PaperlayerView exact revision on arXiv

Source-linked AI summary

Integrating Traffic Noise Emission Modelling into Variable Speed Limit Control

Jiawen Meng, John Pravin Arockiasamy, Alexey Vinel

arXiv:2609.01339v1cs.RO

TL;DR

Road traffic noise is rarely an explicit real-time objective in VSL, despite the environmental burden and the established relationship between speed and noise emissions. The paper integrates aggregated traffic-state estimation with a simplified CNOSSOS-EU indicator and dynamically selects discrete speed stages. In simulation, the adaptive strategy reduced 24-hour equivalent sound level by 2.9 dB(A) while keeping average speed approximately 11.3 km/h higher than static low-speed control.

  • Problem

    Traffic noise is rarely treated as an explicit real-time operational objective in VSL, despite its persistent environmental and public-health burden.

  • Method

    The framework combines real-time traffic sensing, a simplified CNOSSOS-EU-based emission indicator, and dynamic discrete-stage VSL control.

  • Results

    2.9 dB(A) reduction in 24-hour equivalent sound level relative to uncontrolled baseline (S1), with average speed approximately 11.3 km/h higher than static low-speed configuration (S3).

  • Takeaways & Limitations

    Selective, time-varying speed reductions can mitigate traffic-induced noise without imposing a continuous mobility penalty.

  • Takeaways & Limitations

    The simplified aggregated CNOSSOS-based formulation omits tyre, road-surface, powertrain, and vehicle-specific operating factors and assumes reliable traffic-state estimation.

Abstract

from arXiv · show

Road traffic noise remains a major environmental challenge, yet most speed management strategies are static and do not respond to short-term variations in traffic noise emissions. Although variable speed limit (VSL) systems are widely deployed for safety and congestion mitigation, traffic noise is rarely treated as an explicit operational control objective. This paper proposes a noise-aware VSL framework that integrates aggregated traffic-state estimation with a simplified CNOSSOS-EU-based emission indicator. A stage-based controller with time-varying reference thresholds dynamically adjusts discrete speed-limit levels in response to estimated emission conditions. The framework is evaluated using microscopic traffic simulation calibrated with empirical motorway data and replicated across multiple stochastic realisations. Over a 24-hour evaluation period, the adaptive strategy reduces the receiver-based equivalent sound level by 2.9 dB(A) relative to unrestricted traffic conditions, while maintaining an average vehicle speed approximately 11.3 km/h higher than a permanently imposed low-speed regime. Period-wise analysis shows that speed reductions are activated selectively when emission levels approach calibrated targets, rather than enforcing a constant intermediate limit. Traffic stability indicators reveal moderate increases in speed variability compared with unrestricted operation, but substantially lower braking intensity than under uniform low-speed enforcement. These results demonstrate the feasibility of integrating environmental performance indicators into operational speed control, providing a practical complement to conventional infrastructure-based noise mitigation measures.

I. INTRODUCTION

Road traffic noise is a persistent environmental and regulatory challenge, while conventional mitigation is static and noise is rarely an explicit real-time VSL objective. This paper introduces a noise-responsive VSL framework and evaluates its noise–traffic trade-off under realistic motorway demand.

  • Motivation: Approximately 92 million people in Europe are exposed to day–evening–night noise levels exceeding 55 dB(A).The cited threshold is associated with adverse cardiovascular and psychological health outcomes.
  • Motivation: Infrastructure-based measures such as noise barriers and low-noise pavements are static, capital-intensive, and difficult to adapt after deployment.These constraints motivate complementary operational strategies that can dynamically influence traffic-induced noise emissions.
  • Research gap: Vehicle speed, traffic flow, and fleet composition are primary determinants in standardised traffic-noise emission models.Empirical evidence links lower operating speeds with measurable reductions in traffic-noise levels.
  • Research gap: Traffic noise is rarely treated as an explicit real-time operational objective in VSL research, which more often targets fuel consumption or air-pollutant emissions.Previous traffic-management studies primarily used offline scenario analyses or predefined simulated control strategies.
  • Contributions: The proposed framework integrates real-time traffic sensing with a simplified CNOSSOS-EU-based emission indicator to select speed stages dynamically.The controlled motorway is divided into upstream perception and adjustment and downstream noise-sensitive target zones.
  • Contributions: The study provides an operational pathway for standardised noise-emission modelling in VSL and quantitative evidence on noise reduction and traffic-performance impacts.Microscopic simulation is calibrated with BASt detector data to represent realistic demand and fleet composition.

