Advanced Traffic Signal Control System Installed in Phuket City, Kingdom of Thailand

Similar documents
Evaluation of Actuated Right Turn Signal Control Using the ITS Radio Communication System

DESIGN OF VEHICLE ACTUATED SIGNAL FOR A MAJOR CORRIDOR IN CHENNAI USING SIMULATION

A Fuzzy Signal Controller for Isolated Intersections

Evaluation of Connected Vehicle Technology for Concept Proposal Using V2X Testbed

OPAC Adaptive Engine Pinellas County Deployment

Next Generation of Adaptive Traffic Signal Control

Figures. Tables. Comparison of Interchange Control Methods...25

STREAM (Strategic Realtime Control for Megalopolis-Traffic)

Area Traffic Control System (ATCS)

Abilene District Traffic Signal Timing and Capacity Analysis

0-6920: PROACTIVE TRAFFIC SIGNAL TIMING AND COORDINATION FOR CONGESTION MITIGATION ON ARTERIAL ROADS. TxDOT Houston District

Validation Plan: Mitchell Hammock Road. Adaptive Traffic Signal Control System. Prepared by: City of Oviedo. Draft 1: June 2015

Keywords- Fuzzy Logic, Fuzzy Variables, Traffic Control, Membership Functions and Fuzzy Rule Base.

Agenda. Morning. TS2 Cabinet Components and Operation. Traffic Signal Ring Structure. Afternoon. Basic Preemption/Priority

Israel Railways No Fault Liability Renewal The Implementation of New Technological Safety Devices at Level Crossings. Amos Gellert, Nataly Kats

Adaptive signal Control. Tom Mathew

Frequently Asked Questions

Density Based Traffic Control with Emergency Override

Roadmap to Successful Deployment of Adaptive Systems

Area Traffic Control

Core Input Files + Engines. Node/Link/Activity Location Demand Type/ Vehicle Type VOT Table/ Emission Table. DTALite. Movement Capacity File

Road Traffic Estimation from Multiple GPS Data Using Incremental Weighted Update

Development and Application of On-Line Strategi for Optimal Intersection Control (Phase Ill) 1II II! IIi1111 III. I k I I I

Self-Organizing Traffic Signals for Arterial Control

Traffic Signal Timing Coordination. Innovation for better mobility

IMPROVEMENTS TO A QUEUE AND DELAY ESTIMATION ALGORITHM UTILIZED IN VIDEO IMAGING VEHICLE DETECTION SYSTEMS

Real Time Traffic Light Control System Using Image Processing

Automatic Routing of Traffic Signaling using Image Processing

Signal Coordination for Arterials and Networks CIVL 4162/6162

Signal Patterns for Improving Light Rail Operation By Wintana Miller and Mark Madden DKS Associates

THE CHALLENGES OF USING RADAR FOR PEDESTRIAN DETECTION

SOUND: A Traffic Simulation Model for Oversaturated Traffic Flow on Urban Expressways

An Operational Test of Adaptive Signal Control. Campbell Road Corridor Richardson, Texas

City of Surrey Adaptive Signal Control Pilot Project

Intelligent Traffic Signal Control System Using Embedded System

Constructing a Traffic Control Process Diagram

Computer Simulation for Traffic Control

Estimating Vehicle Trajectories on a Motorway by Data Fusion of Probe and Detector Data

PROBE DATA FROM CONSUMER GPS NAVIGATION DEVICES FOR THE ANALYSIS OF CONTROLLED INTERSECTIONS

Intelligent Technology for More Advanced Autonomous Driving

Traffic Controller Timing Processes

Development of Hybrid Image Sensor for Pedestrian Detection

Available online at ScienceDirect. Procedia Engineering 142 (2016 )

Evaluating a Signal Control System Using a Real-time Traffic Simulator Connected to a Traffic Signal Controller

WHITE PAPER BENEFITS OF OPTICOM GPS. Upgrading from Infrared to GPS Emergency Vehicle Preemption GLOB A L TRAFFIC TE CHNOLOGIE S

Digital inertial algorithm for recording track geometry on commercial shinkansen trains

