pith. sign in

arxiv: 2605.19655 · v1 · pith:LIJ63RBHnew · submitted 2026-05-19 · 📡 eess.SY · cs.SY

Equalized Coverage in Motion Control Performance Prediction for Self-Adaptive Road Vehicles

Ole Reuter , Richard Schubert , Marvin Loba , Markus Maurer This is my paper

Pith reviewed 2026-05-20 04:42 UTC · model grok-4.3

classification 📡 eess.SY cs.SY
keywords conformal predictionquantile regressionautomated drivingmotion controlactuator failurecapability monitoringself-adaptive vehicles
0
0 comments X

The pith

A conformalized quantile regression model with equalized coverage predicts whether automated vehicles can maintain low lateral deviation despite actuator degradations or failures.

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

The paper proposes a lightweight prediction model to monitor if an automated vehicle can follow a planned trajectory with low lateral deviation under different actuator conditions, including nominal, degraded, and failed states. This is important for capability monitoring in self-adaptive systems to ensure safe operation when components degrade. The authors use conformalized quantile regression and equalized coverage methods to provide statistical guarantees that hold not just overall but also for specific data regimes like different failure modes. During runtime, this predictor helps determine the admissible action space for behavior generation.

Core claim

We propose a lightweight prediction model based on conformalized quantile regression that predicts whether an automated vehicle can be controlled with sufficiently low lateral deviation from a planned trajectory under nominal, degraded, and failed actuator conditions, employing equalized coverage methods to ensure conditional statistical guarantees.

What carries the argument

Conformalized quantile regression with equalized coverage, which provides prediction intervals with statistical guarantees that adapt to different regimes in the data.

If this is right

  • The model can serve as a heuristic for determining the admissible action space during runtime behavior generation.
  • It supports capability-based behavior in automated driving systems by evaluating remaining performance under degradation.
  • The approach addresses monitoring needs for system elements that degrade and fail.
  • Statistical guarantees hold conditionally for different actuator conditions rather than only marginally.

Where Pith is reading between the lines

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

  • If the model is accurate, it could be integrated into real-time decision making to avoid unsafe maneuvers under failure.
  • Extending this to other vehicle dynamics or sensor degradations might broaden the applicability to broader self-adaptive systems.
  • Testing on real-world data from diverse driving scenarios would validate the conditional coverage in practice.

Load-bearing premise

Statistical coverage guarantees must hold within specific data regimes like different actuator conditions, not only on average across all cases.

What would settle it

If empirical tests show that the actual coverage rate drops below the target level for data points corresponding to failed actuators, that would falsify the method's effectiveness.

Figures

Figures reproduced from arXiv: 2605.19655 by Markus Maurer, Marvin Loba, Ole Reuter, Richard Schubert.

Figure 1
Figure 1. Figure 1: Simplified illustration of a functional architecture taken from [2], based on [3, 4], originally adapted from [5]. System boundaries of the automated [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 1
Figure 1. Figure 1: To mathematically represent deviations from the [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Interval length distributions for predictions on [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Visualization of the application scenario for degradation D2 and [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
read the original abstract

Automated driving systems require monitoring mechanisms to ensure operation as intended, especially when system elements degrade and/or fail. Hence, capability monitoring is crucial in order to evaluate the system's remaining performance and implement capability-based behavior. In this paper, we investigate the dynamics of a highly over-actuated automated vehicle under actuator degradations and failures, affecting the vehicle's motion control capabilities. We propose a lightweight prediction model based on conformalized quantile regression that predicts whether an automated vehicle can be controlled with sufficiently low lateral deviation from a planned trajectory under nominal, degraded, and failed actuator conditions. We recognize that statistical guarantees should hold not only across all data (marginal coverage) but also for different regimes within the data (conditional coverage). We therefore employ equalized coverage methods to address this challenge. During runtime behavior generation our predictor can provide a heuristic for determining the admissible action space. Its application and limitations are discussed in this paper.

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 / 2 minor

Summary. The manuscript proposes a lightweight prediction model based on conformalized quantile regression with equalized coverage to predict whether an automated vehicle can maintain sufficiently low lateral deviation from a planned trajectory under nominal, degraded, and failed actuator conditions. The approach is motivated by the need for capability monitoring in self-adaptive road vehicles and aims to deliver statistical guarantees that hold conditionally across actuator regimes rather than only marginally.

