Analysis of the impact of heterogeneous platoon for mixed traffic flow: control strategy, fuel consumption and emissions

Gen Li; Le Li; Yangsheng Jiang; Yi Wang; Yunxia Wu; Zhihong Yao

REVIEW 4 major objections 5 minor 3 references

Analysis of the impact of heterogeneous platoon for mixed traffic flow: control strategy, fuel consumption and emissions

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that combining a variable time-gap spacing policy for platoon leaders with constant spacing for followers yields the lowest fuel consumption and emissions in dense mixed traffic once CAVs exceed 80% penetration.

desk verdict Useful comparative simulation of new combination spacing strategies, but the headline ranking rests on zero-delay, O=1 idealization and an unspecified HV model. read the letter →

arxiv 2411.15238 v1 pith:K7KYNIGD submitted 2024-11-22 eess.SY cs.SY

Yunxia Wu , Le Li , Zhihong Yao , Yi Wang , Gen Li , Yangsheng Jiang This is my paper · ORCID

classification eess.SYcs.SY

keywords connectedautomatedvehiclesplatoonmixedtrafficflowspacingcontrolstrategyfuelconsumptionpollutantemissionscar-followingmodelstringstability

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to determine whether combining different longitudinal spacing-control strategies within connected-automated-vehicle (CAV) platoons can improve the fuel consumption and pollutant emissions of mixed traffic flow containing human-driven vehicles. It builds a mixed-traffic model with a probability distribution over vehicle types, proposes ten control strategies built from four spacing policies, and tests them in ring-road simulations at densities from 5 to 100 veh/km and CAV penetration rates from 0 to 100%. Its central finding is that at medium-to-high densities (55-95 veh/km) and penetration above 80%, the combinations VTG1-CS, VTG2-CS, and CTG-CS, where the platoon leader uses a variable time-gap policy and followers use constant spacing, keep the flow stable and safe while yielding the lowest fuel consumption and pollutant emissions. The paper thereby provides evidence for a design rule: the leader of a CAV platoon should buffer the traffic with a variable gap, while followers should close ranks with constant spacing.

What carries the argument

The load-bearing device is the combination spacing strategy: the platoon leader follows a variable time-gap policy (VTG1, VTG2, or CTG) while all followers use constant spacing (CS), so the leader absorbs speed fluctuations through a speed-dependent gap and followers maintain tight fixed gaps. This is paired with a Markov-chain probability model of vehicle distributions in mixed traffic (leader-following HV, platoon leader, platoon follower), and with a stability criterion for homogeneous CAV flow expressed through the $\mathcal{H}_\infty$ norm of the speed-disturbance transfer function. Simulation then feeds speed and acceleration trajectories into a vehicle-specific-power fuel model and an instantaneous emission model to compare strategies.

What would settle it

Re-run the ring-road simulation with a nonzero communication delay (e.g., 0.2 s) in the LPF information flow, or with platoon intensity $O<1$ so CAVs are randomly scattered instead of grouped, and check whether VTG1-CS, VTG2-CS, and CTG-CS still produce the lowest fuel consumption and near-zero emissions at densities 55-95 veh/km and penetration 80-100%; if the ranking changes, the paper's central claim fails under those conditions.

Watch

Extended reading notes

Core claim

The paper's discovery is that combination spacing strategies outperform uniform ones in mixed traffic, and specifically that VTG1-CS, VTG2-CS, and CTG-CS dominate at high CAV penetration. In a single-lane ring-road simulation with all CAVs grouped into platoons of at most four vehicles (platoon intensity $O=1$), these three strategies keep the average fuel consumption curve nearly flat as density rises, keep CO2, NOx, VOC, and PM emissions close to zero at penetration rates of 80-100%, and preserve string stability via leader-predecessor-follower communication. The BS-BS strategy, which uses balanced spacing for every vehicle, behaves oppositely: its fuel consumption and emissions worsen as penetration increases. Under low density (15 veh/km) the ten strategies are nearly indistinguishable, so the proposed combinations matter only when traffic is dense enough for spacing policy to constrain the flow.

Load-bearing premise

The load-bearing premise is that communication and sensor delay are zero (Assumption 1) and that all CAVs are clustered into platoons (platoon intensity $O=1$); if realistic delays or randomly distributed CAVs are introduced, the ranked advantage of VTG1-CS, VTG2-CS, and CTG-CS could shrink or disappear.