II. METHODOLOGY

The methodology combines aggregated traffic-state sensing, a simplified CNOSSOS-EU emission indicator, and segment-based motorway modelling. Traffic flow and fleet composition are aggregated by vehicle category and used to estimate emissions without explicit propagation modelling.

  • Framework architecture: The motorway is partitioned into an upstream perception and adjustment zone (E1) and a downstream noise-sensitive evaluation zone (E2).Speed adaptations are largely completed before vehicles enter E2.
  • Traffic-state estimation: Vehicle observations at E1 are aggregated over a sliding window of length T_win and updated every control interval T_ctrl.Traffic-state information is assumed available from roadside detection or connected-vehicle technologies.
  • Traffic-state estimation: For each CNOSSOS vehicle category m, time-averaged flow Q_m is computed and combined with fleet composition as input to the emission model.The aggregated traffic inputs support a segment-level emission indicator.
  • Noise emission model: The simplified CNOSSOS-EU formulation retains sensitivity to traffic flow, speed, and fleet composition while omitting explicit propagation modelling.CNOSSOS-EU is presented as the harmonised European framework for environmental noise assessment.
  • Noise emission model: Single-vehicle noise is separated into rolling and propulsion components before A-weighting and energetic summation across octave bands.The formulation applies to vehicle category m at speed v.
  • Noise emission model: Segment-level directional emission is computed from observed category flow Q_m and corresponding mean operating speed v_m, then summed energetically across categories.This produces the total segment-level emission indicator used by the control logic.

D. Variable Speed Limit Control

The VSL controller evaluates discrete speed-limit stages against a time-dependent noise threshold and selects the least restrictive compliant stage. Persistence rules and nighttime enforcement are used to limit oscillation and reflect stricter nighttime conditions.

  • Stage selection: The active speed-limit stage is selected so the predicted noise emission indicator does not exceed the time-dependent reference L_target(t).Candidate stages are evaluated using observed traffic flows and stage-implied speeds.
  • Stage design: The controller uses N = 5 discrete speed-limit stages ordered from less restrictive to more restrictive operation.Passenger-vehicle limits converge toward HGV limits before both are reduced jointly.
  • Stage design: Passenger vehicles and HGVs receive separate limits, with CNOSSOS light vehicles and powered two-wheelers assigned to the passenger-vehicle limit.Medium-heavy and heavy categories are assigned to the HGV limit.
  • Stage selection: The desired stage is the least restrictive candidate satisfying the emission constraint; if none qualifies, stage N − 1 is applied.This preserves the highest feasible speed stage under the stated constraint.
  • Stability logic: A desired-stage direction must persist for N_persist = 5 consecutive control intervals before the controller moves one stage.The rule is intended to prevent oscillatory behaviour caused by short-term traffic fluctuations.
  • Nighttime operation: During the core nighttime period outside transition windows, the most restrictive stage is enforced directly.This reflects stricter nighttime noise limits and avoids unnecessary switching under very low demand.

III. SIMULATION AND EVALUATION

The evaluation models a calibrated three-kilometre motorway corridor in SUMO using BASt-derived demand and fleet classes. Multiple behavioural profiles and independent random seeds support assessment across realistic stochastic traffic conditions.

