BACK-PRESSURE CONTROL ON URBAN TRAFFIC CYBER-PHYSICAL SYSTEM

• Main Author: Farhan Fathurrahman - Nim: 13314011, I Wayan Hari Sukmaranu - Nim: 13314019, M.
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/28883
• Institution: Institut Teknologi Bandung

<p align="justify">The high population growth of Bandung City which is accompanied by the increasing number of vehicles causes the occurrence of congestion problems. In addition, the implementation of traffic control systems that still can not adapt to long queue of vehicles would be detrimental to certain phases that have accumulated longer queue lengths but gained less green time. This problem further observed using back-pressure control method to determine the green time value in each phase. There are two types of back-pressure algorithms used, namely cycle time back-pressure and slotted time back-pressure. <br /> <br /> The controller is designed to be applicable to cyber-physical systems (CPS) that consist of cyber system and physical system connected through a CPS network. The cyber system is composed of controller and logger while the physical system is composed of sensors, traffic lights, and traffic itself. The controller test is done in Hardware In The Loop (HIL) simulation which consists of three subsystems, namely virtual traffic on PTV Vissim software with simulation location Simpang Lima Bandung, controller in form of microcontroller, and data logger. The back-pressure algorithm will be programmed in the microcontroller to calculate the allocation of green time in each phase. The allocation of green time is then sent to control virtual traffic and stored on the data logger via MQTT wireless data communication protocol. <br /> <br /> The controller test is done by comparing the queue length and vehicle throughput value of back-pressure algorithm against the fixed time method. Simulation results show that the controller with back-pressure algorithm produces lower queue length than fixed time method. In addition, the quality of MQTT communication on the system was tested using a benchmark tool and obtained the latency and success ratio of the communication is very good.<p align="justify"> <br />