CCTV FIXED CAMERA GRID — INTERSECTION

REC ● LIVE

View:

| Vision:

| 0 active

CAMERA CONTROL PANELS

per-camera power · resolution · FPS · model

🚦 SIGNAL-HEAD TIMING (optional)

click to read the real signal timing from cameras ▸

🎥 Vision / Camera Feed (Simulated)

simulated YOLO-style detection — click to expand ▸

Approach:

—

Car: 0

Bus: 0

Truck: 0

Motorcycle: 0

Avg conf: —

% Heavy (bus+truck): —

CAMERA TOPOLOGY MATRIX

—

PTZ:

SPLIT-PHASE INTERSECTION SCHEMATIC

Idle

— hover a CCTV cell to highlight its approach

V2X COOPERATIVE TELEMETRY · OBU/RSU IoT BUS

● C-V2X PC5 · DSRC

BSM Rx

SPAT Tx

PSM Rx

RSU

Fusion Gain

ICP Conflicts

Latency

0 ms

SINR

0 dB

GLOSA advisories 0

V2X-TSP request NONE

GREEN-WAVE CORRIDOR · 3-INTERSECTION NETWORK

SYNCED

Optional bus priority (TSP) — extends/advances green for approaching transit.

🤖 AUTOMATED TRAFFIC ANALYST AI INSIGHT STREAM

System nominal. Awaiting telemetry…

Active Phase —

0 sec remaining

Idle — press Start Analysis

Lost Time L = — s

WEBSTER OPTIMUM CYCLE Y = —

Co = (1.5·L + 5) / (1 − Y) · awaiting demand…

NTCIP RING & BARRIER

⚠ EMERGENCY OVERRIDE

🛑 CONFLICT MATRIX FAIL-SAFE ENGAGED — ALL APPROACHES FLASHING RED/YELLOW

Total Vehicles

System V/C Ratio

0.00

Avg Delay (s)

HCM LOS

Incr. Delay d₂ (s)

EVP Triggers

Queue Clearance

100%

V2X Fusion

Sys FPS / Skipped

0 / 0

Per-Lane Delay Trend

📋 EVALUATOR — Existing Signal Audit

monitoring…

Defects & Remedial Actions

INTERSECTION DESIGN SCHEMATIC

—

🛰 Upload Aerial Blueprint

SIGNAL TIMING DESIGN

HCM / Webster Calculations

PHASING PLAN

APPROACH TIMING

DESIGN SUMMARY

HISTORICAL ANALYTICS DASHBOARD

V/C Ratio Trend (Historical)

LOS Grade Trend (Historical)

Delay Trend (Historical)

Queue Length Trend (Historical)

Chart A · Active Queue Length & Arrival Rate (dual-axis)

Chart B · Avg Delay per Vehicle + HCM LOS thresholds

Phase Transition Event Log

Corridor Coordination

3-signal arterial green-wave analysis (HCM/TRB coordination)

Spacing A→B (m) Spacing B→C (m) Band Speed (km/h) Common Cycle (s) Offset B (s) Offset C (s)

Time–Space Diagram — Through Movement Green Bands

Bandwidth

—

Progression Grade

—

Progression quality (continuous green band / cycle)

—

Reports & Export Center

All engineering reports in one place — download, export or print.

📚 Engineering Codes & Academic References

This simulator's adaptive control logic is grounded in the following authoritative standards and peer-reviewed literature. Each entry notes which idea we adopted for the in-app calculations.

Verified links / DOIs are provided for every academic source; open them to confirm the methods cited.

A · Standards & foundational models

A1 · Highway Capacity Manual (HCM), 7th Edition (2022) — Ch.19 Signalized Intersections

Transportation Research Board (TRB), National Academies of Sciences, Engineering & Medicine.

Idea used: the control-delay decomposition d = d₁·PF + d₂ + d₃ (Eq. 16-9/19-18) drives our per-lane delay & LOS grade. We implement uniform delay d₁, the incremental/overflow term d₂ (time-dependent Kimber–Hollis form), and the LOS A–F thresholds by control delay (s/veh). Full saturation flow adjustment chain: fHV × fP × fRT × fLT × fW × fLU (Exhibit 19-5).

A2 · Webster, F.V. (1958) — "Traffic Signal Settings", Road Research Technical Paper No. 39

Road Research Laboratory, HMSO, UK.

Idea used: the optimum-cycle formula Cₒ = (1.5·L + 5) / (1 − Y) with Y = Σ(qᵢ/sᵢ) over each phase's critical lane, and effective green (Cₒ − L) distributed in proportion to yᵢ/Y. Saturation flow is HCM-adjusted for heavy vehicles (fHV = 100/(100+%HV·(Eₜ−1))) and demand inflated by PHF.