Editorial extensions

If this is right

At CAV penetration of 80% or more, deploying VTG1-CS, VTG2-CS, or CTG-CS in a CAV platoon is sufficient to hold fuel consumption and pollutant emissions near their low-density levels even when density reaches 55-95 veh/km.
The BS-BS strategy should be avoided at high CAV penetration in dense traffic because it increases fuel consumption and emissions as the share of CAVs grows.
The choice of spacing strategy matters only above a density threshold; at low density (15 veh/km), any of the ten strategies gives similar consumption and emissions, so control effort can be relaxed.
The three leading combination strategies also preserve traffic stability and safety under the tested assumptions, meaning the fuel and emission benefits are not bought by degraded flow.
At penetration rates below 80%, no single combination dominates; CTG-CTG is relatively good at medium density with low penetration, and the ranking changes, so strategy selection must be penetration-aware.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Beyond the paper: the zero-delay assumption is the main limitation; extending the car-following model with communication delay, as the authors themselves suggest, could erode the advantage of LPF-based CS combinations, since prior work cited in the paper shows LPF platoon stability is sensitive to delay.
Beyond the paper: the probability distribution model could be reused to predict string stability or capacity in mixed flow under different penetration and platoon intensity, not just fuel and emissions.
Beyond the paper: a testable field hypothesis is that in real platoon deployments, a variable-gap leader plus constant-gap followers should produce lower per-vehicle fuel use than uniform CTG or uniform CS when CAV share is high and traffic is dense.
Beyond the paper: the paper fixes platoon intensity at $O=1$ and platoon size at 4; random CAV distributions ($O<1$) could weaken the advantage, so the design rule likely applies most to coordinated, platoon-friendly traffic management.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Here's the short version: this is a useful comparative simulation study, not a breakthrough. The new thing is Table 6's combination spacing strategies (VTG1-CS, VTG2-CS, VTG1-CTG, etc.) applied to mixed traffic, and the systematic fuel/emission comparison at three densities and five penetration rates. The authors deserve credit for unifying controller parameters across strategies, verifying their vehicle-distribution probability model, and openly listing the zero-delay limitation.

The results are plausible: at penetration >=80% and 55-95 veh/km, the three CS-based combos (with CTG/VTG leaders) give the lowest fuel and emissions in their simulated ring. But the central claim is only as strong as the idealization. Assumption 1 sets communication delay to zero, and Section 5.3.1 fixes platoon intensity O=1, which maximizes the number of CS followers using leader information. The stability test in Section 4.3.3 is for a homogeneous CAV stream, not for the mixed platoon with a CTG/VTG leader and CS followers; the 'stability and safety' phrase in the abstract is never backed by a measured safety or string-stability metric in the mixed-flow simulation. The HV car-following model is never specified, which matters because HVs make up 20% of vehicles at p=0.8. There are no error bars, no sensitivity to delay, platoon size, or controller parameters, and no validation against empirical data. The paper's own limitation (1) concedes the delay issue. So the headline ranking might survive under realistic delays, but it is untested; if LPF string stability degrades with 0.1-0.5s delay, as the cited Zhang et al. work suggests, the advantage of the CS combos could shrink.

The citation pattern is fine—they build on Zheng et al. and their own prior work, and they say so. No circularity: the fuel/emission numbers come from simulation, not from fitting a target.

Who should read this: people calibrating or designing CACC spacing policies for mixed traffic, and researchers wanting a benchmark table of combination strategies. It deserves a serious referee, but the referee should push for an explicit HV model, a delay sensitivity analysis, a safety/string-stability metric, and ideally a variation on O.

Referee Report

4 major / 5 minor

Summary. This paper develops a mixed traffic flow model for CAV platoons, derives a Markov-chain-type probability distribution for vehicle types (LV1, LV2, PV), and proposes ten spacing strategies: four single strategies (CTG, VTG1, VTG2, BS) and six combinations with CS or CTG as follower control. The strategies are compared in single-lane ring-road simulations using VSP-based normalized fuel consumption and the Int Panis et al. emission model. The main claim is that at medium-to-high densities (55–95 veh/km) and CAV penetration above 80%, the VTG1-CS, VTG2-CS, and CTG-CS strategies can ensure traffic flow stability and safety while significantly reducing fuel consumption and emissions. The probability distribution model is verified against simulation with high R² values except for the degenerate LV1 case, where RMSE is used.