Deployment and Testing of Optimized Autonomous and Connected Vehicle Trajectories at a Closed- Course Signalized Intersection

Chapter 39. Vehicle Actuated Signals Introduction Vehicle-Actuated Signals Basic Principles

LMD8000 PROGRAMMING GUIDE

76-GHz High-Resolution Radar for Autonomous Driving Support

Agenda. TS2 Cabinet Components and Operation. Understanding a Signal Plan Maccarone. Basic Preemption/Priority

GNSS and M2M for Automated Driving in Japan Masao FUKUSHIMA SIP Sub-Program Director ITS Technical Consultant, NISSAN MOTOR CO.,LTD May. 15.

AUSTRALIAN JOURNAL OF BASIC AND APPLIED SCIENCES

Texas Transportation Institute The Texas A&M University System College Station, Texas

THE EXPANSION OF DRIVING SAFETY SUPPORT SYSTEMS BY UTILIZING THE RADIO WAVES

INNOVATIVE DEPLOYMENT OF DYNAMIC MESSAGE SIGNS IN SAFETY APPLICATIONS

Traffic Control for a Swarm of Robots: Avoiding Group Conflicts

Intelligent Transport Systems and GNSS. ITSNT 2017 ENAC, Toulouse, France 11/ Nobuaki Kubo (TUMSAT)

CONCURRENT OPTIMIZATION OF SIGNAL PROGRESSION AND CROSSOVER SPACING FOR DIVERGING DIAMOND INTERCHANGES

Development of a 24 GHz Band Peripheral Monitoring Radar

Humans and Automated Driving Systems

EVALUATING AN ADAPTIVE SIGNAL CONTROL SYSTEM IN GRESHAM. James M. Peters, P.E., P.T.O.E., Jay McCoy, P.E., Robert Bertini, Ph.D., P.E.

DEVELOPMENT OF A MICROSCOPIC TRAFFIC SIMULATION MODEL FOR INTERACTIVE TRAFFIC ENVIRONMENT

Getting Through the Green: Smarter Traffic Management with Adaptive Signal Control

The Collaborative Digital Process Methodology achieved the half lead-time of new car development

Route-based Dynamic Preemption of Traffic Signals for Emergency Vehicle Operations

Wireless technologies Test systems

SIMULATION BASED PERFORMANCE TEST OF INCIDENT DETECTION ALGORITHMS USING BLUETOOTH MEASUREMENTS

The real-time urban traffic control system CRONOS: Algorithm and experiments

DEVELOPMENT AND EVALUATION OF AN ARTERIAL ADAPTIVE TRAFFIC SIGNAL CONTROL SYSTEM USING REINFORCEMENT LEARNING. A Dissertation YUANCHANG XIE

FPGA Implementation of VHDL Based Traffic Light Controller System

A STUDY OF FREEWAY TRAFFIC INFORMATION REPORTED VIA COMMERCIAL RADIO. Conrad L. Dudek. John D. Friebele. and. Roy C. Lautzenheiser

DEVELOPMENT OF AN ALGORITHM OF AUTOMATICALLY SETTING CRITICAL SPEEDS ON URBAN EXPRESSWAYS

Adaptive Signal System Mt. Juliet, TN. SR-171 (Mt. Juliet Road)

Model-based Design of Coordinated Traffic Controllers

Chapter 10. Non-Intrusive Technologies Introduction

Traffic Management for Smart Cities TNK115 SMART CITIES

Transportation Data Potpourri in Frisco

(51) Int Cl.: G09B 29/00 ( ) G01C 21/00 ( ) G06T 1/00 ( ) G08G 1/005 ( ) G09B 29/10 ( ) H04Q 7/34 (2006.

Development of Explosion-proof Autonomous Plant Operation Robot for Petrochemical Plants

Administering Saturated Signalized Networks Through Fuzzy Technique

DATACAR ADVANCED MULTILANE TRAFFIC MONITORING SYSTEM

Texas Transportation Institute The Texas A&M University System College Station, Texas

Presented by: Hesham Rakha, Ph.D., P. Eng.

