pith. sign in

arxiv: 2605.21145 · v1 · pith:4IQJSQ6Dnew · submitted 2026-05-20 · 💻 cs.DC · cs.AR

Cloud-Native Operation of Roadside Infrastructure Enabling Demand-Driven Collective Perception via V2X

Lukas Zanger , Fabian Thomsen , Guido Linden , Jean-Pierre Busch , Lennart Reiher , Lutz Eckstein This is my paper

Pith reviewed 2026-05-21 01:38 UTC · model grok-4.3

classification 💻 cs.DC cs.AR
keywords cloud-native architectureroadside infrastructuredemand-driven orchestrationcollective perceptionV2X communicationenergy efficiencyC-ITSKubernetes
0
0 comments X

The pith

Demand-driven orchestration deploys collective perception services in time for nearby vehicles while keeping them inactive otherwise to reduce energy consumption and congestion.

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

This paper shows how a cloud-native Kubernetes cluster spanning roadside units and a cloud server can orchestrate the dynamic deployment of services for roadside infrastructure. The key idea is to activate resource-heavy applications like V2X collective perception only when a connected vehicle approaches, based on detection. Experiments in a real test field in Aachen confirm that the service starts quickly enough for the vehicle to use the enhanced perception data. When no vehicles are present, the service stays off, which cuts energy use, eases channel load, and reduces hardware wear over time. Readers interested in scalable smart transport systems would care because this offers a way to expand roadside capabilities without constant high costs or interference.

Core claim

The paper establishes that demand-driven orchestration on the described cloud-native architecture allows the V2X-based collective perception application to activate in time upon vehicle detection in real-world conditions, enabling benefits for the vehicle, while the application remains inactive in the absence of demand, thereby reducing energy consumption, channel congestion, and hardware wear, with supporting estimates from V2X recordings in the Aachen test field.

What carries the argument

The demand-driven orchestration approach that dynamically deploys the collective perception service on the Kubernetes cluster when a vehicle is detected nearby via V2X.

If this is right

  • The service activates sufficiently early for the vehicle to receive and benefit from the collective perception data.
  • Without demand the application stays inactive, directly lowering energy consumption at the roadside units.
  • Channel congestion is reduced because unnecessary V2X messages are not transmitted when no vehicles require the service.
  • Hardware wear decreases due to less continuous operation of the infrastructure components.
  • Overall scalability for large C-ITS deployments improves by avoiding always-on resource usage.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This method could be applied to other roadside services such as dynamic traffic light control or hazard warnings to achieve similar efficiency gains.
  • Long-term deployment in larger areas might reveal cumulative energy savings that affect overall power grid demands for infrastructure.
  • Combining this with predictive vehicle routing data could allow even earlier activation to minimize any activation delays.

Load-bearing premise

The orchestration system can detect vehicles and deploy the service with low enough latency and high reliability to ensure activation before the vehicle enters the relevant area, under the test field conditions.

What would settle it

A test run where the time between vehicle detection and full service activation exceeds the time it takes for the vehicle to reach the roadside unit's coverage zone, or repeated failures to activate on demand.

Figures

Figures reproduced from arXiv: 2605.21145 by Fabian Thomsen, Guido Linden, Jean-Pierre Busch, Lennart Reiher, Lukas Zanger, Lutz Eckstein.

Figure 1
Figure 1. Figure 1: When a connected vehicle enters a geofenced area, services enabling [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Kubernetes cluster comprising four sRISU nodes and one server node, collectively hosting the collective perception application. On each sRISU, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: End-to-end latency is defined as the time between [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: CAM occurrence matrix for the V2X recording: The blocks [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 4
Figure 4. Figure 4: Trajectories are aggregated based on the recorded CAM messages. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
read the original abstract

Intelligent roadside infrastructure is a key enabler for cooperative intelligent transport systems (C-ITS), supporting vehicles equipped with automated driving systems (ADS), e.g., through enhanced environment perception. With a growing number and an expanding functional scope of roadside units, scalable and efficient operation becomes a challenge. This paper presents a cloud-native architecture for the operation of distributed roadside infrastructure based on a Kubernetes cluster spanning roadside units and a cloud server. Building on this architecture, a demand-driven orchestration approach is implemented to dynamically deploy resource-intensive services only when required. As a representative use case, a V2X-based collective perception application is deployed on-demand when a connected vehicle is nearby. The approach is validated in a real-world experiment in our test field in Aachen, demonstrating that the collective perception application starts in time for the vehicle to benefit from it. Without any demand, the application remains inactive, reducing energy consumption, channel congestion, and hardware wear. Beyond the primary evaluation, V2X recordings from the test field are analyzed to estimate the energy-saving potential of demand-driven operation. In summary, the results demonstrate the practical feasibility of cloud-native, demand-driven operation of roadside infrastructure and indicate its potential to improve scalability and (energy) efficiency in future C-ITS deployments.