Significance. If the headline result were robust, the paper would provide a useful systematic comparison of combination spacing strategies for CAV platoons in mixed traffic, with practical implications for platoon control design. The use of established fuel and emission models, the verification of the vehicle-distribution probability model (R² close to 1 for O=0 and O=1, Table 1), and the breadth of the ten-strategy comparison are genuine strengths. However, the central claims are established only under an idealized simulation setup (zero communication delay, O=1 all-CAV platooning, unspecified HV car-following behavior), and the statements about stability and safety are not backed by any measured stability or safety metric. The current evidence supports a conditional, simulation-specific ranking rather than the broad dominance claim in the abstract.

major comments (4)

[§3.3, §5.3.1] The car-following model for human-driven vehicles is never specified. Section 3.3 defines HV car-following modes but gives no equation or parameter set for HV behavior, and Section 5.3.1 lists only CAV-relevant parameters. Since the headline result is stated for p≥0.8 and not p=1, the HV dynamics are part of the simulated mixed flow, and the fuel/emission comparisons at p=0.8 depend on an unspecified HV model. This makes the results non-reproducible and weakens the quantitative ranking. The HV model should be stated explicitly, and its parameters justified.
[§4.3.3, abstract, §6] The claim that VTG1-CS, VTG2-CS, and CTG-CS can 'effectively ensure traffic flow stability and safety' is not supported by the reported analyses. Equation (25) is a stability test only for homogeneous CAV traffic with identical strategies and is used only to check the VTG1/VTG2 controller parameters; no stability condition is derived or measured for the mixed traffic flow or for the combination strategies themselves. Moreover, no safety metric (e.g., time-to-collision, minimum gap, collision count) and no flow-stability metric (e.g., speed or acceleration variance) is reported in the simulation results. The wording of the abstract and conclusions should be limited to what the simulations actually show, or the corresponding metrics should be added.
[Assumption 1, §5.3.1, §4.1] The main simulations assume zero communication delay and platoon intensity O=1, which maximizes the number of vehicles following the CS law under LPF topology. The recommended VTG1-CS, VTG2-CS, and CTG-CS strategies rely on leader acceleration, velocity, and position through Eq. (9), and the authors themselves cite Zhang et al. (2020a) showing that LPF platoon stability is sensitive to communication and sensor delay. With nonzero delay or with random CAV distributions (O=0), the relative advantage of these three strategies could shrink or disappear. Since this setting is load-bearing for the central ranking, the paper should include sensitivity experiments with realistic delays and at least one non-platoon intensity (e.g., O=0), or explicitly restrict the conclusions to the idealized setting.
[§5.3] The simulation results are presented without any uncertainty quantification or sensitivity analysis. Each strategy is compared using a single deterministic simulation run; there are no error bars, multiple random seeds, or perturbations of the controller parameters, time gaps, and BDBM parameters listed in Tables 2–5. The word 'significantly' in the abstract and conclusions is therefore not statistically supported. Adding at least a small Monte Carlo or parameter-perturbation analysis would strengthen the comparative ranking substantially.

minor comments (5)

[§4.1, Eq. (9)] In Eq. (9), the symbol e_{i,leader,CS}(t) appears in two consecutive terms with the same notation; the last term presumably contains the time derivative of e_{i,leader,CS}(t). Please clarify the notation.
[§4.3.3, Table 4] Table 4 reports a single value 0.0624 for the VTG1 stability condition, but Eq. (26) is an inequality involving three partial derivatives. Please state how the number is computed and which side of the inequality is evaluated.
[§5.3.2] The text uses 'NFT' where the index is defined as 'NFF' (normalized fuel factor) in Eq. (33). Please correct the typo.
[References] Several references are incomplete: 'Hung et al. (n.d.)' and 'Yang et al. (n.d.)' lack publication details, and 'Y. Qin (2019)' is cited in an inconsistent author–date style.
[§5.3.1] The simulation setup says the ring road is 1000 m and vehicles are initially evenly distributed, but the number of vehicles is not stated as a function of density. Please specify how many vehicles are used for each density value and how platoons of maximum size 4 are formed when the number of CAVs is small.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular reduction in the central simulation results; the fuel/emission rankings are genuine simulation outputs, and the only self-citations are auxiliary and non-load-bearing.