PUBLICATION 213. Think Safety First

Use of Probe Vehicles to Increase Traffic Estimation Accuracy in Brisbane

Comparison of Simulation-Based Dynamic Traffic Assignment Approaches for Planning and Operations Management

Freeway Performance Measurement System (PeMS)

Texas Transportation Institute The Texas A&M University System College Station, Texas

Region-wide Microsimulation-based DTA: Context, Approach, and Implementation for NFTPO

)454 1 TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU

1. Travel time measurement using Bluetooth detectors 2. Travel times on arterials (characteristics & challenges) 3. Dealing with outliers 4.

Appendix CANADA / Québec Montréal tunnels Ville-Marie & Viger

USE OF WEB BASED REAL TIME NOISE DATA TRASMISSION FOR ACOUSTIC INVESTIGATION AND MAPPING

Appendix B: Transportation B-10 Toll Plaza Analysis

Arterial Traffic Signal Optimization: A Person-based Approach

Transportation and Traffic Theory: Flow, Dynamics and Human Interaction

RHODES: a real-time traffic adaptive signal control system

ULTRASONIC TRANSMITTER & RECEIVER

Experimental Study of Infrastructure Radar Modulation for. Vehicle and Pedestrian Detection

Transcription:

INFORMATION & COMMUNICATION SYSTEMS Advanced Traffic Signal Control System Installed in Phuket City, Kingdom of Thailand Hajime SAKAKIBARA, Masanori AOKI and Hiroshi MATSUMOTO Along with the economic development, traffic has increased in the Kingdom of Thailand and traffic congestion has become a problem in urban areas. To solve this problem, some districts of the major cities of Bangkok and Chiang-Mai are equipped with centralized traffic control systems. Most signalized intersections, however, are not connected to the centralized systems, and signal controllers give predetermined green signal times to the vehicles at intersections.the predetermined green times do not fit the traffic demand that changes dynamically during peak period, so the traffic police is forced to control traffic manually.sei has designed a new signal control system that is combined with an image processing vehicle detector and an advanced signal controller. The image processing vehicle detector was modified to detect information useful for signal control, and the signal controller was improved so as to calculate the green times that fit the changing traffic demand. The new signal control system is installed at six main intersections in Phuket City by January 2004, and the system has realized smooth and safe traffic at these intersections. This paper introduces the new signal control system in Phuket City. 1. Introduction Sumitomo Electric Industries, Ltd. (SEI) has installed a traffic signal control system, which combines newest traffic signal controller with spatial-measurement-type image processing type detector (hereinafter referred to as the image processing type detector or IDET ), to the six most congested intersections in the City of Phuket in August 2003 and January 2004. The system has realized safe and smooth traffic flows at these intersections. In Thailand, the centralized traffic control systems, which have the central processing computers set at the police headquarters, are installed in some cities including Bangkok and Changmai. The other cities employ the stand-alone fixed-time signal controllers, which operate based on the pre-set signal parameters. The traffic control system installed in Phuket employs the traffic signal controllers that incorporate some of the functions of the centralized system and are equivalent to the systems operated in Japan. 2. Traffic Condition in Thailand The GDP growth rate of Thailand maintains a high level reaching approximately 5%, and the number of vehicles owned is also increasing. In every city including Bangkok, heavy traffic congestion occurs at main intersections in the morning and evening peak periods. Because most traffic signal controllers installed at these intersections are the fixed-time controllers, they are not able to respond to changes in traffics. Traffic police officers are therefore forced to control traffic manually during peak traffic periods. Photo 1. Manual Traffic Control by Police Officer 3. Significance of New Traffic Signal Control System A common method for solving such problem is to carry out the optimal control of traffic signals at a traffic control center using information collected by the vehicle detectors installed at roadsides. The traffic control centers in Japan and Thailand were constructed under this concept. Construction of traffic control centers, however, requires huge amount of cost and time. The traffic signal control system installed in Phuket calculates optimal traffic control by the computer loaded in each traffic signal controller, thus realizing the decrease of traffic congestion at an inexpensive cost. 4. System Configuration In this section, the configuration of the system installed in Phuket city (hereinafter referred to as the Phuket type system) is explained by comparing the system with the centralized system. 54 Advanced Traffic Signal Control System Installed in Phuket City, Kingdom of Thailand

