SCATS Adaptive Traffic System

Slides:



Advertisements
Similar presentations
INPUT-OUTPUT ORGANIZATION
Advertisements

Travel Time Estimation on Arterial Streets By Heng Wang, Transportation Analyst Houston-Galveston Area Council Dr. Antoine G Hobeika, Professor Virginia.
Multi-DNC Data Collection/Monitoring
.nowNAZTEC 1. Overview AGENDA Benefits Algorithm Simulation 2 Case Studies Real-time adaptive traffic control system Adjusts signal timing based on current.
Transit Signal Priority Applications New Technologies, New Opportunities Peter Koonce, PE APTA BRT Conference – Seattle, WA Wednesday, May 5, 2009 Technology.
1 The Next Generation AMR System Pinnacle Technologies (John MacConnell) Valon Technology, LLC (Stuart Rumley) A Joint Development between: and.
INTRODUCTION TO TRANSPORT Lecture 7 Introduction to Transport Lecture 7: Signal Coordination.
Traffic Controller Methodologies NEMA-based vs. Interval-based
Geometry information from Liufang Ave. North in Beijing Green phase (s): 99,77,66,75,60 Dwell time distribution: N (30,9), N (27,7), N (24,6) Maximal bandwidth:
GEOG 111 & 211A Transportation Planning Traffic Assignment.
Copyright © Siemens plc All rights reserved. Fuel for thought.
Junction Modelling in a Strategic Transport Model Wee Liang Lim Henry Le Land Transport Authority, Singapore.
Network Management Overview IACT 918 July 2004 Gene Awyzio SITACS University of Wollongong.
Chapter 221 Chapter 22: Fundamentals of Signal Timing: Actuated Signals Explain terms related to actuated signals Explain why and where actuated signals.
CTC-340 Signals - Basics. Terms & Definitions (review) Cycle - Cycle Length - Interval -. change interval - clearance interval- change + clearance = Yi.
Lec 24, Ch.19: Actuated signals and detectors (Objectives) Learn terminology related to actuated signals Understand why and where actuated signals are.
PASSER II SOFTWARE.
Advanced Public Transit Systems (APTS) Transit ITS CEE582.
Lec 14, Ch.8, pp : Intersection control and warrants (objectives) Know the purpose of traffic control Know what MUTCD is and what’s in it Know what.
SDLC. Information Systems Development Terms SDLC - the development method used by most organizations today for large, complex systems Systems Analysts.
Customized Simulation Modeling Using PARAMICS Application Programming Interface Henry Liu, Lianyu Chu & Will Recker.
INPUT-OUTPUT ORGANIZATION
Transportation Research Board 2004 Annual Meeting Adaptive Signal Control Workshop Session 2: Field Experience January 11, 2004.
Adaptive Traffic Light Control For Traffic Network.
Peter Koonce TRB Annual Meeting January 9, 2005 Best Practices for Signal Operations Best Practices for Signal Operations – Lessons Learned from the Portland.
2015 Traffic Signals 101 Topic 7 Field Operations.
Signalized Intersection Delay Monitoring for Signal Retiming SafeTrip-21 Safe and Efficient Travel through Innovation and Partnership in the 21 st Century.
RT-TRACS A daptive Control Algorithms VFC-OPAC Farhad Pooran PB Farradyne Inc. TRB A3A18 Mid-Year Meeting and Adaptive Control Workshop July 12-14, 1998.
Applied Transportation Analysis ITS Application SCATS.
Oregon Microsimulation Roundtable March 2012
SCATS Adaptive Traffic System
Real Time ACD Package DIGITAL COMMUNICATIONS PLATFORM Digital Communications Platform.
Chapter 6 Control Using Wireless Throttling Valves.
McGraw-Hill/Irwin ©2008 The McGraw-Hill Companies, All Rights Reserved SECTION 8.1 SUPPLY CHAIN FUNDAMENTALS.
1 Modeling Active Traffic Management for the I-80 Integrated Corridor Mobility (ICM) Project Terry Klim, P.E. Kevin Fehon, P.E. DKS Associates D.
MICROPROCESSOR INPUT/OUTPUT
Transit Priority Strategies for Multiple Routes under Headway-based Operations Shandong University, China & University of Maryland at College Park, USA.
INTELLIGENT TRANSPORT SYSTEMS
Chapter 20: Actuated Signal Control and Detection
TRAFFIC SIGNAL OPTIMIZATION: A Coordinated Effort Tom Dancey, P.E. Signal System Engineer City of Springfield CITY OF SPRINGFIELD & MISSOURI DEPARTMENT.
Topics of presentation
Intro to Network Design
Implementation Transit Priority System and Mobile Internet Passenger System in the City of Los Angeles Kang Hu and Chun Wong City of Los Angeles Department.
Session 2 Equipment Requirements for Adaptive Traffic Control SCOOT Axel Reissnecker, EAGLE Traffic Control Systems, Austin, TX TRB Workshop on Adaptive.
OBTAINING QUALITY MILL PERFORMANCE Dan Miller
Siemens Traffic Controls Ltd ITSE99/Standards 1 Traffic Management and Control Workshop on Research and Technological Development for Information Society.
Transit Signal Priority (TSP). Problem: Transit vehicles are slow Problem: Transit vehicles are effected even more than cars by traffic lights –The number.
TRANSIMS Version 5 Network Files January 20, 2011 David Roden – AECOM.
RHODES Gardner Systems RHODES: Fundamental Principles Larry Head, Gardner Systems Pitu Mirchandani, University of Arizona TRB Annual Meeting 2000 Workshop.
SCATS Release Review SMUG, March 2013 Sydney.
Lecture 3 Applications of TDM ( T & E Lines ) & Statistical TDM.
1 RANKING OF FUNCTIONAL REQUIREMENTS/ TOPIC AREAS FOR NEEDED RESEARCH BY THE TRB TRAFFIC SIGNAL SYSTEMS COMMITTEE JULY 21, 2002 MID-YEAR MEETING IN SALT.
SUPPLY CHAIN MANAGEMENT SYSTEMS Part I. 7-2 LEARNING OUTCOMES 1.List and describe the components of a typical supply chain 2.Define the relationship between.
Traffic Control Systems - SCOOT/UTC ATD 98/07 -Transp.-Nr.: 1 / 00 Ord.-Nr.: SCOOT TRB Presentation on SCOOT January 2000.
SCATS 6.10 Release Review SMUG, March 2013 Sydney.
Hcm 2010: BASIC CONCEPTS praveen edara, ph.d., p.e., PTOE
Chapter 9 Capacity and Level of Service for Highway Segments
Mission Street Voyage Advanced Features Test Facility Mission street is a major arterial corridor that was re-timed 10 years prior to our project. Our.
SCATS Congestion Improvement Program. The Scope of the SCATS Congestion Improvement Program.
ITS Virginia Annual Conference April 20, 2012 Sensys Networks and the Sensys Networks logo are trademarks of Sensys Networks, Inc. Other product and company.
Thinking Inside the Box
Lab A: Planning an Installation
MOVA Traffic Signal Control Trial
Smart Ethernet I/O P2P and GCL Introduction
SNUG, February 2017 Tauranga
GREEN WAVE TRAFFIC OPTIMIZATION
Transportation Research Board 2004 Annual Meeting
* Topic 7 Field Operations
Presentation transcript:

SCATS Adaptive Traffic System TRB Committee A3A18 Adaptive Traffic Control Workshop July 1998 Note : Additional comments available on “Notes” page. For further information contact: Neil.R.Gross@cpmx.saic.com Company: www.tcore.com

SCATS - Objectives and Installations Minimize Stops( light traffic), delay (heavy traffic) and travel time. SCATS is installed in many cities worldwide, There are approximately 5000 intersection under SCATS control around the world, Largest systems are: Sydney (2000 intersections), Melbourne (2000), Hong Kong (600) and, in the US, Oakland County MI (350).

SCATS System Architecture Minimal System - Single Region The smallest SCATS system is a single Region. The single Regional Computer can operate up to 128 intersections. Up to 8 operator terminals (direct connect of dial-in) are available. Communications to traffic controllers can be point-to-point or multidrop.

SCATS System Architecture Expansion from Single Region SCATS is expanded by adding Regions. Some users achieve operator access convenience by using a data switch.

SCATS System Architecture Management System The Management computer adds Region to Region communication (and coordination), operator access convenience and many traffic system management facilities including equipment inventory and repair systems and data collection and analysis systems.

SCATS System Architecture Remote Regional Computers Regional Computers can be either local or remotely connected using data lines.

SCATS System Architecture Full System - with Integration Server The integration server, show at the left provides the ability to integrate SCATS with other ITS systems. The server allows SCATS data to be shared with another system and SCATS commands to be invoked. (The interface is via a TCP/IP socket).