Significance. If the conditional coverage property is empirically verified, the work would provide a practical, real-time tool for performance prediction in safety-critical autonomous driving, enabling better-informed behavior generation under actuator degradation. The focus on equalized coverage directly targets a known limitation of standard conformal prediction and is well-motivated for regime-specific reliability in over-actuated vehicle control.

major comments (1)
  1. [§4] §4 (Numerical Experiments) and associated tables/figures: the central claim that equalized coverage yields reliable conditional guarantees across the three actuator regimes (nominal, degraded, failed) is load-bearing for the paper's contribution to capability monitoring. However, coverage metrics appear to be reported only in aggregate form without separate empirical coverage rates, calibration plots, or per-regime tables for each actuator condition. This leaves the conditional coverage advantage over ordinary conformal prediction unverified and weakens support for the runtime heuristic application.
minor comments (2)
  1. [§3] The description of how equalized coverage is adapted to the specific actuator regimes could be expanded with a short algorithmic outline or pseudocode for clarity.
  2. [Figures] Figure captions should explicitly state the number of trials or data points used for each coverage evaluation to aid reproducibility.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback on our manuscript. We have addressed the major comment regarding the presentation of conditional coverage results and revised the manuscript accordingly to strengthen the empirical support for our claims.

read point-by-point responses

  1. Referee: [§4] §4 (Numerical Experiments) and associated tables/figures: the central claim that equalized coverage yields reliable conditional guarantees across the three actuator regimes (nominal, degraded, failed) is load-bearing for the paper's contribution to capability monitoring. However, coverage metrics appear to be reported only in aggregate form without separate empirical coverage rates, calibration plots, or per-regime tables for each actuator condition. This leaves the conditional coverage advantage over ordinary conformal prediction unverified and weakens support for the runtime heuristic application.

    Authors: We agree that explicit per-regime verification is important for supporting the conditional coverage claim and its relevance to capability monitoring. In the revised manuscript, we have added a new table (Table 4) reporting empirical coverage rates separately for the nominal, degraded, and failed actuator regimes, along with calibration plots (new Figure 7) that visualize coverage as a function of the conformity score for each regime. These additions directly compare the equalized coverage approach against standard conformal prediction and demonstrate improved conditional performance in the degraded and failed cases. We believe this addresses the concern and bolsters the justification for using the predictor as a runtime heuristic. revision: yes

Circularity Check

0 steps flagged

No circularity: proposal relies on established conformal methods without self-referential reductions

full rationale

The paper proposes a lightweight prediction model based on conformalized quantile regression with equalized coverage to predict low lateral deviation under nominal, degraded, and failed actuator conditions. No equations, derivations, or parameter-fitting steps are shown that would reduce any prediction to a fitted input by construction. The abstract explicitly states that equalized coverage methods are employed to address conditional coverage, but this is presented as an application of known techniques rather than a self-defined or self-cited uniqueness result. The central claim remains independent of any load-bearing self-citation chain or renaming of empirical patterns within the visible text.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Review is based on abstract only; no specific free parameters, axioms, or invented entities are described in the provided text.

axioms (1)
  • domain assumption Conformal prediction assumptions hold for the vehicle trajectory deviation data across nominal, degraded, and failed regimes.
    The coverage guarantees rely on these statistical assumptions being valid for the motion control data.

pith-pipeline@v0.9.0 · 5689 in / 1098 out tokens · 35214 ms · 2026-05-20T04:42:00.089520+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

58 extracted references · 58 canonical work pages · 2 internal anchors

  1. [1]

    AUTOtech.agil: Architecture and Technologies for Orchestrating Automotive Agility,

    R. van Kempen et al., “AUTOtech.agil: Architecture and Technologies for Orchestrating Automotive Agility,” inProc. 32. Aachen Colloq. Sustain. Mobility, 2023

    work page 2023

  2. [2]

    Schubert et al.,Architectural Requirements for Self-Aware and Self-Adaptive Automated Driving Systems: A Literature Review, to be published, 2026, Accessed: Feb