  • Simulation setup: The simulated corridor is a 3 km unidirectional, three-lane motorway divided into warm-up, E1, and E2 segments of 1 km each.Geometry and demand derive from BASt permanent counting data on the A66 (R1) in 2023.
  • Demand and fleet: Hourly traffic volumes and BASt vehicle classes are mapped to four CNOSSOS-EU categories implemented in SUMO through hourly vehicle-flow definitions.The categories are light, medium heavy, heavy, and powered two-wheelers.
  • Demand and fleet: The demand profile contains two daytime peaks and an increased HGV share during nighttime hours.This pattern is shown in the BASt-derived diurnal traffic profile.
  • Driver modelling: Driver heterogeneity is represented by conservative (30%), normal (50%), and aggressive (20%) behavioural profiles within the Krauss car-following model.Remaining model parameters use SUMO defaults.
  • Evaluation protocol: The simulation spans 25 hours, with the first hour discarded to yield a 24-hour evaluation period.Each scenario is replicated using five independent random seeds and results are averaged across replications.

B. Experimental Scenarios

Three control configurations compare unrestricted, dynamically noise-responsive, and permanently low-speed operation. The dynamic strategy uses period-specific CNOSSOS-based targets with smoothed transitions to select speed stages.

  • Scenario definitions: Three scenarios compare unrestricted operation, noise-aware dynamic VSL, and a permanently imposed 60 km/h low-speed regime.S1 is the mobility reference, S2 dynamically selects stages from 100/80 to 60/60 km/h, and S3 provides a fixed low-noise reference.
  • Target calibration: The time-varying control target is derived from period-specific average baseline emission levels computed with the CNOSSOS-based model.Assessment periods are day, evening, and night.
  • Target calibration: ΔL = 1.5 dB keeps the controller within the effective range of the defined speed-stage set.Larger values would permanently activate the most restrictive stage, whereas near-zero values would rarely trigger adjustments.
  • Target calibration: Linear interpolation within symmetric transition windows produces a continuous piecewise-linear target profile Ltarget(t).Figure 3 shows the baseline emission indicator with mean ±1σ across five seeds and the calibrated target profile.

D. Control Parameter Settings

Controller parameters are fixed across scenarios after target calibration. Traffic states are aggregated over Twin, and stage updates run every Tctrl subject to persistence and smoothing settings.

  • Parameter settings: All controller parameters are fixed across scenarios and summarised in Table I.The settings follow the calibrated target profile.
  • Parameter settings: Traffic states are aggregated over Twin and the controller executes every Tctrl.Stage updates follow the persistence condition defined for the controller.
  • Parameter settings: Tsmooth denotes the transition window used to connect period-specific control targets.The parameter corresponds to the transition window defined in Section III-C.

E. Noise Evaluation Method

Noise evaluation uses microscopic vehicle contributions at a virtual receiver within segment E2. Energetic summation produces a sliding 60 s equivalent continuous A-weighted sound level under common propagation assumptions.

  • Receiver evaluation: Acoustic performance is evaluated with receiver-based A-weighted sound pressure levels in segment E2.Unlike the aggregated control indicator, evaluation uses instantaneous positions of individual vehicles.
  • Receiver evaluation: The virtual receiver is positioned at E2’s midpoint, 25 m laterally from the roadway centreline and 4 m above ground.Vehicles within E2 contribute according to instantaneous position, speed, and category at 10 s intervals.
  • Receiver evaluation: Each vehicle’s instantaneous receiver level uses its vehicle-category source level, source-receiver distance, and a −11 dB free-field divergence term.The distance is three-dimensional.
  • Receiver evaluation: Energetic summation combines vehicle contributions, and Leq,60 quantifies noise over a sliding 60 s window.Atmospheric absorption, ground effects, and shielding are neglected consistently across scenarios.

IV. RESULTS

Across the 24-hour evaluation, dynamic VSL reduces receiver noise relative to unrestricted traffic while remaining less restrictive than permanent low-speed control. Period-wise and time-series results show intermediate acoustic performance with moderate control-related variability.