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.

Referee Report

1 major / 1 minor

Summary. The paper proposes a cloud-native Kubernetes-based architecture spanning roadside units and a cloud server for operating distributed roadside infrastructure in C-ITS. It implements demand-driven orchestration to deploy a resource-intensive V2X collective perception service only upon detection of a nearby connected vehicle. The approach is validated via a real-world experiment in the Aachen test field, claiming that the service activates in time for the vehicle to benefit while remaining inactive otherwise to reduce energy use, channel congestion, and hardware wear; V2X recordings are additionally analyzed to estimate energy-saving potential.

Significance. If the orchestration latency claims are quantitatively substantiated, the work offers a practical path toward scalable and energy-efficient C-ITS by avoiding always-on operation of roadside services. The real-world deployment and experimental validation constitute a clear strength, providing direct evidence of feasibility beyond simulation.

major comments (1)
  1. In the description of the Aachen test-field validation and results: the claim that the collective perception application 'starts in time for the vehicle to benefit from it' is load-bearing for the central contribution, yet no quantitative measurements are supplied for (a) the interval from first V2X detection to service-ready state, (b) vehicle speeds or distances at the moment of detection, or (c) the minimum lead time needed for the perception benefit to be realized. Without these values the demonstration is compatible with both 'sufficiently fast' and 'marginally too slow' interpretations, leaving the low-latency assumption of the Kubernetes cluster untested in the reported data.
minor comments (1)
  1. Abstract: the statement that V2X recordings 'are analyzed to estimate the energy-saving potential' would benefit from a brief indication of the analysis method or the magnitude of the estimated savings to give readers an immediate sense of the quantitative support.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the detailed and constructive review. The major comment identifies a key area where our validation can be strengthened with additional quantitative details from the Aachen test field experiments. We address this below and will revise the manuscript to incorporate the requested measurements and analysis.

read point-by-point responses

  1. Referee: In the description of the Aachen test-field validation and results: the claim that the collective perception application 'starts in time for the vehicle to benefit from it' is load-bearing for the central contribution, yet no quantitative measurements are supplied for (a) the interval from first V2X detection to service-ready state, (b) vehicle speeds or distances at the moment of detection, or (c) the minimum lead time needed for the perception benefit to be realized. Without these values the demonstration is compatible with both 'sufficiently fast' and 'marginally too slow' interpretations, leaving the low-latency assumption of the Kubernetes cluster untested in the reported data.

    Authors: We agree that explicit quantitative values would make the timeliness claim more robust and allow readers to independently assess the orchestration performance. In the revised manuscript, we will add a new subsection (or expanded results paragraph) that reports: (a) measured end-to-end times from first V2X Cooperative Awareness Message reception to the collective-perception pod reaching Ready state, obtained from Kubernetes event logs and our monitoring stack; (b) observed vehicle speeds and distances at detection, drawn from the test-field V2X recordings and GPS traces collected during the experiments; and (c) an estimate of the minimum lead time required by the perception service, based on the time needed to generate and disseminate extended environmental models to the approaching vehicle. These additions will be supported by a table of representative runs and a short discussion of how the observed latencies compare with the derived lead-time requirement, thereby directly testing the low-latency behavior of the Kubernetes cluster under realistic roadside conditions. revision: yes

Circularity Check

0 steps flagged

No circularity: claims rest on experimental deployment and recordings

full rationale

The paper describes a Kubernetes-based cloud-native architecture for demand-driven orchestration of V2X collective perception services on roadside units. Central claims are supported by a real-world Aachen test-field deployment showing timely service activation and by post-hoc analysis of V2X recordings to estimate energy savings. No equations, fitted parameters renamed as predictions, self-citation load-bearing uniqueness theorems, or ansatz smuggling appear in the derivation chain. The results are therefore independent of any reduction to their own inputs and remain self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper relies on standard assumptions from cloud computing and networking domains rather than introducing new fitted parameters or invented entities; the central claim rests on experimental feasibility rather than mathematical derivation.