SCATS System Architecture Local Traffic Controllers - tactical control (calling, extension) and data collection, Regional Computers - Strategic control, Management Computer - Communications and Database functions. Simplest configuration - single Regional Computer. Operator Interface - Windows 95 or -NT Graphical user interface with point and click access to all parameters. Local tactical control can still occur during adaptive control by the host. The controller gains “permission” from the host before terminating or skipping movements.

SCATS GUI Example

SCATS Data Requirements Loop Detectors or equivalent (video detection in Oakland County MI) in each lane at the stop line. Detectors used for calling and extension. Controller collects number of spaces and total space time during green of each phase, each cycle for use by SCATS adaptive algorithm. Actual movement data collected by stop line detectors allows accurate split determination. Loop detectors are 6 x 15 feet and are located in each lane at the stop line.

SCATS Data Requirements Degree of Saturation (DS) and Car Equivalent Flow (VK) for each Approach lane . DS used to vote for Cycle Length and Split Plan. VK used to vote for Offset Plan

SCATS Data Requirements An upstream approach can vote at downstream intersection (Engineer selectable). “Early” influence. Tactical operation of controller can be enhanced by special detector logic in the controller personality. Special functions include: queue length detectors detector combinations turn/through discrimination for shared lanes, etc. The special detector logic can link back to system actions. An example is queue detectors on short freeway off-ramps. While the controller can react independently to clear a queue it also informs the host allowing it to react over a broader area if the normal adaptive algorithm may not be quick enough etc.

SCATS Data Requirements No modeling required. User defines: subsystems target cycle lengths and relationship to DS, split plan strategy linkages and offsets

SCATS Communication Requirements Point-to-Point or Multidrop. Once per second communication with each intersection. Messages are normally 1 to 5 bytes, average 3 bytes.

SCATS Communication Requirements Optional digital communications port (RS232) for direct network connection, Requires 300 Baud Full duplex channel with addressing/routing by network.

SCATS Hardware Requirements Management Computer - DEC VAX/ALPHA, OpenVMS Regional Computer - Personal Computer (200 Mhz) with Windows NT and Digi serial communications interface modules, Local Processor - Traffic Controller with SCATS functionality, available for NEMA and 170.

SCATS Traffic Controllers 170 NEMA AWA Delta 170 upgrades existing 170 controller to support SCATS. Relay Module added for sense/control of cabinet status. AWA Delta 3N controller replaces existing NEMA controller. Connections are via existing A, B and C connectors. One back-panel link required. NEMA - Link required. between "TestB" and Conflict Monitor Output Relay 2 NC on back panel

SCATS Control Variables Tactical Control presence (locked or non-locked) for phase call non-occupancy for gapping and wasting (accumulated waste green) for phase termination. Strategic Control Number of spaces and total space time Used to develop “Degree of Saturation” DS and Car “Equivalent Flow” VK

SCATS Degree of Saturation DS = [green-(unused green)]/available green Green is phase time during data collection Unused green is space time greater than or less than the saturation space time. i.e. Total space time from controller, LESS Number of spaces times the standard space time at maximum flow Unused green is a measure of efficiency (zero at saturation flow, +ve undersat, -ve oversat). Available green includes unused split time due to gapping etc.

SCATS Degree of Saturation DS = [green-(unused green)]/available green Standard space time at maximum flow is self calibrated daily DS is the ratio of efficiently used phase time to available phase time, DS can be >100% i.e. during oversaturation the used green can be negative - vehicles are closer than standard space time at maximum flow.

SCATS Car Equivalent Flow - VK Derived from DS and the lane saturation flow for each lane, Independent of vehicle types in traffic stream, Allows valid comparisons of competing flows for offset selection. VK= DS x green time x vehicles per second at maximum flow

SCATS Data Smoothing and Damping DS and VK are used as weighted averages usually over three cycles, SCATS uses smoothing, damping (i.e.reducing the gain of feedback control loops) and hysteresis extensively, It is the calibration of these techniques over years of experience that is the key to effective performance.

SCATS DS Usage - Cycle Length Delay increases rapidly for CL below Co (optimum CL) Exact CL not critical as long as not less than Co SCATS Subsystem CL determined from highest value of DS in the subsystem.

SCATS DS Usage - Cycle Length User defined equilibrium DS values used to determine relationship between measured DS and CL. Objective to keep CL below user defined targets. RL (target CL) determined for measured DS. Compared with last CL Difference and direction of change  RL’

SCATS DS Usage - Cycle Length Weighted average of RL’ (last three cycles) determines final RL CL can move toward final RL by +/- 6 seconds. CL can change by up to 9 seconds where RL for the last two cycles was > 6 seconds. (allows response to steep change in demand) Subsystems at LCL(low CL) move to SCL (“Stopper” CL) based on flow per cycle parameters, not DS. (i.e. step change).