  • Aggregate performance: Over 24 hours, S1 records 67.3 dB(A) and 93.5 km/h, while S3 reduces noise by 4.7 dB but lowers mean speed to 57.2 km/h.These configurations establish the unrestricted and permanent low-speed reference endpoints.
  • Aggregate performance: 2.9 dB(A) lower Leq is achieved by S2, with an average speed of 68.5 km/h.S2 avoids the full mobility penalty associated with permanent low-speed operation.
  • Period-wise analysis: 2.7 dB and 2.8 dB reductions occur during daytime and evening, respectively, relative to S1.S2 maintains substantially higher mean speeds than S3 in both periods.
  • Period-wise analysis: At night, S2 enforces the most restrictive stage but remains slightly noisier than S3 because smoothing and persistence create gradual period-boundary transitions.S3 applies the low-speed limit continuously.
  • Temporal analysis: The time-series ordering S1 > S2 > S3 persists throughout the day.Figure 4 reports mean ±1σ receiver levels across five random seeds.
  • Temporal analysis: Nighttime fluctuations are larger across all scenarios because low demand makes individual vehicle passages dominate the 60 s equivalent level.The passage attributes these fluctuations to statistical effects rather than control instability.

C. Control Behaviour and Traffic Stability

The adaptive VSL strategy changes speed stages selectively and maintains stable control behaviour while introducing less traffic disruption than permanent low-speed enforcement.

  • Control Behaviour: Stage changes occur in structured increments without rapid back-and-forth switching, indicating stable discrete control behaviour.Figure 5 shows a representative S2 trajectory across assessment periods.
  • Control Behaviour: S2 stage transitions remain moderate, averaging 8.2 during daytime and 6.2 during evening.Daytime operation concentrates mainly in Stages 2 and 3, while evening operation is more dispersed, with Stages 1 and 3 most frequent.
  • Traffic Stability: Passenger-car speed standard deviation rises from 0.78 m/s under S1 to 1.6 m/s under S2 and 2.39 m/s under S3.Maximum deceleration follows the same scenario ordering: 1.76, 3.02, and 3.55 m/s2, respectively.
  • Traffic Stability: S2 introduces additional variability relative to unrestricted operation but does not produce excessive braking or pronounced stop-and-go behaviour.S3 induces substantially stronger braking responses and larger speed dispersion than S2.
  • Traffic Stability: Acceleration–deceleration reversals remain nearly unchanged across scenarios, suggesting oscillatory behaviour primarily reflects underlying car-following dynamics rather than VSL logic.Overall, S2 achieves noise mitigation with a moderate and controlled impact on traffic stability.

V. CONCLUSION AND DISCUSSION

The paper integrates a simplified noise-emission indicator into operational motorway speed control and evaluates its environmental and traffic effects. The adaptive strategy reduces noise while preserving more mobility and causing less disruption than static low-speed enforcement, but broader validation and richer modelling remain necessary.

  • Conclusion: The framework embeds a simplified CNOSSOS-EU-based emission indicator into operational motorway speed control for real-time noise mitigation.Environmental performance metrics are incorporated directly into the regulation logic.
  • Conclusion: S2 reduces the 24-hour equivalent sound level by 2.9 dB(A) relative to S1 while preserving an average speed 11.3 km/h higher than S3.S2 also exhibits milder traffic stability impacts than permanent restriction.
  • Discussion: Selective, time-varying speed reductions mitigate traffic-induced noise without imposing a continuous mobility penalty.Compared with uniform low-speed regulation, adaptive control provides environmental benefits with reduced disruption to traffic dynamics.
  • Limitations: The current implementation assumes reliable traffic-state estimation and uses an aggregated emission formulation that omits several vehicle, tyre, road-surface, and powertrain factors.These omissions may cause deviations between predicted and realised emission levels.
  • Limitations: The evaluation compares S2 only with an uncontrolled baseline and a static low-speed regime, without traffic-oriented VSL controllers or explicit overall-system optimisation.Future extensions could incorporate travel time, energy use, and air pollutant emissions in multi-objective control.
  • Limitations: Validation is limited to a single-corridor motorway case study with controlled boundary conditions, leaving scalability under heterogeneous networks and coordinated multi-segment control unresolved.Larger-network studies are needed to assess operational robustness.
  • Conclusion: The study demonstrates feasibility and establishes a practical basis for environmentally informed VSL strategies.This conclusion follows from embedding environmental performance indicators within motorway traffic control.
Loading 2609.01339v1…