1.3 ASCT Detection Layout & System Health Monitoring

1.3 ASCT Detection Layout & System Health Monitoring

No component of an Adaptive Signal Control Technology (ASCT) installation is more critical than the traffic detection infrastructure. While standard semi-actuated traffic signals function acceptably with basic presence detection that merely confirms a vehicle is waiting, ASCT algorithms rely directly on continuous numerical streams of volume counts, lane occupancy, vehicle speeds, and queue discharge headways. A malfunction in a single detection zone can degrade an entire coordinated arterial corridor.

---

1. Detection Fidelity: ASCT vs. Basic Actuated Control

The Count Accuracy Threshold

• Basic Actuation: Standard NEMA TS1/TS2 actuated controllers utilize detector inputs primarily to place phase calls and extend green time via passage timers. If a detector undercounts or overcounts by 15% to 20%, the controller continues to operate safely, as passage timers bridge small detection gaps.
• ASCT Optimization: ASCT optimization engines use volume counts to calculate link capacities, flow ratios ($y = V/s$), degrees of saturation ($DS$), and Cyclic Flow Profiles. A 10% error in volume count directly introduces a 10% error in calculated saturation, resulting in incorrect cycle length expansion, split misallocation, and broken progression bands.
• Industry Mandate: ASCT deployments strictly mandate a sustained volume count accuracy of $\ge 95%$ across all traffic volume regimes, and an occupancy accuracy within $\pm 2%$ under all environmental conditions (day, night, rain, fog, and sun glare).

Mathematical Formulation of Detector Occupancy

Occupancy is the fundamental parameter used by ASCT to determine traffic density and queue presence. Mathematically, Percent Occupancy ($Occ$) over an evaluation time window $T$ is defined as: $Occ = \left( \frac{\sum_{k=1}^{M} t_{\text{occ},k}}{T} \right) \times 100\%$ Where:

• $M$ = total number of vehicle actuations during time window $T$.
• $t_{\text{occ},k}$ = duration of time that vehicle $k$ continuously activates the detection zone (seconds).
• $T$ = total duration of the observation interval (typically 60 seconds or one cycle length).

Relationship Between Occupancy, Volume, Speed, and Density

Occupancy is directly related to macroscopic traffic density ($k$, vehicles per mile) and average travel speed ($S$, mph): $Occ \approx \frac{V \cdot (L_v + L_d)}{S \cdot 52.8}$ Where:

• $V$ = traffic volume (vehicles per hour).
• $L_v$ = average vehicle length (typically assumed as 18 to 20 feet for passenger vehicles).
• $L_d$ = physical length of the detection zone along the roadway (feet).
• $S$ = average vehicle speed (mph).
• $52.8$ = unit conversion factor (5,280 ft/mile divided by 100% occupancy).

Operational Thresholds:

• $Occ < 15%$: Free-flow arterial conditions; vehicle headways are large.
• $Occ = 15% - 25%$: Optimal flow at or near roadway capacity.
• $Occ > 30%$: Traffic flow is breaking down; standing queues are beginning to encroach into the detection zone.
• $Occ > 50%$: Severe congestion, standing queues, and downstream bottleneck spillback.

---

2. ASCT Detection Layout & Functional Zones

A comprehensive ASCT arterial installation organizes detection into three distinct functional zones:

A. Upstream Departure / System Detectors

• Physical Placement: Installed on the departure side of the upstream intersection, typically 50 to 150 feet downstream of the upstream stop bar, or 500 to 1,000 feet upstream of the downstream target intersection.
• Operational Purpose: Captures clean vehicle departure counts and platoon shapes immediately as vehicles exit the upstream intersection. Because this zone is positioned beyond the physical reach of downstream standing queues, it provides uncorrupted Cyclic Flow Profiles (CFPs) essential for SCOOT and ACS Lite dispersion calculations.

B. Advance Dilemma-Zone / Platoon Arrival Detectors

• Physical Placement: Positioned upstream of the stop line at a distance calculated from the approach design speed ($v$). For high-speed approaches ($v \ge 45\text{ mph}$), placement follows standard dilemma-zone equations: $d_{\text{advance}} = 1.47 \cdot v \cdot t_{\text{pr}} + \frac{v^2}{30 \cdot (0.35 \pm G)}$ where $t_{\text{pr}}$ is perception-reaction time (typically 1.0 to 1.5 seconds) and $G$ is roadway grade. For 45 mph, advance detectors are positioned approximately 330 to 400 feet upstream.
• Operational Purpose: Detects the arrival trajectory of oncoming vehicle platoons, enabling the ASCT engine to calculate the exact timing for phase hold commands, ensuring green bands open prior to platoon arrival and protecting vehicles in the dilemma zone.

C. Stop-Bar Presence Detectors

• Physical Placement: Located immediately at the stop line, covering each individual approach lane across a length of 40 to 60 feet (utilizing sequential 6 ft $\times$ 6 ft inductive loops or equivalent video/radar presence zones).
• Operational Purpose: Measures instantaneous queue presence, confirms queue discharge clearance, calls secondary/left-turn phases, and measures the Degree of Saturation ($DS$) during green intervals (crucial for SCATS and InSync).

D. Downstream Departure Receiving Loops

• Physical Placement: Located on the receiving lanes immediately entering the departure link.
• Operational Purpose: Verifies that vehicles serviced during the green interval successfully clear the intersection, detecting downstream queue spillback before it backs up into the intersection grid.