    R. Schubert et al.,Architectural Requirements for Self-Aware and Self-Adaptive Automated Driving Systems: A Literature Review, to be published, 2026, Accessed: Feb. 23 2026

    work page 2026

  3. [4]

    Conformal Prediction of Motion Control Performance for an Automated Vehicle in Presence of Actuator Degradations and Failures,

    R. Schubert et al., “Conformal Prediction of Motion Control Performance for an Automated Vehicle in Presence of Actuator Degradations and Failures,” inProc. Int. Conf. Intell. Transp. Syst., DOI: 10.1109/ITSC58415.2024.10920241, IEEE, 2024

    work page doi:10.1109/itsc58415.2024.10920241 2024

  4. [5]

    Towards a Functional System Architecture for Automated Vehicles

    S. Ulbrich et al., “Towards a functional system architecture for automated vehicles,”arXiv:1703.08557, 2017

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  5. [6]

    A Taxonomy to Unify Fault Tolerance Regimes for Automotive Systems: Defining Fail-Operational, Fail-Degraded, and Fail-Safe,

    T. Stolte et al., “A Taxonomy to Unify Fault Tolerance Regimes for Automotive Systems: Defining Fail-Operational, Fail-Degraded, and Fail-Safe,”IEEE Trans. Intell. V ehicles, 2021

    work page 2021

  6. [7]

    Supporting Safe Decision Making Through Holistic System-Level Representations & Monitoring – A Summary and Taxonomy of Self-Representation Concepts for Automated Vehicles,

    M. Nolte, I. Jatzkowski, S. Ernst, and M. Maurer, “Supporting Safe Decision Making Through Holistic System-Level Representations & Monitoring – A Summary and Taxonomy of Self-Representation Concepts for Automated Vehicles,”arXiv:2007.13807, 2020

    work page arXiv 2007

  7. [8]

    Representing the Unknown – Impact of Uncertainty on the Interaction between Decision Making and Trajectory Generation,

    M. Nolte, S. Ernst, J. Richelmann, and M. Maurer, “Representing the Unknown – Impact of Uncertainty on the Interaction between Decision Making and Trajectory Generation,” inProc. 21st Int. Conf. Intell. Transp. Syst. (ITSC), DOI: 10.1109/itsc.2018.8569490, 2018, pp. 2412–2418

    work page doi:10.1109/itsc.2018.8569490 2018

  8. [9]

    End-to-end system architectures in autonomous driving: Comparative analysis against modular design and technological exploration,

    L. Yuan, “End-to-end system architectures in autonomous driving: Comparative analysis against modular design and technological exploration,” inProc. Int. Conf. Mach. Learn. Automat., doi: 10.54254/2755-2721/102/20241031, vol. 102, 2024, pp. 141–147

    work page doi:10.54254/2755-2721/102/20241031 2024

  9. [10]

    Conformalized Quantile Regression,

    Y . Romano, E. Patterson, and E. Candes, “Conformalized Quantile Regression,” inProc. Advances Neural Inf. Process. Syst., vol. 32, 2019

    work page 2019

  10. [11]

    With malice toward none: Assessing uncertainty via equalized coverage,

    Y . Romano, R. F. Barber, C. Sabatti, and E. Cand`es, “With malice toward none: Assessing uncertainty via equalized coverage,”Harvard Data Sci. Rev., vol. 2, no. 2, p. 4, 2020, DOI: 10.1162/99608f92.03f00592

    work page doi:10.1162/99608f92.03f00592 2020

  11. [12]

    P. J. Antsaklis and K. M. Passino,An Introduction to Intelligent and Autonomous Control. Kluwer Academic Publishers, 1993

    work page 1993

  12. [13]

    Flexible Automatisierung von Straßenfahrzeugen mit Rechnersehen,

    M Maurer, “Flexible Automatisierung von Straßenfahrzeugen mit Rechnersehen,” PhD Thesis, Uni. der Bundw., 2000

    work page 2000

  13. [14]

    Komponenten zur automatischen Fahrzeugfuhrung in sehenden (semi-)autonomen Fahrzeugen,

