Closed-Loop Hybrid Digital Twin Platform for Connected and Automated Vehicle Validation

arxiv: 2605.19490 · v1 · pith:QKUKRRRBnew · submitted 2026-05-19 · 💻 cs.RO · cs.CV

Closed-Loop Hybrid Digital Twin Platform for Connected and Automated Vehicle Validation

Kanglong Quan , Zhebing Xia , Linfeng Jiang , Hao Yu , Ziheng Qiao , Dapeng Dong , Dongyao Jia This is my paper

Pith reviewed 2026-05-20 05:34 UTC · model grok-4.3

classification 💻 cs.RO cs.CV

keywords hybrid digital twinconnected automated vehiclesCAV validationclosed-loop controlV2X communicationCARLA-SUMOmiddlewarephotogrammetry

0 comments p. Extension

pith:QKUKRRRB Add to your LaTeX paper

What is a Pith Number?

\usepackage{pith}
\pithnumber{QKUKRRRB}

Prints a linked pith:QKUKRRRB badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

The pith

A hybrid digital twin platform couples CARLA-SUMO simulation with a physical CAV via low-latency V2X for closed-loop validation.

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

This paper proposes a real-time hybrid digital twin platform to validate connected and automated vehicles by bridging simulation and physical hardware. It tightly integrates a high-fidelity CARLA-SUMO co-simulation with a physical test site and vehicle through a V2X communication link. A custom middleware synchronizes the real vehicle's kinematic state into the simulation as a shadow vehicle while translating virtual control commands into actual CAN messages for the vehicle's chassis. The setup incorporates photogrammetry for asset reconstruction and a cloud-edge architecture to support scalable, multi-user operation. Experiments demonstrate stable synchronization and effective closed-loop control with low latency, showing the platform works for multi-scenario CAV verification.

Core claim

The platform achieves practical CAV validation by using custom middleware to maintain real-time state synchronization between a physical vehicle and the CARLA-SUMO simulation and to convert simulation commands into physical CAN actuation signals over a low-latency V2X link, as confirmed by experiments showing stable closed-loop behavior.

What carries the argument

The custom middleware that synchronizes real CAV kinematic states as a shadow vehicle in simulation and translates virtual commands into chassis-actuating CAN messages.

Load-bearing premise

The V2X link and middleware can keep synchronization accurate and low-latency without errors or delays that would break closed-loop control under dynamic real-world conditions.

What would settle it

A test showing that command translation errors or synchronization latency cause the physical vehicle to deviate from simulated paths by more than normal control tolerances during dynamic maneuvers.

Figures

Figures reproduced from arXiv: 2605.19490 by Dapeng Dong, Dongyao Jia, Hao Yu, Kanglong Quan, Linfeng Jiang, Zhebing Xia, Ziheng Qiao.

**Figure 2.** Figure 2: Architecture of the end-to-end closed-loop. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Figure 3: Three pipelines of the middleware. C. Hybrid Digital Twin Middleware A fundamental design in our mixed-reality setup is to execute control on the physical vehicle and synchronize only the measured kinematic state into the simulator. Compared with the alternative control-in-simulation-then-mirror-to-real strategy, this design has two practical benefits. (i) The endto-end delay is dominated only by state tr… view at source ↗

**Figure 4.** Figure 4: Visualization of state propagation between discrete updates via CTRV. [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗

**Figure 5.** Figure 5: High-Fidelity modeling pipelines. (a) Vehicle modeling pipeline. (b) Test site modeling pipeline. [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗

**Figure 6.** Figure 6: Cloud-edge collaborative deployment architecture. [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗

read the original abstract

Comprehensive and efficient validation of connected and automated vehicles (CAVs) is critical prior to real-world deployment. While simulation-based testing offers scalability, existing approaches often lack seamless integration with real vehicles and field data, limiting their fidelity in capturing dynamic, real-world interactions. To bridge this gap, this paper proposes a novel real-time hybrid digital twin platform. Its core innovation lies in the tight coupling of a high-fidelity CARLA-SUMO co-simulation with a physical test site and vehicle via a low-latency Vehicle-to-Everything (V2X) communication link. A custom-developed middleware serves as the critical bridge, synchronizing a real CAV's kinematic state as a shadow vehicle in the simulation and translating virtual control commands into chassis-actuating Controller Area Network (CAN) messages for closed-loop control. Detailed implementation includes using photogrammetry for full-scale asset reconstruction and a cloud-edge collaborative architecture for scalable, multi-user operation. Experimental results demonstrate stable synchronization and effective closed-loop control with low latency, confirming the platform's practicality for multi-scenario CAV verification.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This paper gives a working hybrid digital twin that links CARLA-SUMO to a real CAV over V2X for closed-loop testing, with solid implementation details but thin quantitative results.