---

3. Automated Detector Health Diagnostics & Failure Modes

Because external environmental conditions and physical roadway wear degrade sensor fidelity, modern ASCT supervisory software executes automated real-time health diagnostic algorithms. Technicians must understand the four primary detector failure modes:

1. Chattering / Rapid Oscillation

• Physical Cause: Loose terminal connections in cabinet backpanels, cracked loop wire insulation rubbing against conduit walls, moisture inside splice kits, or a defective detector amplifier capacitor.
• Symptom: The sensor outputs dozens of rapid on/off pulses per second without vehicle presence.
• Diagnostic Identification: Algorithmic flagging when a detector registers $>10\text{ pulses/second}$ or $>1,000\text{ actuations/hour}$ accompanied by an average occupancy of $<2%$.

2. Stuck-On / Constant Call

• Physical Cause: Direct short-to-ground in lead-in cable, damaged loop wire, amplifier relay welded closed, or a video detection zone tracking a parked vehicle, shadow, or camera sway.
• Symptom: Continuous 100% active call state transmitted to the controller.
• Diagnostic Identification: Algorithmic flagging when continuous presence exceeds a programmed maximum threshold (e.g., $>15\text{ minutes}$ during off-peak periods, or $>30\text{ minutes}$ during peak hours) without a momentary dropout.
• Catastrophic Impact on ASCT ("Runaway Green"): If an upstream system detector fails stuck-on, the ASCT optimization engine interprets the continuous call as an infinite, high-density vehicle platoon approaching the intersection. The algorithm responds by driving the cycle length to its maximum limit and expanding the phase split to its ceiling, starving cross-street and left-turn phases of green time and causing gridlock.

3. Pulse Dropout / Open Circuit

• Physical Cause: Physical severance of inductive loop wire due to road milling, utility trenching, or open splice connections.
• Symptom: Zero actuations recorded; infinite resistance across loop terminals.
• Diagnostic Identification: Algorithmic flagging when a detector registers zero counts over an extended window (e.g., 2 hours) during daytime peak hours while adjacent parallel lanes record heavy traffic.

4. Volume-to-Occupancy Inconsistency

• Physical Cause: Camera lens dirt/spider webs, optical occlusion by heavy commercial vehicles, or radar multipath reflection.
• Symptom: Disproportionate readings, such as 300 vehicles/hour with 80% occupancy (indicating severe camera freeze) or 1,200 vehicles/hour with 1% occupancy (indicating electrical noise).

---

4. Fail-Soft Protocols & Fallback Hierarchies

When automated diagnostics detect a compromised sensor, the ASCT supervisory architecture initiates a deterministic Fail-Soft Hierarchy to prevent corridor breakdown:

[Level 1: Virtual Data Synthesis]
  - Synthesize missing volume/occupancy from adjacent lanes or historical trends
       │ (If failure persists or involves critical single-lane movement)
       ▼
[Level 2: Approach Split Locking / Recall]
  - Place affected phase on Soft Recall or fixed historical split
  - Continue adaptive optimization on remaining healthy movements
       │ (If critical advance/upstream arterial detection drops)
       ▼
[Level 3: Local Intersection Isolation]
  - Drop intersection from adaptive network
  - Revert controller to local actuated-coordinated Time-of-Day (TOD) plan
       │ (If central communications or corridor master drops offline)
       ▼
[Level 4: Corridor-Wide Time-Base Fallback]
  - Entire corridor reverts to local time-base coordinated backup plans
  - Progression preserved via internal controller real-time clocks

---

5. Preventative Maintenance Protocols & Field Tolerances

To ensure sustained ASCT operational fidelity, field technicians must adhere to rigorous maintenance specifications:

Inductive Loop Diagnostic Testing

• Insulation Resistance to Ground (Megger Test): Must be tested using a 500V DC megohmmeter between the loop lead-in conductor and earth ground (with lead-in disconnected from the cabinet detector rack):
  • $>500\text{ M}\Omega$: Excellent, pristine new installation condition.
  • $100 - 500\text{ M}\Omega$: Good, fully acceptable operational health.
  • $10 - 100\text{ M}\Omega$: Deteriorating condition; moisture intrusion in conduit or splice kit.
  • $<10\text{ M}\Omega$: Critical failure threshold; immediate splice excavation and loop re-sealing required.
• Loop DC Series Resistance: Measured with a calibrated multimeter across the two lead-in terminals; acceptable range is $0.5\text{ to }5.0\ \Omega$. Resistance $>10\ \Omega$ indicates corroded splices or loose terminal screws.
• Inductance ($L$): Nominal loop system inductance must measure between $50\ \mu\text{H}$ and $700\ \mu\text{H}$. Inductance below $50\ \mu\text{H}$ causes amplifier oscillator instability; inductance above $700\ \mu\text{H}$ severely reduces sensitivity to high-bed trucks and motorcycles.

Non-Intrusive Detection Maintenance (Video, Thermal, Radar)

• Video & Thermal Sensors: Semi-annual cleaning of optical glass faceplates to remove diesel exhaust film, road spray, and spider webbing; verify sun-shield position; inspect cabinet video surge suppressors; verify camera pan-tilt alignment against fixed reference landmarks.
• Radar / Wavetronix Sensors: Verify boresight targeting alignment using precision scopes; inspect radome cover for cracking or UV degradation; recalibrate range-bin lane boundaries following any asphalt resurfacing or lane restriping projects.