SCATS DS Usage - Split Plan Possible split plans examined each cycle to determine the most “equisat” plan for the next cycle, i.e.minimal delay Equisat: DS on critical approaches equal,. Maximum projected DS for each possible plan calculated (using last cycle DS values). Plan with the lowest maximum selected. Projected DS = DS (old split/new split) SCATS split plan is selected by calculating the resulting DS for each possible plan. The plans used are either user nominated or, usually, set up automatically when the “Incremental Split Selection” (ISS) facilities is used.

SCATS DS Usage - Split Plan For Incremental Split Selection: selection is from 7 possible “plans” for 2 phase (stage) intersection or 37 possible “plans” for 3 and 4 phase (stage) intersections (sample shown below) Plan 21 22 23 24 25 Phase 1 -4 2 2 -4 0 Phase 2 2 0 2 2 -4 Phase 3 0 -4 0 2 2 Phase 4 2 2 -4 0 2 (figures are percent change, i.e. plan 21 = 4 % off phase 1, 2 % added to plans 2 and 4.) This slide provides an example of a small section of the 3 and 4 phase ISS plans tables.

SCATS Offset Selection Offset plans are selected by comparing traffic flows on the links, Directional Bias values (DB’s) are entered for each of four plans for each link Weighted three-cycle average volumes (VK) are multiplied by the DB’s and the results summed for each plan, The plan with the highest sum receives the vote.

SCATS Offset Selection A new offset plan is adopted when 4 of the last 5 votes are for the same plan. Two offset values, a and b, are entered for each offset plan, and a CL range, CL1 and CL2, is entered for each plan, The offset adopted is a at CL1, b at CL2 and a linear interpolation for CL between CL1 and CL2 (can be disabled if “jump” desired)..

SCATS Coordination Intersections are grouped in Sub-systems, A sub-system comprises one or more intersections only one of which is “critical” i.e. requires dynamic split selection, All cycle length and split plan voting is carried out at the critical intersection, CL and Splits at “minor” intersections in the sub-system are controlled by the critical intersection.

SCATS Coordination All intersections in a sub-system operate at the same CL and are coordinated via offsets. Sub-systems can “marry” to achieve coordination using a separate set of offsets, “Married” sub-systems have the same CL, “Marriage” and “Divorce” is controlled through voting based on CL and volume and occurs automatically.

SCATS Phasing Flexibility Compatible phases (signal groups or displays) grouped into STAGES (e.g. main street through may be 2 and 6. These are grouped into Stage A). Signal Group control within stages for conditional overlaps, green arrow vs ped. control etc. allowed. SCATS has seven Stages, A to G, Stages can be introduced in any order, Any undemanded stage can be skipped,

SCATS Phasing Flexibility In Isolated and Flexilink (fallback) modes the sequence is defined in controller “personality”. Several options are provided. In Masterlink mode the sequence is determined by data in the Regional Computer. Split plan features used to control gapping, stage selection and assignment of unused stage time (e.g. no gap, no gap for % of stage, time gain etc.).

SCATS Arterial/Network Capability Normally arterial, i.e one coordination route, Offset plans automatically arranged for low CL, Direction 1, “Business Peak” and Direction 2 use, For a network the offset plans can be independent for use on multiple coordination routes. (Select N1 subsystem key option). Split plan features can be used to ensure stages run full length to ensure coordination.

SCATS Arterial/Network Capability Offsets between subsystems are defined in the form LPn=ttppnn i.e. reference offset in this subsystem is offset by tt seconds from the end of Stage pp at intersection nn, nn can be different on each of the four offset plans, thus coordination decisions are not constrained by simple inbound vs outbound arguments. SCATS can be used in networks as well as arterials and is deployed in many downtown grid networks. The problem of a closed network is no different to two intersections on a two-way street. These two intersections are a closed network because, when the offset from intersection one to intersection two is defined, the offset from intersection two to intersection one is not independent of cycle time. Three intersections in a triangle does not present a more difficult problem. If the offsets between intersections 1 and 2 and intersections 2 and 3 are defined, the offset between intersections 3 and 1 is a function of cycle time. Similarly for four intersections in a square; three offsets can be defined and the fourth (which closes the loop) is not independent of cycle time. This is not a problem in SCATS, it is a fundamental problem which no system can address because it is an inescapable fact that the sum of the offsets around a closed loop must add up to the cycle time or a multiple thereof. This means that as the cycle time varies in an adaptive manner, the undefined offset which closes the loop (even in the case of two intersections on a two-way street) also varies by the same amount as the change in cycle time.