A3 · Zakariya & Rabia (2016) — "Estimating the minimum-delay optimal cycle length based on a time-dependent delay formula"

Alexandria Engineering Journal 55(3):2509–2514. · doi:10.1016/j.aej.2016.07.029

Idea used: Webster over-estimates cycle length at high saturation; we cap/clamp the computed Cₒ toward the HCM time-dependent minimum-delay range for high v/c so the recommended plan stays realistic under congestion.

A4 · NTCIP 1202 — Object Definitions for Actuated Traffic Signal Controller Units

NTCIP (ITE / AASHTO / NEMA).

Idea used: our Dual-Ring state machine mirrors the NTCIP phase/ring model and the Multi-Ring Conflict Matrix enforces the same non-conflicting-phase safety guarantees required of field controllers. Ring-barrier architecture, force-off, and coordination plan structures follow NTCIP 1202 v03 object definitions.

B · Adaptive control & reinforcement learning

B1 · Feng, Lu, Li, Tu & Li (2024) — "Advances in reinforcement learning for traffic signal control: a review of recent progress"

Intelligent Transportation Infrastructure. · doi:10.1093/iti/liaf009

Idea used: contrasts fixed-time / actuated / ATSC (SCOOT, SCATS) vs. RL. Motivates our adaptive green-split heuristic that responds to real-time per-lane demand rather than preset rules.

B2 · Saadi, Abghour, Chiba et al. (2025) — "A survey of reinforcement and deep RL for coordination in intelligent traffic light control"

Journal of Big Data 12:84. · doi:10.1186/s40537-025-01104-x

Idea used: state/action/reward design (queue- and delay-based reward). Informs our congestion metric per approach and the "serve the most-congested lane" priority rule.

B3 · MDPI Infrastructures (2025) — "Traffic Signal Control via Reinforcement Learning: A Review on Applications and Innovations"

Infrastructures 10(5):114. · doi:10.3390/infrastructures10050114

Idea used: reward-shaping that penalises unnecessary phase switching — reflected in our minimum-green enforcement and anti-flicker phase-hold logic.

B4 · MDPI Electronics (2024) — "AI-Based Adaptive Traffic Signal Control System: A Comprehensive Review"

Electronics 13(19):3875. · doi:10.3390/electronics13193875

Idea used: single- vs multi-intersection taxonomy (FL, DP, RL, DRL, hybrid) and reported gains (e.g. MARLIN-ATSC 27–39% wait-time reduction) as design benchmarks for our KPIs.

C · Vision-based detection, counting & deployment

C1 · Wu, Chen, Wang, Chen & Lin (2026) — "Deployable Vision-Based Traffic Signal Adaptation Using YOLO and Edge Computing"

IEEE ACDSA 2026. · doi:10.1109/acdsa67686.2026.11468159

Idea used: extract vehicle count + class + delay from camera feeds and feed a parameterised timing model (interpretable, no training). This is exactly our rule-based, in-browser pipeline; the paper reports 20–30% delay reduction vs fixed-time.

C2 · Aswini, Kumar, Revanth, Faheem & Srinivasu (2026) — "Smart Traffic Signal Control Using Deep RL and Real-Time Video Analytics"

IJERST 22(1). · doi:10.62643/ijerst.2026.v22.n1(2).pp214-218

Idea used: four directional polygon zones (N/S/E/W) mapping the junction to software lanes and computing a normalised congestion level every few seconds — mirrored by our per-approach lane zones and windowed arrival-rate estimate.

C3 · Bandi et al. (2025) — "YOLO based Adaptive Traffic Signals Control & Real-Time Violation Monitoring"

IEEE ICICNIS 2025. · doi:10.1109/icicnis66685.2025.11315786

Idea used: per-lane density counting + emergency-vehicle preemption (immediate green). Informs our multi-class counting and priority/TSP hooks.

C4 · Springer NCA (2025) — "Smart traffic management system using YOLOv11 for real-time vehicle detection and dynamic flow optimization"

Neural Computing & Applications. · doi:10.1007/s00521-025-11434-9

Idea used: multi-class vehicle recognition + density-based signal adjustment; confirms counting all classes (car, truck, bus, motorcycle, bicycle) rather than cars only, as done in our detector class map.

D · Signal timing, clearance & pedestrian standards

D1 · ITE Traffic Engineering Handbook, 7th Edition (2016) — Chapter 4E: Traffic Signal Timing

Institute of Transportation Engineers (ITE).

Idea used: yellow change interval Y = t + v/(2a+2Gg) and all-red clearance R = (W+L)/v derived from approach speed. We implement computeChangePeriod() per ITE 4E.05 with configurable approach speed and intersection width.