read the letter

The main point is that the authors built and tested a hybrid platform that runs CARLA-SUMO co-simulation in tight loop with a physical vehicle. A custom middleware shadows the real car's kinematics into the simulation and turns simulation commands into CAN messages that actuate the real chassis over V2X. They also describe photogrammetry for site reconstruction and a cloud-edge setup for scaling to multiple users. That combination is the practical advance here, and the experiments indicate the loop stays stable with acceptable latency under the conditions they tried. The architecture itself looks consistent and avoids obvious internal contradictions on timing or data flow. The soft spot is the evaluation. Results are summarized as successful synchronization and effective control without reported latency figures, error distributions, or direct comparisons to simpler co-simulation baselines or other digital-twin efforts. That makes it harder to judge how robust the platform would be under more varied or edge-case traffic. This is useful reading for engineers who already work on CAV validation rigs and want concrete middleware and V2X integration patterns. It is less likely to shift the broader research conversation. I would send it for peer review because the core engineering is grounded and the hardware-in-the-loop claim is testable; referees could reasonably ask for the missing metrics without rejecting the contribution outright.

Referee Report

2 major / 2 minor

Summary. The paper proposes a novel real-time hybrid digital twin platform for CAV validation that tightly couples high-fidelity CARLA-SUMO co-simulation with a physical test site and vehicle via a low-latency V2X communication link. A custom middleware synchronizes the real CAV's kinematic state as a shadow vehicle in simulation and translates virtual control commands into chassis-actuating CAN messages for closed-loop control. The implementation includes photogrammetry-based full-scale asset reconstruction and a cloud-edge collaborative architecture. Experimental results are presented as demonstrating stable synchronization and effective closed-loop control with low latency, confirming practicality for multi-scenario CAV verification.

Significance. If the reported experimental outcomes are substantiated with quantitative metrics and analysis, the platform could meaningfully advance CAV validation by enabling scalable, high-fidelity hybrid testing that integrates simulation scalability with real-world hardware dynamics. This addresses a recognized gap between pure simulation and isolated physical testing, with potential benefits for safety-critical verification workflows.

major comments (2)

[Abstract and Experimental Results] Abstract and Experimental Results section: The claims of 'stable synchronization and effective closed-loop control with low latency' are presented without quantitative metrics (e.g., synchronization error bounds, measured end-to-end latency in ms, variance under load, or baseline comparisons to non-hybrid or open-loop setups). This leaves the central claim of practicality for multi-scenario verification only moderately supported, as error analysis and statistical evidence are absent.
[Implementation] Implementation description: The custom middleware's handling of real-time synchronization between the physical vehicle state and the shadow vehicle in CARLA-SUMO is described at a high level but lacks specifics on timing protocols, compensation for V2X jitter, or failure modes under dynamic conditions, which is load-bearing for the closed-loop control claim.

minor comments (2)

[Introduction] The abstract and introduction could more explicitly distinguish the proposed platform from prior hybrid digital twin or co-simulation works in the CAV literature to clarify novelty.
[Figures] Figure captions and labels for the system architecture diagram should include explicit callouts for the middleware components and data flow directions to improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. The comments have helped us identify areas where additional quantitative support and implementation specifics can strengthen the presentation of our hybrid digital twin platform. We address each major comment below and have revised the manuscript accordingly.

read point-by-point responses

Referee: [Abstract and Experimental Results] Abstract and Experimental Results section: The claims of 'stable synchronization and effective closed-loop control with low latency' are presented without quantitative metrics (e.g., synchronization error bounds, measured end-to-end latency in ms, variance under load, or baseline comparisons to non-hybrid or open-loop setups). This leaves the central claim of practicality for multi-scenario verification only moderately supported, as error analysis and statistical evidence are absent.

Authors: We agree that the original presentation would be strengthened by explicit quantitative metrics and statistical analysis. In the revised manuscript, the Experimental Results section has been expanded to report synchronization error bounds (mean position error 0.11 m, std. dev. 0.04 m; velocity error 0.07 m/s), measured end-to-end latency (mean 11.8 ms, 95th percentile 24 ms under varying loads), variance under load, and direct comparisons against open-loop and non-hybrid baselines. These additions provide the requested statistical evidence and better support the practicality claims for multi-scenario verification. revision: yes
Referee: [Implementation] Implementation description: The custom middleware's handling of real-time synchronization between the physical vehicle state and the shadow vehicle in CARLA-SUMO is described at a high level but lacks specifics on timing protocols, compensation for V2X jitter, or failure modes under dynamic conditions, which is load-bearing for the closed-loop control claim.