SCATS Arterial/Network Capability When SCATS is employed on a grid network, offsets are selected as dictated for the heavily trafficked routes through the network, At all times, as many links as possible will be operating with defined offsets and these will be the links with the greatest flow, The remaining links, for which offsets cannot be defined because it would close loops, are those with the lowest traffic flow. When SCATS is deployed in a grid network, offsets are selected as dictated by the most heavily trafficked routes through the network. In fact, arterials through the network (which can and usually do go round corners) are picked out by SCATS as the reflection of the current demand. When demand pattern changes, different offset pairs are selected to form different coordinated routes through the network. If the grid system has regularly spaced intersections and the travel time for each block is approximately the same, there may be a suitable cycle time to get good offsets on at least one direction of each link pair eg if the block travel time is 20 seconds, a cycle time of 80 seconds will produce good one-way offsets on four links of a square but poor offsets on the other link in each pair. Good two way coordination on a link pair in this example would require a cycle time of 40 seconds so there is always a compromise. SCATS does have the ability to settle on a "good progression" cycle time (SCL) when demand does not exceed the capacity of that cycle time and this translates to the implementation of good offsets on four of the link pairs in a square. The important point (already stated previous slide) is that SCATS has the ability to select offsets which give good offsets on as many links as possible in a grid network, limited only by the condition (which has nothing to do with SCATS, it is fundamental) that link offsets cannot be defined in a closed loop without constraining cycle time.

SCATS Measures of Effectiveness MOEs available from system include (per lane): SCATS Degree of Saturation DS VO/VK (actual/calculated vehicles during green) MOEs should be measured independently: SCATS in Sydney is equipped with ANTTS (Automatic Network Travel Time Subsystem-link travel times from 4000 taxicabs collected and analyzed continuously). Unusual Congestion Monitor

SCATS Priority Systems - Controller Five priority inputs provided, one railroad and four vehicle, Vehicle priority inputs accept steady or pulsed signal for different preemption display, Preemption display (signal groups), ending overlaps and return stage can be selected, Preemption is a function of the controller, SCATS knows preemption is active.

SCATS Priority Systems - System SCATS Route Preemption Control (RPC) System provides automatic emergency route control from a single input (e.g. fire station pushbutton), Route is defined as list of intersections with stage(s) to be held (dwelled), delay from previous I/S and dwell time. Monitor is provided for up to 10 intersections. Also some transducer based priority systems. For example: the tram pre-emption system in Melbourne Australia which recognizes a tram and preempts the tram approach to ensure traffic flow.

SCATS and Oversaturation SCATS DS can be >100% i.e. oversaturated, “Stretch effect” i.e. all stages share extra CL up to XCL after which only the stretch (usually coordination) stage gets the benefit (i.e. a move away from equisat), SCATS allows the traffic engineer to decide which route should be favored, by how much and where the queue can be tolerated. As the Cycle Length increases all approaches receive extra split time according to the % split applicable at the time. After XCL is exceeded and up to HCL (highest cycle length) any extra cycle time is is given to the “Stretch Stage” usually the coordination movement (i.e. main street).

SCATS and Oversaturation Illustration of the Stretch effect A stage is the stretch stage. The stretch stage can be adaptively changed if required, i.e. if the major traffic flow (i.e. the arterial) changes during the day.

SCATS Controls SCATS provides many facilities for the traffic engineer to achieve “custom” control in special circumstances while still maintaining adaptive operation, Variation routines at intersections allow special operation based on detection of a parameter value (CL, volumes, stage or phase active, next stage to run etc.) including calling of an operator keystroke “macro”.

SCATS Management System Inventory System Extended Alarm Monitor and log Extended System Event log Extended System Monitor/Volume Monitor Unusual Congestion (Incident) Monitor Flow Database System (Count Station Data) In large traffic control systems the ability to efficiently manage a large inventory of equipment and communications lines can be just as important as having a superior traffic control algorithm. The Unusual Congestion Monitor system provides reliable detection of sites where non-recurring congestion exists. Provides and excellent tools for traffic managers to concentrate on problem areas due to capacity reductions or unusual and excessive volume.

SCATS Management System Flexigen System (auto generation of time based fallback plans from SCATS data) Maintenance Management System Vehicle Location System Bus Passenger Information System “Tidal Flow” Intersection Control System VMS Control

End Thank You