. Azimifar, T. Todd, G. Karakostas, and A. Khezrian, "Vehicle-to-Vehicle Forwarding in Green Roadside Infrastructure," IEEE Transactions on Vehicular Technology, pp. 1-1, 2015.
[9]	E. T. Leal, O. Chiotti, and P. D. Villarreal, "Software Agents for Management Dynamic Inter-Organizational Collaborations," Latin America Transactions, IEEE (Revista IEEE America Latina), vol. 12, pp. 330-341, 2014.
[10]	T. Miller, L. Bin, L. Sterling, G. Beydoun, and K. Taveter, "Requirements Elicitation and Specification Using the Agent Paradigm: The Case Study of an Aircraft Turnaround Simulator," Software Engineering, IEEE Transactions on, vol. 40, pp. 1007-1024, 2014.
[11]	V. D. Khairnar and K. Kotecha, "Performance of Vehicle-to-Vehicle Communication using IEEE 802.11p in Vehicular Ad-hoc Network Environment," International Journal of Network Security & Its Applications, vol. 5, pp. 143-170, 2013.
[12]	G. Burnett and T. Horberry, "A reference level for assessing the acceptable visual demand of in-vehicle information systems," Behaviour & Information Technology, vol. 29, pp. 527-540, 2010.
[13]	G. Gehlen and G. Mavromatis, "A Web Service based Middleware for Mobile Vehicular Applications," in Vehicular Technology Conference (VTC), 2006, pp. 1-5.
[14]	H. Dong, X. Zhao, L. Qu, X. Chi, and X. Cui, "Multi-Hop Routing Optimization Method Based on Improved Ant Algorithm for Vehicle to Roadside Network," Journal of Bionic Engineering, vol. 11, pp. 490-496, 7// 2014.
[15]	N. Cordeschi, D. Amendola, and E. Baccarelli, "Reliable Adaptive Resource Management for Cognitive Cloud Vehicular Networks," Vehicular Technology, IEEE Transactions on, vol. 64, pp. 2528-2537, 2015.
[16]	T. Engel, H. Granzer, B. F. Koch, M. Winter, P. Sampatakos, I. S. Venieris, et al., "AQUILA: adaptive resource control for QoS using an IP-based layered architecture," Communications Magazine, IEEE, vol. 41, pp. 46-53, 2003.
[17]	H. Abbott and D. Powell, "Land-vehicle navigation using GPS," Proceedings of the IEEE, vol. 87, pp. 145-162, 1999.
[18]	Z. F. Syed, P. Aggarwal, N. Xiaoji, and N. El-Sheimy, "Civilian Vehicle Navigation: Required Alignment of the Inertial Sensors for Acceptable Navigation Accuracies," Vehicular Technology, IEEE Transactions on, vol. 57, pp. 3402-3412, 2008.
[19]	A. Frei, A. Popovici, and G. Alonso, "Eventizing Applications in an Adaptive Middleware Platform," IEEE Distributed Systems Online, vol. 6, pp. 1-1, 2005.
[20]	S. Vinoski, "The performance presumption [middleware evaluation]," Internet Computing, IEEE, vol. 7, pp. 88-90, 2003.
[21]	K. Donghee, L. Chang-Eun, P. Jun Hee, M. KyeongDeok, and L. Kyungshik, "Scalable message translation mechanism for the environment of heterogeneous middleware," IEEE Transactions on Consumer Electronics, vol. 53, pp. 108-113, 2007.
[22]	T. Finin, J. Weber, G. Wiederhold, M. Genesereth, R. Fritzson, D. McKay, et al., "Specification of the KQML agent-communication language," 1994.
[23]	Y. I. Wijata, D. Niehaus, and V. S. Frost, "A scalable agent-based network measurement infrastructure," Communications Magazine, IEEE, vol. 38, pp. 174-183, 2000.
[24]	G. Milagres, E. Moreira, J. Pimentão, P. Sousa, and A. Garcao, "Dealing with security within DEEPSIA project," WSEAS Transactions on Systems, vol. 2, pp. 444-452, 2003.
[25]	B. Bellalta, E. Belyaev, M. Jonsson, and A. Vinel, "Performance Evaluation of IEEE 802.11p-Enabled Vehicular Video Surveillance System," Communications Letters, IEEE, vol. 18, pp. 708-711, 2014.