4-1 Centralized Type In Japan, more than 180,000 intersections are signalized, and 35% of the signalized intersections are connected with the traffic control centers by communication lines (as of year 2003). An example of the configuration of a centralized type system is shown in Fig. 1. For Right Turn Actuation Macro Control Traffic Control Center Vehicle Detector For Traffic Volume For Queue Length Measurement Measurement 30m 150m 300m 500m Fig. 1. System Configuration of Centralized System Hereafter every 250m (1) Vehicle Detectors Ultrasonic detector, a vehicle detector that emits ultrasonic wave, is used widely. As shown in Photo 2, the detector measures pulse reflection time and detects the existence of vehicle beneath. The general installation points of vehicle detectors are shown in Fig. 1. A vehicle detector actuated by the movement of a vehicle turning right is installed over a right turn lane at around 30 m from stop line. A detector for traffic volume measurement is placed at 150 m from stop line. The detectors for vehicle queue length measurement are placed at 300 m, 500 m, and at every 250 m from then on. As previously explained, ultrasonic detector measures the traffic volume of the lane directly beneath it, thus, the measurement accuracy for congestion queue length depends on the detector intervals. If detectors are installed at short intervals, it is possible to obtain more accurate data. However, costs for installation and maintenance increase as well. Based on the authors experiences and simulation result, it is economically recommendable to install detectors at 250 m intervals. Collected data will then be transmitted to the traffic control center to convert into predicted queue length, hourly traffic volume, and other traffic information used for traffic signal control. (2) Traffic Signal Control Method The function of a traffic signal controller consists of the macro control function that operates every 2.5 minutes or every cycle, and the micro control function that operates every 0.1 seconds. The macro control operates on central equipment and determines traffic signal parameters based on traffic volume data and congestion information collected by the detectors. The micro control operates on local traffic signal controller. The system detects the vehicles that enter into intersection and determines the optimal green time based on the standard green time determined by macro control, as shown in Fig. 2. For example, data on right-turning vehicles collected by a detector is transferred to the traffic signal controller for right-turning-vehicle-actuated control. Gap-actuated control like right-turning-vehicle-actuated control is not the only function of micro control. Public transportation priority control, which preferentially allocates green time to public transportation vehicles, and dilemma control, which either extends or shortens green time when vehicle enters into intersection at high speed and reduces accident occurrence rate, are the other functions of micro control. Standard Value Determined by Macro Control Green RED Adjustable Range (Adjustable range permitted to micro control) Fig. 2. Macro and Micro Controls 4-2 Phuket Type The system configuration is shown in Fig. 3. In the Phuket type system, traffic signal controller and detectors operate without being connected to the center. Vehicle detectors are also installed near intersections for collecting traffic volume information. (1) Vehicle Detectors Vehicle detector for right-turn-vehicle-actuated control is installed at 20-30 m from the stop line on a right turn lane and detector for vehicle volume measurement Vehicle existing Vehicle not existing For Right Turn Actuation Vehicle Detector 20-30m 40-60m For Traffic Volume Measurement Photo 2. Ultrasonic Vehicle Detector Fig. 3. System Configuration of Phuket Type System SEI TECHNICAL REVIEW NUMBER 60 JUNE 2005 55