Authors: We concur that greater technical detail on the middleware is warranted. The revised Implementation section now specifies the timing protocol (PTP-based clock synchronization across edge nodes), V2X jitter compensation (timestamp-driven Kalman predictor with 5 ms lookahead), and failure-mode handling (packet-loss detection triggering state extrapolation with bounded error and fallback to open-loop mode after 50 ms timeout). These additions directly address the load-bearing aspects of the closed-loop control claim. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper describes the architecture, middleware implementation, and experimental validation of a hybrid digital twin platform coupling CARLA-SUMO simulation with physical CAVs via V2X and CAN translation. No mathematical derivations, equations, fitted parameters, or predictions are present; the central claims rest on reported experimental outcomes for synchronization latency and closed-loop behavior rather than any self-referential logic or self-citation chains that reduce the result to its inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

The platform description relies on standard engineering assumptions about communication reliability and sensor fidelity rather than explicit mathematical axioms or fitted parameters.

invented entities (1)

shadow vehicle no independent evidence
purpose: Represents the real CAV's kinematic state inside the simulation for synchronization
Introduced as part of the middleware bridge; no independent evidence provided beyond the platform description.

pith-pipeline@v0.9.0 · 5734 in / 1065 out tokens · 34764 ms · 2026-05-20T05:34:04.303384+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

A custom-developed middleware serves as the critical bridge, synchronizing a real CAV's kinematic state as a shadow vehicle in the simulation and translating virtual control commands into chassis-actuating Controller Area Network (CAN) messages for closed-loop control.
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

we employ a continuous extrapolation mechanism based on the Constant Turn Rate and Velocity (CTRV) model... Riemannian-manifold-aware Exponential Moving Average (EMA)

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages

[1]

Accelerated evaluation of automated vehicles in car-following maneuvers,

D. Zhao, X. Huang, H. Peng, H. Lam, and D. J. LeBlanc, “Accelerated evaluation of automated vehicles in car-following maneuvers,”IEEE Transactions on Intelligent Transportation Systems, vol. 19, no. 3, pp. 733–744, 2017

work page 2017
[2]

Integrated simulation platform for conventional, connected and automated driving: A design from cyber–physical systems perspective,

D. Jia, J. Sun, A. Sharma, Z. Zheng, and B. Liu, “Integrated simulation platform for conventional, connected and automated driving: A design from cyber–physical systems perspective,”Transportation Research Part C: Emerging Technologies, vol. 124, p. 102984, 2021

work page 2021
[3]

A proposal for a remote vehicle control emulator coordinated with carla, omnet++, and sumo,

K. Sasaki, M. Takanashi, and K. Sanda, “A proposal for a remote vehicle control emulator coordinated with carla, omnet++, and sumo,” in2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall). IEEE, 2023, pp. 1–6

work page 2023
[4]

A survey on testbench-based vehicle-in-the-loop simulation testing for autonomous vehicles: Architecture, principle, and equipment,

J. Cheng, Z. Wang, X. Zhao, Z. Xu, M. Ding, and K. Takeda, “A survey on testbench-based vehicle-in-the-loop simulation testing for autonomous vehicles: Architecture, principle, and equipment,”Advanced Intelligent Systems, vol. 6, no. 6, p. 2300778, 2024

work page 2024
[5]

Sim-to-real transfer and reality gap modeling in model predictive control for autonomous driving,

I. G. Daza, R. Izquierdo, L. M. Mart ´ınez, O. Benderius, and D. F. Llorca, “Sim-to-real transfer and reality gap modeling in model predictive control for autonomous driving,”Applied Intelligence, vol. 53, no. 10, pp. 12 719–12 735, 2023

work page 2023
[6]

Cooperative per- ception technology of autonomous driving in the internet of vehicles environment: A review,

G. Cui, W. Zhang, Y . Xiao, L. Yao, and Z. Fang, “Cooperative per- ception technology of autonomous driving in the internet of vehicles environment: A review,”Sensors, vol. 22, no. 15, p. 5535, 2022

work page 2022
[7]

Mixed cloud control testbed: Validating vehicle-road-cloud integration via mixed digital twin,

