How InSync Works
“InSync provides seamless operations between intersections with minimum delays and drive times. No engineering staff time is necessary to address changing traffic patterns in this growing section of Little Rock. We are extremely pleased with the system.”
Even the latest traffic controllers on the market still conform to the operational concepts of analog electromechanical controllers from the early twentieth century. Modern controllers are constrained in their sequencing, green splits, offsets, cycle lengths and relative cycle lengths, all of which reduce a system’s ability to serve actual traffic demand.
In contrast to this, InSync uses a digital state machine. It can flexibly adapt to actual traffic conditions based on real-time inputs and algorithmic decisions.
In InSync, a state is a phase or concurrent phase pair. The system chooses the state that best serves traffic conditions on a second-by-second basis based on detection data, the operational objectives specific to each intersection and network of intersections and InSync’s algorithms.
This chart shows the eight north-south sequences used in conjunction with eight east-west (not shown) sequences. Your agency’s preferences shape the options available. InSync simplifies a wide range of signalization possibilities to digital choices so it can quickly adapt to actual demand.
By digitizing the traffic control options available, InSync can dynamically choose and adjust signalization parameters such as the state, sequence and amount of green time to best serve actual traffic conditions. (Using standard sequences, InSync maintains all safety considerations while not being constrained by the ring-and-barrier.)
InSync’s local optimizer focuses on minimizing summed delay at each individual intersection. The system does this by constantly measuring volume (the number of vehicles) and delay (the time vehicles spend waiting) at each intersection, then making instant decisions about how to best reduce those numbers.
A greedy algorithm is the ideal choice for signalization because the finite number of signal states and green time durations (options) allow rapid solutions in order to serve real-time traffic demand.
Given the inputs of number of vehicles in the queue and their delay, the algorithm considers each possible signalization solution as well as its constraints and objectives as defined by the configuration.
InSync must count cars quickly and accurately to keep this perpetual math equation going. On most InSync deployments, Rhythm Engineering’s video detection cameras facing each approach handle this task. Once the cameras are mounted and accessible via a secure Ethernet connection, detection zones are drawn along the contours of each lane and subdivide each detection zone into segments. By counting how many segments have vehicles in them, InSync always knows, second-by-second, how many vehicles are in each lane and how long they’ve been waiting.
InSync:Tesla, which does not use our proprietary detection cameras, measures demand differently. InSync:Tesla learns the prevailing demand for each approach on different days and times, then monitors presence and delay at the stop-bar to either confirm or adjust its expectations and signalization choices in real-time.
Rather than merely recognizing vehicle presence, InSync's optimization process takes into account the wait duration and number of vehicles present in each queue to maximize efficiency in green time allocation. Additionally, the system's digital architecture and artificial intelligence enable a dynamic choice of optimal signal states.
To make traffic flow efficiently, InSync identifies the best phase combinations based on actual traffic demand. This design maximizes efficiency and clears queues without wasting green time on empty approaches.
InSync has three different adaptive characteristics in its local optimization. The system’s flexibility with phasing, sequencing and green time allocation enables it to adapt to actual demand.
Phasing is the first way in which InSync adapts to demand at the local level. InSync uses a digital state machine rather than a fixed timing plan, allowing it to spontaneously choose the phases and sequences that best serve actual traffic demand. The system continuously measures queue and delay at each approach, which allows it to calculate demand. Since InSync is unconstrained by cycles, the system determines priority so it can serve approaches from highest priority to lowest priority. The system chooses from the states (concurrently permissible phasing) available to it and requests the traffic controller actuate green lights accordingly.
The second way InSync adapts to actual traffic demand is with sequencing. The local traffic engineer can select allowable sequences using CentralSync. InSync draws from the available sequences but can skip a particular phase in the sequence whenever there is no demand for service. This provides the best possible use of green time for vehicles waiting at or approaching the intersection.
The third way InSync adapts to demand at the local level is in its green time allocation. In addition to actuating phases, InSync adjusts green time according to the volume of demand and intersection geometry. If there are a low number of vehicles demanding service, less green time is allocated. By not serving green time to empty approaches and instead distributing time to those approaches with demand for service, all approaches benefit.
InSync’s global optimizer ensures vehicles experience as few stops as possible. The global optimizer does so by synchronizing all traffic signals in an InSync network of intersections (such as on a corridor). By creating and serving "green tunnels," InSync’s software improves the travel time and safety of motorists, while reducing the carbon footprint of their daily commutes.
InSync minimizes stops and congestion along a road through coordinating signals to move traffic at a desired speed. As a result, InSync’s global optimizer frees up existing roadway capacity by keeping traffic moving through a series of intersections.
To create progression, the system schedules green tunnels based on demand. These tunnels are synchronized bands of green lights progressing platoons of vehicles through the corridor. Communications between signals allow the downstream signals to anticipate the arrival of platoons given their knowledge of segment travel times. In this way, InSync can predict platoon arrival, serve the coordinated movement a green signal and avoid slowing or stopping progression. Vehicles traveling close to the speed line can expect to travel through the InSync intersections without stopping.
By anticipating the arrival of platoons, InSync reduces delay caused by slowing down, stopping and start up. This attribute allows more vehicles to progress through a series of intersections and reduces fuel use, harmful emissions and the circumstances conducive to traffic accidents.
InSync initiates green tunnels throughout the day. Tunnel frequency and duration are determined by traffic demand within user-defined ranges. As traffic needs change, InSync can vary the duration and frequency of green tunnels to best support traffic conditions.
The fourth adaptive property of InSync (in addition to phasing, sequencing and green time allocation) is its flexibility with tunnel duration, which is the ability to alter the green time served to a coordinated movement. If the traffic volume on the coordinated movement approach at an intersection decreases and reaches a minimum threshold, InSync can truncate the duration of the green tunnel at that intersection. The intersections downstream also have the independent ability to truncate the tunnel if demand decreases. This ensures coordinated green tunnels, while scheduled, remain adaptive to real-time traffic demand. The extra time is distributed to the local optimizer to minimize intersection delay.
The fifth adaptive property of InSync is period length – the time between serving the coordinated movements. A period is the time interval from the start of a tunnel to the start of the same tunnel later in time. If configured to allow dynamic period lengths, InSync can alter the frequency of green tunnels within a certain range. By adapting period length to real-time traffic demand, InSync generates more or fewer green tunnels inside a given timeframe.
In the periods between green tunnels, InSync switches to the local optimizer to intelligently serve the best phase pairs and green times based on real-time, actual traffic demand. InSync serves these minor movements at each individual intersection (local optimizer) before the next platoon of vehicles moves through the intersection (global optimizer).