is installed at 40-60 m upstream from the stop line. For the system in Phuket, image processing type detectors were used instead of ultrasonic detectors. The functions of image processing type detectors and the reason for their replacement will be explained in Section 5. (2) Traffic Signal Controller The latest local traffic signal controller has macro and micro control functions. Moreover, the controller can be connected with a traffic control center to operate as a part of a centralized system. The functions of traffic signal controllers will be described in Section 6. 5. Image Processing Type Vehicle Detector This section explains the image processing type detectors used in the Phuket type system. 5-1 Overview A camera unit can be installed at 8-10 m above the road surface and can collect traffic data listed in Table 1. In the Phuket type system, the camera units are installed Measurement Item Traffic Volume / Occupancy Speed Vehicle Type (L/S) Vehicle Presence Queue Length Table 1. Measurement Items Number of Measurable Lanes 2 lanes on traffic signal poles as shown in Fig. 4. When using ultrasonic detectors, setting up of a new pole is required for each unit. When using image processing type detectors, however, each unit can be installed on a traffic signal pole, and therefore is preferable for aesthetic and cost reasons. 5-2 Improvement of Functions On the construction of the Phuket type traffic signal control system, SEI improved the image processing type detector s function to make traffic signal controller operate more appropriately according to traffic condition. (1) Vehicle Existence Detection It is important in micro control to understand the behavior of vehicles that approach to the intersection. In general, a vehicle passes the stop line every 2 seconds on each lane under congested condition (as shown in Fig. 5). This traffic performance is called saturation flow. In gap-actuated control, the distance between two vehicles is measured and if the distance in seconds is 3-4 seconds (as shown in Fig. 6), the traffic controller judges that the saturation flow stopped and ends green signal accordingly. In the case of ultrasonic detectors, the detectors are installed at 20-30 m upstream from the stop line on a right turn lane and at 40-60 m upstream from the stop line on a through-traffic lane for detecting the above mentioned traffic condition. Traffic condition between stop line and detectors is an estimated value. The image processing type detectors installed in Phuket can detect vehicle presence lane by lane in a Vehicle passes stop line in every 2 sec.per lane Fig. 5. Definition of Saturation Flow Volume, Speed, Type Ultrasonic vehicle detector Presence Image processing vehicle detector Space gap : More than 3-4 sec. Fig. 4. Overview Fig. 6. Determination of Green Signal Termination 56 Advanced Traffic Signal Control System Installed in Phuket City, Kingdom of Thailand

detection area, which is 20-60 m upstream from stop line, by a single camera unit. This leads to the perfect detection of vehicle existence at intersections, and allows accurate termination of green signal time. (2) Exit-Blocking Detection It is preferable to end green time and assign it to the other direction when spillback from downstream intersections reaches to the controlled intersection and vehicles cannot pass trough the intersection. An image processing type detector measures the speed of each vehicle in the detection area. When vehicles are not moving for a certain period of time in the detection area, the detector turns off vehicle existence signal, or in other words, the detector judges that there are no vehicle passing through the intersection and ends green signal. In this way the system can eliminate unnecessary green time. 6. Traffic Signal Controller As explained earlier, some of the functions of the centralized system are incorporated into local traffic signal controllers in the Phuket Type system. Table 2 is the comparison of macro control between the Centralized Type and Phuket Type systems. The overview of macro control realized in the Phuket type system is explained below. Because micro control is the original function of traffic signal controllers, the controllers in the Phuket tye system have this function from the beginning. Control Type Control Item Table 2. Comparison of Macro Control Centralized Phuket Type Split Control Actuated Control Actuated Control Cycle Control Actuated Control Fixed Time / Actuated Control Offset Control Actuated Control Fixed Time Control 6-1 Split Control The most effective signal control technique for reducing congestion at intersections is split control. In a centralized system, the split value of each signal aspect is calculated to become equal, based on traffic volume information during smooth traffic and on congestion queue length information during traffic congestion. In the Phuket type system, because detectors for measuring congestion queue length were not installed, it is not possible to recognize the occurrence of congestion. During traffic congestion, the results of micro control are reflected on split calculation as follows. 1 In a certain signal aspect, when green time reached the maximum adjustable range, it is judged that green time is not sufficient or congestion has occurred. 2 The standard green time for that direction will be increased at the next split calculation (macro control). 6-2 Cycle Length Control In Phuket city, at intersections where no offset control is required since distances between intersections are long, the cycle-less control method, which adjusts signal cycle lengths according to the incoming traffic volume of each cycle, was employed. For those intersections that require offset control, cycles were adjusted based on time of day (TOD) data, due to the reason explained in Section 6-3. 6-3 Offset Control At intersections where the distances between are less than 300 m, offset control is employed to adjust the start time of green signal at each intersection, so that vehicles are able to pass through the intersections with minimum stops. If the cycle length of each intersection differs, it is impossible to maintain the difference in signal start time. The cycle length of each intersection must hence be equalized. Furthermore, because the time difference of the start of green signal is adjusted on the second time scale, the clock in each controller must be adjusted in like manner. In the case of centralized system, signal cycle lengths, traffic information, and signal control parameters are administrated collectively. Therefore, green signal start time will not deviate from the standard time. On the other hand, in the Phuket type system, traffic information cannot be administrated collectively via a traffic control center, and signal control will be executed based on preinstalled time-of-day data. On this occasion, if the built-in clock of each signal controller is inaccurate, the green signal start time will deviate from the standard time. Thus, the signal controllers in Phuket are equipped with a GPS system to adjust the clock every hour. 7. Effect of Installed Systems With regard to the three intersections in which the system developed by SEI was installed in August 2003, an investigation to check the effectiveness of the system was carried out by Chulalongkorn University after the installation. As for one intersection, however, ex-ante data could not be collected since other construction work was taking place at the time of traffic investigation. Therefore, the system s effectiveness was investigated and compared for the two intersections only. (1) Congestion Status At one intersection, traffic congestion improved dramatically. At the other intersection, however, an intersection located 80 m down the street affected that intersection, and the intersection s effectiveness therefore could not be proven. The problem was solved in January 2004 when a traffic signal controller was installed at the downstream intersection and offset control was realized. (2) Road Users Evaluation More than 90% of the road users were satisfied with SEI s system. They were satisfied especially with traffic SEI TECHNICAL REVIEW NUMBER 60 JUNE 2005 57