axioms (1)
  • domain assumption A Kubernetes cluster can reliably span and orchestrate services across physically distributed roadside units and a central cloud server with acceptable latency.
    Invoked in the architecture description to enable dynamic deployment.

pith-pipeline@v0.9.0 · 5770 in / 1248 out tokens · 45918 ms · 2026-05-21T01:38:32.929189+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

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

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

34 extracted references · 34 canonical work pages

  1. [1]

    Intelligent transportation systems using roadside infrastructure: A literature survey,

    C. Creß, Z. Bing, and A. C. Knoll, “Intelligent transportation systems using roadside infrastructure: A literature survey,”IEEE Transactions on Intelligent Transportation Systems, vol. 25, no. 7, pp. 6309–6327, 2024

    work page 2024

  2. [2]

    [Online]

    Cloud Native Computing Foundation. [Online]. Available: https: //www.cncf.io/about/who-we-are/

    work page

  3. [3]

    Burns, J

    B. Burns, J. Beda, and K. Hightower,Kubernetes up & Running: Dive into the Future of Infrastructure. Sebastopol, United States: O’Reilly Media, 2019

    work page 2019

  4. [4]

    [Online]

    Cloud Native Glossary: Microservices Architecture. [Online]. Available: https://glossary.cncf.io/microservices-architecture/

    work page

  5. [5]

    [Online]

    Cloud Native Glossary: Containerization. [Online]. Available: https: //glossary.cncf.io/containerization/

    work page

  6. [6]

    [Online]

    Cloud Native Glossary: Container Orchestration. [Online]. Available: https://glossary.cncf.io/container-orchestration/

    work page

  7. [7]

    [Online]

    Cloud Native Glossary: Cluster. [Online]. Available: https://glossary .cncf.io/cluster/

    work page

  8. [8]

    [Online]

    K3s - Lightweight Kubernetes. [Online]. Available: https://docs.k3s.io/

    work page

  9. [9]

    European Telecommunications Standards Institute (ETSI),Intelligent Transport Systems (ITS); ITS-G5 Access layer in the 5 GHz frequency band, European Standard ETSI EN 303 797 V2.1.1, Nov. 2023

    work page 2023

  10. [10]

    European Telecommunications Standards Institute (ETSI),Intelligent Transport Systems (ITS); Vehicular Communications; Basic Set of Applications; Part 2: Specification of Cooperative Awareness Basic Service, European Standard ETSI EN 302 637-2 V1.4.1, Apr. 2019

    work page 2019

  11. [11]

    European Telecommunications Standards Institute (ETSI),Intelligent Transport Systems (ITS); Vehicular Communications; Basic Set of Applications; Collective Perception Service; Release 2, Technical Specification ETSI TS 103 324 V2.1.1, Jun. 2023

    work page 2023

  12. [12]

    V2AIX: A Multi-Modal Real-World Dataset of ETSI ITS V2X Messages in Public Road Traffic,

    G. Kuepperset al., “V2AIX: A Multi-Modal Real-World Dataset of ETSI ITS V2X Messages in Public Road Traffic,” in2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC). Edmonton, AB, Canada: IEEE, Sep. 2024, pp. 392–398

    work page 2024

  13. [13]

    Robot Operating System 2: Design, architecture, and uses in the wild,

    S. Macenskiet al., “Robot Operating System 2: Design, architecture, and uses in the wild,”Science Robotics, vol. 7, no. 66, p. eabm6074,

    work page

  14. [14]

    Available: https://www.science.org/doi/abs/10.1126/sc irobotics.abm6074

    [Online]. Available: https://www.science.org/doi/abs/10.1126/sc irobotics.abm6074

    work page doi:10.1126/sc

  15. [15]

    Cloud-Native Infrastructure: Powering the Next Generation of Autonomous Vehicles,

    Anuj Harishkumar Chaudhari, “Cloud-Native Infrastructure: Powering the Next Generation of Autonomous Vehicles,”Journal of Computer Science and Technology Studies, vol. 7, no. 3, pp. 187–193, May 2025

    work page 2025

  16. [16]

    Leveraging public cloud infrastructure for real- time connected vehicle speed advisory at a signalized corridor,

    H.-W. Denget al., “Leveraging public cloud infrastructure for real- time connected vehicle speed advisory at a signalized corridor,”Inter- national Journal of Transportation Science and Technology, vol. 17, Mar. 2025

    work page 2025

  17. [17]