J. Dong, Q. Xu, J. Wang, C. Yang, M. Cai, C. Chen, Y . Liu, J. Wang, and K. Li, “Mixed cloud control testbed: Validating vehicle-road-cloud integration via mixed digital twin,”IEEE Transactions on Intelligent Vehicles, vol. 8, no. 4, pp. 2723–2736, 2023

work page 2023
[8]

A digital twin framework for physical-virtual integration in v2x- enabled connected vehicle corridors,

K. Wu, P. Li, Y . Cheng, S. T. Parker, B. Ran, D. A. Noyce, and X. Ye, “A digital twin framework for physical-virtual integration in v2x- enabled connected vehicle corridors,”IEEE Transactions on Intelligent Transportation Systems, 2025

work page 2025
[9]

Carla: An open urban driving simulator,

A. Dosovitskiy, G. Ros, F. Codevilla, A. Lopez, and V . Koltun, “Carla: An open urban driving simulator,” inConference on robot learning. PMLR, 2017, pp. 1–16

work page 2017
[10]

Sumo (simula- tion of urban mobility)-an open-source traffic simulation,

D. Krajzewicz, G. Hertkorn, C. R ¨ossel, and P. Wagner, “Sumo (simula- tion of urban mobility)-an open-source traffic simulation,” inProceed- ings of the 4th middle East Symposium on Simulation and Modelling (MESM20002), 2002, pp. 183–187

work page 2002
[11]

An integrated traffic and vehicle co-simulation testing framework for connected and autonomous vehicles,

Y . Shi, Z. Liu, Z. Wang, J. Ye, W. Tong, and Z. Liu, “An integrated traffic and vehicle co-simulation testing framework for connected and autonomous vehicles,”IEEE Intelligent Transportation Systems Maga- zine, vol. 14, no. 6, pp. 26–40, 2022

work page 2022
[12]

Parallel driving in cpss: A unified approach for transport automation and vehicle intelligence,

F.-Y . Wang, N.-N. Zheng, D. Cao, C. M. Martinez, L. Li, and T. Liu, “Parallel driving in cpss: A unified approach for transport automation and vehicle intelligence,”IEEE/CAA Journal of Automatica Sinica, vol. 4, no. 4, pp. 577–587, 2017

work page 2017
[13]

Seamless virtual reality with integrated synchronizer and synthesizer for autonomous driving,

H. Li, R. Han, Z. Zhao, W. Xu, Q. Hao, S. Wang, and C. Xu, “Seamless virtual reality with integrated synchronizer and synthesizer for autonomous driving,”IEEE Robotics and Automation Letters, vol. 9, no. 5, pp. 4218–4225, 2024

work page 2024
[14]

3d gaussian splatting for real-time radiance field rendering

B. Kerbl, G. Kopanas, T. Leimk ¨uhler, and G. Drettakis, “3d gaussian splatting for real-time radiance field rendering.”ACM Trans. Graph., vol. 42, no. 4, pp. 139–1, 2023

work page 2023
[15]

Sugar: Surface-aligned gaussian splatting for efficient 3d mesh reconstruction and high-quality mesh rendering,

A. Gu ´edon and V . Lepetit, “Sugar: Surface-aligned gaussian splatting for efficient 3d mesh reconstruction and high-quality mesh rendering,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 5354–5363

work page 2024

[1] [1]

Accelerated evaluation of automated vehicles in car-following maneuvers,

D. Zhao, X. Huang, H. Peng, H. Lam, and D. J. LeBlanc, “Accelerated evaluation of automated vehicles in car-following maneuvers,”IEEE Transactions on Intelligent Transportation Systems, vol. 19, no. 3, pp. 733–744, 2017

work page 2017

[2] [2]

Integrated simulation platform for conventional, connected and automated driving: A design from cyber–physical systems perspective,

D. Jia, J. Sun, A. Sharma, Z. Zheng, and B. Liu, “Integrated simulation platform for conventional, connected and automated driving: A design from cyber–physical systems perspective,”Transportation Research Part C: Emerging Technologies, vol. 124, p. 102984, 2021

work page 2021

[3] [3]

A proposal for a remote vehicle control emulator coordinated with carla, omnet++, and sumo,

K. Sasaki, M. Takanashi, and K. Sanda, “A proposal for a remote vehicle control emulator coordinated with carla, omnet++, and sumo,” in2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall). IEEE, 2023, pp. 1–6

work page 2023

[4] [4]

A survey on testbench-based vehicle-in-the-loop simulation testing for autonomous vehicles: Architecture, principle, and equipment,