D2 · MUTCD, 11th Edition (2023) — Chapter 4C: Signal Phasing & Timing + Chapter 4E: Pedestrian Timing

Manual on Uniform Traffic Control Devices (FHWA, USDOT).

Idea used: pedestrian WALK interval (min 7 s), FDW = crossing width / 1.07 m/s (3.5 ft/s), and pedestrian clearance timing per §4E.06. Our cfgWalk and cfgXWidth inputs feed the pedestrian phase calculation. Also implements all 7 MUTCD signal warrants (W1–W7) for signal warrant analysis.

D3 · AASHTO Policy on Geometric Design of Highways and Streets, 8th Edition (2018)

American Association of State Highway and Transportation Officials.

Idea used: approach storage length and queue clearance design. Our cfgApproachLength auto-calculates storage capacity (length / 7.5 m/veh) per AASHTO guidance for intersection approach design.

D4 · MUTCD Chapter 8D — Railroad Crossing Preemption

Manual on Uniform Traffic Control Devices (FHWA, USDOT).

Idea used: railroad preemption sequence follows MUTCD §8D.07: pre-signal phase → track clearance green → hold → resume normal operation. Our initiateRailPreempt() and preemptTick() implement this state machine.

E · V2X cooperative intelligence & connected vehicles

E1 · SAE J2735 — Dedicated Short Range Communications (DSRC) Message Set Dictionary

Society of Automotive Engineers (SAE International).

Idea used: BSM (Basic Safety Message) at 10 Hz, SPAT (Signal Phase and Timing), MAP (Intersection Geometry), PSM (Personal Safety Message), and RSA (Road Side Alert) message types structure our full V2X telemetry feed. The V2X object and renderV2XFeed() parse and display all SAE J2735 messages.

E2 · IEEE 1609 — Wireless Access in Vehicular Environments (WAVE)

Institute of Electrical and Electronics Engineers.

Idea used: WAVE/DSRC protocol stack for V2X communication. Our RSU (Road Side Unit) status, SINR, MCS, and latency metrics (8–22 ms) reflect the IEEE 1609.x framework for vehicle-to-infrastructure communication. Link quality model includes packet loss and SINR jitter.

E3 · Dempster-Shafer Evidence Theory — Sensor Fusion

Shafer, G. (1976) — A Mathematical Theory of Evidence, Princeton University Press.

Idea used: our Sensor Fusion Engine (SEE) uses Dempster-Shafer confidence blending: video detection (0.6 weight) fused with V2X BSM data (0.4 weight) via OR-fusion logic. This recovers occlusion dead-zones where video under-counts but V2X confirms stopped vehicles.

F · Corridor coordination & green wave

F1 · TRB Special Report 287 (2010) — "Capacity Analysis of Actuated Traffic Signal Systems"

Transportation Research Board, National Academies.

Idea used: our corridor coordination model (3-signal arterial with green-wave offset) follows TRB progression offset methodology: offset = distance / design speed. The computeGreenWaveOffset() and computeCorridorBand() functions implement the time-space diagram analysis for bandwidth maximization.

F2 · Robertson & Bretherton (1991) — "Optimal Control of Traffic Signals — The SCOOT Method"

TRRL Report LR 394, Transport Research Laboratory, UK.

Idea used: SCOOT's online offset optimization principle informs our TOD-based coordination plans (5 plans: AM Peak, Midday, PM Peak, Evening, Weekend) with master clock and force-off logic in coordinationTick().

G · Emissions, safety & environmental modeling

G1 · EPA MOVES (Motor Vehicle Emission Simulator) — Emission Factors

United States Environmental Protection Agency.

Idea used: our EmissionsOptimizer class models CO₂, NOx, HC, and fuel consumption per vehicle type (car, truck, bus, motorcycle) using simplified EPA MOVES emission factors. The multi-objective function J = w₁×delay + w₂×stops + w₃×emissions allows configurable optimization weighting.

G2 · Kalman, R.E. (1960) — "A New Approach to Linear Filtering and Prediction Problems"

Journal of Basic Engineering, 82(1):35–45.

Idea used: our KalmanFilter2D class implements a 2D Kalman filter for bounding-box track smoothing between AI inference frames. State vector [x, y, vx, vy] with predict-update cycle improves count accuracy by filtering jitter and bridging detection gaps.

Note: the in-browser detectors (COCO-SSD / MobileNet via TensorFlow.js) approximate the YOLO pipelines above; custom YOLO weights can be supplied for field-grade accuracy. Reported percentage gains come from the cited studies' own simulations/deployments, not from this demo.

SmartFlow Intersection — Adaptive Traffic Signal Control