order and safety at right turning situation. The road users evaluation showed that they preferred adaptive control to manual control by police officers. Cycle length Stage 1 Stage 2 Stage 3 Stage 4 8. Conclusion By constructing a traffic control center and installing vehicle detectors at roadsides, the traffic signal control that optimally corresponds to various traffic conditions can be realized. It was reported that by constructing a control center and connecting intersections with the center, traffic congestion was improved by 15-20%. The construction of centralized system, however, requires a large amount of time and cost. There is a large difference in adaptability to changes in traffic conditions between the intersections controlled using TOD data and intersections controlled collectively via a control center. To compensate for this difference, the Phuket type system that can adapt to traffic condition changes has been developed. The traffic signal controllers used in this system are expandable for use in the centralized system. Therefore, it is possible to first install the system into the congested intersections at different locations and then construct a control center after the number of the system-installed intersections has increased to a certain extent. The authors are expecting to install this system into the intersections with decentralized traffic control in the near future. {Explanatory Note} *1 Traffic Signal Control Parameters: Traffic signal control parameters are composed of cycle length, split and offset. One cycle is a complete sequence of signal aspects, and the time required for one cycle is called cycle length. Cycle length is normally adjusted to become long at high traffic volume conditions and adjusted to become short at low traffic volume conditions. Split is a percentage of a cycle length allocated to Split Offset Time Red Signal 50% 7% 38% 5% Detector the green signal aspect and is calculated based on the traffic demand ratio of each direction. Offset is the time difference between the start time of the green signal at one intersection and the start time of the green signal at the other corresponding intersection. The offset is adjusted so that the vehicles can pass intersections with minimum delay time and minimum number of stops. References (1) Kazuaki Goto, Masakatsu Higashikubo, Masanori Aoki A Spatial Image Processing Traffic Flow Sensor and it's applications for Signal Control, Surveillance and Warning System, Trans. IEE of Japan, Vol.121-D, No.1, Jan., 2001 Contributors H. SAKAKIBARA Assistant Manager, Traffic Systems Development Section, Systems Technology Department/Tokyo, Systems Development Division, Sumitomo Electric Field Systems Co., Ltd. M. AOKI Ph.D., Assistant General Manager, Equipment Development Department, Systems & Electronics Division H. MATSUMOTO Traffic Systems Engineering Group, Systems Engineering Department, Systems & Electronics Division 58 Advanced Traffic Signal Control System Installed in Phuket City, Kingdom of Thailand