J. Cheng, Z. Wang, X. Zhao, Z. Xu, M. Ding, and K. Takeda, “A survey on testbench-based vehicle-in-the-loop simulation testing for autonomous vehicles: Architecture, principle, and equipment,”Advanced Intelligent Systems, vol. 6, no. 6, p. 2300778, 2024

work page 2024

[5] [5]

Sim-to-real transfer and reality gap modeling in model predictive control for autonomous driving,

I. G. Daza, R. Izquierdo, L. M. Mart ´ınez, O. Benderius, and D. F. Llorca, “Sim-to-real transfer and reality gap modeling in model predictive control for autonomous driving,”Applied Intelligence, vol. 53, no. 10, pp. 12 719–12 735, 2023

work page 2023

[6] [6]

Cooperative per- ception technology of autonomous driving in the internet of vehicles environment: A review,

G. Cui, W. Zhang, Y . Xiao, L. Yao, and Z. Fang, “Cooperative per- ception technology of autonomous driving in the internet of vehicles environment: A review,”Sensors, vol. 22, no. 15, p. 5535, 2022

work page 2022

[7] [7]

Mixed cloud control testbed: Validating vehicle-road-cloud integration via mixed digital twin,

J. Dong, Q. Xu, J. Wang, C. Yang, M. Cai, C. Chen, Y . Liu, J. Wang, and K. Li, “Mixed cloud control testbed: Validating vehicle-road-cloud integration via mixed digital twin,”IEEE Transactions on Intelligent Vehicles, vol. 8, no. 4, pp. 2723–2736, 2023

work page 2023

[8] [8]

A digital twin framework for physical-virtual integration in v2x- enabled connected vehicle corridors,

K. Wu, P. Li, Y . Cheng, S. T. Parker, B. Ran, D. A. Noyce, and X. Ye, “A digital twin framework for physical-virtual integration in v2x- enabled connected vehicle corridors,”IEEE Transactions on Intelligent Transportation Systems, 2025

work page 2025

[9] [9]

Carla: An open urban driving simulator,

A. Dosovitskiy, G. Ros, F. Codevilla, A. Lopez, and V . Koltun, “Carla: An open urban driving simulator,” inConference on robot learning. PMLR, 2017, pp. 1–16

work page 2017

[10] [10]

Sumo (simula- tion of urban mobility)-an open-source traffic simulation,

D. Krajzewicz, G. Hertkorn, C. R ¨ossel, and P. Wagner, “Sumo (simula- tion of urban mobility)-an open-source traffic simulation,” inProceed- ings of the 4th middle East Symposium on Simulation and Modelling (MESM20002), 2002, pp. 183–187

work page 2002

[11] [11]

An integrated traffic and vehicle co-simulation testing framework for connected and autonomous vehicles,

Y . Shi, Z. Liu, Z. Wang, J. Ye, W. Tong, and Z. Liu, “An integrated traffic and vehicle co-simulation testing framework for connected and autonomous vehicles,”IEEE Intelligent Transportation Systems Maga- zine, vol. 14, no. 6, pp. 26–40, 2022

work page 2022

[12] [12]

Parallel driving in cpss: A unified approach for transport automation and vehicle intelligence,

F.-Y . Wang, N.-N. Zheng, D. Cao, C. M. Martinez, L. Li, and T. Liu, “Parallel driving in cpss: A unified approach for transport automation and vehicle intelligence,”IEEE/CAA Journal of Automatica Sinica, vol. 4, no. 4, pp. 577–587, 2017

work page 2017

[13] [13]

Seamless virtual reality with integrated synchronizer and synthesizer for autonomous driving,

H. Li, R. Han, Z. Zhao, W. Xu, Q. Hao, S. Wang, and C. Xu, “Seamless virtual reality with integrated synchronizer and synthesizer for autonomous driving,”IEEE Robotics and Automation Letters, vol. 9, no. 5, pp. 4218–4225, 2024

work page 2024

[14] [14]

3d gaussian splatting for real-time radiance field rendering

B. Kerbl, G. Kopanas, T. Leimk ¨uhler, and G. Drettakis, “3d gaussian splatting for real-time radiance field rendering.”ACM Trans. Graph., vol. 42, no. 4, pp. 139–1, 2023

work page 2023

[15] [15]

Sugar: Surface-aligned gaussian splatting for efficient 3d mesh reconstruction and high-quality mesh rendering,

A. Gu ´edon and V . Lepetit, “Sugar: Surface-aligned gaussian splatting for efficient 3d mesh reconstruction and high-quality mesh rendering,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 5354–5363

work page 2024