本論文主要探討多自走車跟隨系統，一主車使用攝影模組做影像處理跟隨標的物而兩台從車利用紅外線分別去追隨前車以達到多車跟隨之目的。主車在硬體上分為兩大部分，以Raspberry Pi和攝影模組做影像處理，及以Arduino做馬達控制與紅外線信標兩部分，兩者間使用UART進行溝通，將影像處理得到的目標物資訊做PID控制將其輸出給Arduino以PWM控制車輪轉速，而紅外線信標將作為標的物使從車得以進行跟隨。從車方面使用多個紅外線接收器判斷信標位置並且使用距離感測器判斷距離，最後將各資訊以I2C傳給Arduino進行自走車的跟隨。

This thesis mainly discusses the following system of multiple autonomous cars. The main autonomous car uses the camera module to do image processing to follow the target, and two slave autonomous cars use infrared sensors to follow the car in front to achieve the purpose of multiple autonomous cars following. The main car is divided into two parts in hardware. Raspberry Pi and camera module are used for image processing, and Arduino is used for motor control and infrared beacon. The two parts use UART to communicate, and the target information obtained by image processing is used for communication. Apply PID control and output it to Arduino to control the wheel speed with PWM, and the infrared beacon will be used as the target so that the slave car can follow. From the slave car side, multiple infrared receivers are used to determine the position of the beacon and the distance sensor is used to determine the distance. Finally, the information is transmitted to the Arduino by I2C to follow the autonomous car in front.

1.李宛真。「可跟隨與避障的雙重卷積神經網路之視覺自走車」。碩士論文，國立中央大學資訊工程學系，2018。https://hdl.handle.net/11296/d99r28。

2.許裕昇。「自動停車之停車位偵測與路徑追蹤控制研究」。碩士論文，國立臺北科技大學車輛工程系所，2009。https://hdl.handle.net/11296/5hhz25。

3.郭仲仁。「自動跟隨前導者軌跡行進的自走車」。碩士論文，國立中央大學資訊工程學系，2016。https://hdl.handle.net/11296/87skv2。

4.陳信宏。「利用深度影像處理技術的智慧型避障自走車之人體跟隨控制設計」。碩士論文，國立暨南國際大學電機工程學系，2018。https://hdl.handle.net/11296/84z95k。

5.賴一翔。「具搜尋與避障之自動跟隨機器人」。碩士論文，國立中央大學電機工程研究所，2009。https://hdl.handle.net/11296/8pt4wr。

6.許傑堯。「人物追隨自走車系統之研製」。碩士論文，國立虎尾科技大學資訊工程系碩士班，2018。https://hdl.handle.net/11296/35m7m3。

7. Chen, Yu-Tso, et al. "A RSSI-based algorithm for indoor localization using ZigBee in wireless sensor network." Proc. of the 15th International Conference on Distributed Multimedia Systems (DMS 2009). 2009.

8. Redmon, Joseph, et al. "You only look once: Unified, real-time object detection." Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.

9. Saunier, Nicolas, and Tarek Sayed. "A feature-based tracking algorithm for vehicles in intersections." The 3rd Canadian Conference on Computer and Robot Vision (CRV'06). IEEE, 2006.

10. NEC 規範下0和1的圖 https://www.sbprojects.net/knowledge/ir/nec.php

11. NEC 協議的資料格式圖 https://www.sbprojects.net/knowledge/ir/nec.php

12. I2C 傳輸的基本結構圖 https://www.circuitbasics.com/basics-of-the-i2c-communication-protocol/

13. HSV和HSV圖 https://medium.com/neurosapiens/segmentation-and-classification-with-hsv-8f2406c62b39

14. 非最大值抑制示意圖 https://docs.opencv.org/master/da/d22/tutorial\_py\_canny.html

15. 滯後閾值法示意圖 https://docs.opencv.org/master/da/d22/tutorial\_py\_canny.html

16. Canny, John. "A computational approach to edge detection." IEEE Transactions on pattern analysis and machine intelligence 6 (1986): 679-698.

17. Yuen, H. K., et al. "Comparative study of Hough transform methods for circle finding." Image and vision computing 8.1 (1990): 71-77.