full rationale

The paper's central claims about fuel consumption and emissions are numerical simulation outcomes obtained from explicit car-following models (Eqs. 9, 12, 17, 21, 27), a VSP/NFR fuel model (Eqs. 30-33), and an emissions model (Eq. 34), under stated density and penetration settings. No parameter in those models is fitted to the fuel/emission target being predicted, and the rankings of VTG1-CS, VTG2-CS, and CTG-CS emerge from the simulation rather than being imposed. The stability condition in Eqs. (25)-(26) is a standard disturbance-transfer-function test used to check parameter values, not to define the comparison. The probability distribution model in Section 3.4 is taken from prior work by overlapping authors, but it is validated against an independent drawing/simulation in Section 3.5 and does not determine the fuel/emission comparisons. The zero-delay assumption (Assumption 1) and the O=1 platoon setting are acknowledged limitations (Section 6, limitation 1) that may weaken external validity, but they are scope conditions, not circular inputs. No equation is shown to be equivalent to its inputs by construction, and no fitted quantity is relabeled as a prediction. Minor self-citations (Sections 3.4 and 4.4) are not load-bearing for the central result.

Assumptions & free parameters 8 free parameters · 7 assumptions · 0 invented entities

The central claim rests on an idealized simulation: zero delay, O=1 platoon intensity, an unstated HV car-following model, and literature parameter values. These are domain assumptions rather than fitted parameters; the main quantitative outcomes are sensitive to them.

free parameters (8)

Platoon intensity O = 1 (simulation setting)
All CAVs are aggregated into platoons in the main simulations, favoring strategies that use leader information.
Maximum platoon size S = 4
Taken from Zhu and Tasic (2021); affects P_LV2 and P_PV and hence platoon composition.
CTG time gaps h_LV, h_PV = 1.1 s / 0.6 s
Chosen for platoon leader and follower; directly sets desired spacing and influences capacity and emissions.
VTG1 parameters c1, mu = 0.6 s / 0.1 s
Set the nonlinear time gap in Eq. (14); affects VTG1-CS performance.
VTG2 parameters d_VTG2, m = 7 m / 8.83 m/s
Set the exponential time gap in Eq. (18); affects VTG2-CS performance.
Common controller gains k_e, k_v, k = 0.1 / 0.98 / 0.7
Unified across CTG, VTG1, and VTG2 for comparability; taken from Qin (2019) but hand-chosen.
CS controller gains q1, q2, q3, q4 = 0.4 / 0.1 / 0.9 / 0.6
From Zheng et al. (2023); used in Eq. (9) for CS-based combinations.
BDBM/BS parameters T, a_max, b, lambda = 2.5 s, 1 m/s^2, 2 m/s^2, 0.5
From Yi et al. (2022); the large safe headway T=2.5 s contributes to BS-BS poor performance.

assumptions (7)

domain assumption Zero communication delay in V2V and sensors (Assumption 1)
The CAV control models ignore communication and sensor delay; prior work cited by the authors notes LPF platoon stability is sensitive to delay.
ad hoc to paper Human-driven vehicles obey a car-following model that is not stated in the paper
Section 3.3 defines HV modes but no HV acceleration model is given; the p=0 baseline and mixed-flow results depend on an unstated HV model.
domain assumption All CAVs are Level 5 and actuation is instantaneous (Assumptions 5 and 6)
No actuator lag; results may overstate stability of aggressive spacing strategies.
ad hoc to paper Platoon intensity O=1 in main simulations (all CAVs form platoons)
Section 5.3.1 fixes O=1; the probability model covers O in [0,1], but no sensitivity is tested.
domain assumption The probability distribution formulas of Jiang et al. (2023) are correct
Section 3.4 adopts the Markov-chain distribution model from the authors' earlier work without re-derivation.
domain assumption Fuel and emission models (VSP/NFR and Int Panis et al. 2006) apply to the simulated fleet
Equations (30)-(34) are empirical regressions from the literature applied without calibration to the simulated vehicles.
domain assumption Single-lane ring road, no lane changes (Assumption 2)
Simplifies dynamics; excludes multi-lane effects that could affect platoon performance.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Analysis of the impact of heterogeneous platoon for mixed traffic flow: control strategy, fuel consumption and emissions." pith.science (2026). https://pith.science/paper/K7KYNIGD