    Dependability-Aware Coordination for Infrastructure- Side Critical V2X Services,

    Y . Duet al., “Dependability-Aware Coordination for Infrastructure- Side Critical V2X Services,” in2025 IEEE 102nd Vehicular Technol- ogy Conference (VTC2025-Fall). Chengdu, China: IEEE, Oct. 2025

    work page 2025

  18. [18]

    Application Management in C-ITS: Orchestrating Demand-Driven Deployments and Reconfigurations,

    L. Zangeret al., “Application Management in C-ITS: Orchestrating Demand-Driven Deployments and Reconfigurations,” in2025 IEEE In- ternational Conference on Intelligent Transportation Systems (ITSC). Gold Coast, Australia: IEEE, 2025

    work page 2025

  19. [19]

    European Telecommunications Standards Institute (ETSI),Intelligent Transport Systems (ITS); Communications Architecture, Technical Specification ETSI EN 302 665 V1.1.1, Sep. 2010

    work page 2010

  20. [20]

    European Telecommunications Standards Institute (ETSI),Intelligent Transport Systems (ITS); Vehicular Communications; GeoNetworking; Part 3: Network Architecture, European Standard ETSI EN 302 636-3 V1.2.1, Dec. 2014

    work page 2014

  21. [21]

    [Online]

    Helm. [Online]. Available: https://helm.sh/

    work page

  22. [22]

    [Online]

    Grafana OSS: Visualize your data, optimize your performance. [Online]. Available: https://grafana.com/oss/grafana/

    work page

  23. [23]

    [Online]

    Kubernetes Dashboard. [Online]. Available: https://kubernetes.io/docs /tasks/access-application-cluster/web-ui-dashboard/

    work page

  24. [24]

    [Online]

    Prometheus: Open source metrics and monitoring for your systems and services. [Online]. Available: https://prometheus.io/

    work page

  25. [25]

    [Online]

    Grafana loki. [Online]. Available: https://grafana.com/oss/loki/

    work page

  26. [26]

    [Online]

    NVIDIA Data Center GPU Manager (DCGM). [Online]. Available: https://developer.nvidia.com/dcgm

    work page

  27. [27]

    [Online]

    Application Management Framework on GitHub. [Online]. Available: https://github.com/ika-rwth-aachen/application manager

    work page

  28. [28]

    Event Detection in C-ITS: Classification, Use Cases, and Reference Implementation

    L. Reiheret al., “Event Detection in C-ITS: Classification, Use Cases, and Reference Implementation.” 16. Uni-DAS e.V . Workshop Fahrerassistenz und automatisiertes Fahren, 2025

    work page 2025

  29. [29]

    Zenoh: Unifying communication, storage and computation from the cloud to the microcontroller,

    A. Corsaroet al., “Zenoh: Unifying communication, storage and computation from the cloud to the microcontroller,” in2023 26th Euromicro Conference on Digital System Design (DSD), 2023

    work page 2023

  30. [30]

    Scalable real-time multi object tracking for automated driving using intelligent infrastructure,

    F. Thomsen, M. Ortmann, and L. Eckstein, “Scalable real-time multi object tracking for automated driving using intelligent infrastructure,” in2023 International Conference on Electrical, Computer and Energy Technologies (ICECET), 2023, pp. 1–6

    work page 2023

  31. [31]

    karl. - A Research Vehicle for Automated and Connected Driving,

    J.-P. Buschet al., “karl. - A Research Vehicle for Automated and Connected Driving,” in2026 IEEE Intelligent Vehicles Symposium (IV). Detroit, United States: IEEE, 2026

    work page 2026

  32. [32]

    Experimental V2X Evaluation for C-V2X and ITS-G5 Technologies in a Real-Life Highway Environment,

    V . Maglogianniset al., “Experimental V2X Evaluation for C-V2X and ITS-G5 Technologies in a Real-Life Highway Environment,”IEEE Transactions on Network and Service Management, vol. 19, Jun. 2022

    work page 2022

  33. [33]

    White Paper on ITS-G5 and Sidelink LTE-V2X Co-Channel Coexistence Mitigation Methods,

    CAR 2 CAR Communication Consortium, “White Paper on ITS-G5 and Sidelink LTE-V2X Co-Channel Coexistence Mitigation Methods,” CAR 2 CAR Communication Consortium, White Paper, 2021

    work page 2021

  34. [34]

    [Online]

    Electricity consumption of private households. [Online]. Available: https://www.destatis.de/DE/Themen/Gesellschaft-Umwelt/Umwelt/U GR/private-haushalte/Tabellen/stromverbrauch-haushalte.html

    work page