@misc{pith2026241115238,
  author       = {Pith},
  title        = {Pith review of: Analysis of the impact of heterogeneous platoon for mixed traffic flow: control strategy, fuel consumption and emissions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/K7KYNIGD}},
  note         = {Machine review of arXiv:2411.15238}
}

read the original abstract

Compared with traditional vehicle longitudinal spacing control strategies, the combination spacing strategy can integrate the advantages of different spacing control strategies. However, the impact mechanism of different combination spacing control strategies on mixed traffic flow has not been analyzed yet. Therefore, this paper proposes various combination spacing control strategies for connected automated vehicles (CAVs). First, a mixed traffic flow model was developed to analyze the characteristics of CAV platoons. On this basis, a probability model of vehicle distribution was derived, and its effectiveness was verified through simulation. Then, multiple spacing combination strategies are proposed based on four spacing control strategies. Finally, numerical experiments were conducted to calculate the average fuel consumption and pollutant emissions of mixed traffic flow under different spacing control strategies, and the impact of platoon spacing control strategies on traffic flow fuel consumption and pollutant emissions was further analyzed. Results show that: (1) the differences in average fuel consumption and pollutant emissions of traffic flow are relatively small under different platoon spacing control strategies under low traffic density (i.e., 15 veh/km); (2) at medium to high traffic densities (i.e., 55-95 veh/km), when the penetration rate of CAVs exceeds 80%, VTG1-CS, VTG2-CS, and CTG-CS strategies can effectively ensure traffic flow stability and safety, and significantly reduce fuel consumption and pollutant emissions.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

3 extracted references · 3 canonical work pages

[67]

https://doi.org/10.16097/j.cnki.1009-6744.2017.02.007 Yao, Z., Wu, Y., Wang, Y., Zhao, B., & Jiang, Y. (2023). Analysis of the impact of 40 maximum platoon size of CAVs on mixed traffic flow: An analytical and simulation method. Transportation Research Part C: Emerging Technologies , 147, 103989. https://doi.org/10.1016/j.trc.2022.103989 Yi, Z., Lu, W., Q...

work page arXiv 2023
[204]

https://doi.org/10.1109/TITS.2003.821341 Seiler, P., Pant, A., & Hedrick, K. (2004). Disturbance propagation in vehicle strings. lEEE Transactions on Automatic Control , 49(10), 1835 –1841. https://doi.org/10.1109/tac.2004.835586 Song, G., & Yu, L. (2009). Estimation of Fuel Efficiency of Road Traffic by Characterization of Vehicle-Specific Power and Spee...

work page arXiv 2004
[292]

-G., Wang, J

https://doi.org/10.1016/j.trb.2017.09.022 Guo, X. -G., Wang, J. -L., Liao, F., & Swee Huat Teo, R. (2016). String stability of heterogeneous leader -following vehicle platoons based on constant spacing policy. 2016 IEEE Intelligent Vehicles Symposium (IV) , 761 –766. https://doi.org/10.1109/IVS.2016.7535473 Hall, R., & Chin, C. (2005). Vehicle sorting for...

work page arXiv 2016

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [67]

https://doi.org/10.16097/j.cnki.1009-6744.2017.02.007 Yao, Z., Wu, Y., Wang, Y., Zhao, B., & Jiang, Y. (2023). Analysis of the impact of 40 maximum platoon size of CAVs on mixed traffic flow: An analytical and simulation method. Transportation Research Part C: Emerging Technologies , 147, 103989. https://doi.org/10.1016/j.trc.2022.103989 Yi, Z., Lu, W., Q...

work page arXiv 2023

[2] [204]

https://doi.org/10.1109/TITS.2003.821341 Seiler, P., Pant, A., & Hedrick, K. (2004). Disturbance propagation in vehicle strings. lEEE Transactions on Automatic Control , 49(10), 1835 –1841. https://doi.org/10.1109/tac.2004.835586 Song, G., & Yu, L. (2009). Estimation of Fuel Efficiency of Road Traffic by Characterization of Vehicle-Specific Power and Spee...

work page arXiv 2004

[3] [292]

-G., Wang, J

https://doi.org/10.1016/j.trb.2017.09.022 Guo, X. -G., Wang, J. -L., Liao, F., & Swee Huat Teo, R. (2016). String stability of heterogeneous leader -following vehicle platoons based on constant spacing policy. 2016 IEEE Intelligent Vehicles Symposium (IV) , 761 –766. https://doi.org/10.1109/IVS.2016.7535473 Hall, R., & Chin, C. (2005). Vehicle sorting for...

work page arXiv 2016