    K.-H. Siedersberger, “Komponenten zur automatischen Fahrzeugfuhrung in sehenden (semi-)autonomen Fahrzeugen,” PhD Thesis, Universit ¨at der Bundeswehr, 2003

    work page 2003

  14. [15]

    Verhaltensentscheidung f ¨ur automatische Fahrzeuge mit Blickrichtungssteuerung,

    M. Pellkofer, “Verhaltensentscheidung f ¨ur automatische Fahrzeuge mit Blickrichtungssteuerung,” PhD Thesis, Uni. der Bundw., 2003

    work page 2003

  15. [16]

    Werte- und f ¨ahigkeitsbasierte Bewegungsplanung f ¨ur autonome Straßenfahrzeuge – Ein systemischer Ansatz,

    M. Nolte, “Werte- und f ¨ahigkeitsbasierte Bewegungsplanung f ¨ur autonome Straßenfahrzeuge – Ein systemischer Ansatz,” PhD Thesis, TU Braunschweig, 2024

    work page 2024

  16. [17]

    Fertigkeiten- und F ¨ahigkeitengraphen als Grundlage des sicheren Betriebs von automatisierten Fahrzeugen im ¨offentlichen Straßenverkehr,

    A. Reschka, “Fertigkeiten- und F ¨ahigkeitengraphen als Grundlage des sicheren Betriebs von automatisierten Fahrzeugen im ¨offentlichen Straßenverkehr,” PhD Thesis, TU Braunschweig, 2017

    work page 2017

  17. [18]

    Zum Fahrman ¨overbegriff im Kontext automatisierter Straßenfahrzeuge,

    I. Jatzkowski et al., “Zum Fahrman ¨overbegriff im Kontext automatisierter Straßenfahrzeuge,” Tech. Rep., 2021

    work page 2021

  18. [19]

    Towards a skill- and ability-based development process for self-aware automated road vehicles,

    M. Nolte et al., “Towards a skill- and ability-based development process for self-aware automated road vehicles,” inProc. ITSC, DOI: 10.1109/itsc.2017.8317814, 2017, pp. 1–6

    work page doi:10.1109/itsc.2017.8317814 2017

  19. [20]

    Towards Safety Concepts for Automated Vehicles by the Example of the Project UNICARagil,

    T. Stolte et al., “Towards Safety Concepts for Automated Vehicles by the Example of the Project UNICARagil,” inProc. 29th Aachen Colloq. Sustain. Mobility, 2020

    work page 2020

  20. [21]

    Crash-Prone Fault Combination Identification for Over-Actuated Vehicles During Evasive Maneuvers,

    A. Da Silva, C. Birkner, R. N. Jazar, and H. Marzbani, “Crash-Prone Fault Combination Identification for Over-Actuated Vehicles During Evasive Maneuvers,”IEEE Access, vol. 12, pp. 37 256–37 275, 2024, DOI: 10.1109/access.2024.3374524

    work page doi:10.1109/access.2024.3374524 2024

  21. [22]

    Efficient

    E. Ratner, C. J. Tomlin, and M. Likhachev, “Operating with Inaccurate Models by Integrating Control-Level Discrepancy Information into Planning,” inProc. Int. Conf. Robot. Automat. (ICRA), DOI: 10.1109/icra48891.2023.10161389, 2023, pp. 7823–7829

    work page doi:10.1109/icra48891.2023.10161389 2023

  22. [23]

    Active Learning of Abstract Plan Feasibility,

    M. Noseworthy et al., “Active Learning of Abstract Plan Feasibility,” in Proc. Robot. Sci. Syst. XVII, 2021

    work page 2021

  23. [24]

    Recursively Feasible Probabilistic Safe Online Learning with Control Barrier Functions,

    F. Castaneda et al., “Recursively Feasible Probabilistic Safe Online Learning with Control Barrier Functions,” inProc. arXiv preprint, arXiv:2208.10733, arXiv, 2023

    work page arXiv 2023

  24. [25]

    Learning When to Trust a Dynamics Model for Planning in Reduced State Spaces,

    D. McConachie, T. Power, P. Mitrano, and D. Berenson, “Learning When to Trust a Dynamics Model for Planning in Reduced State Spaces,”Robot. Automat. Lett., vol. 5, no. 2, pp. 3540–3547, 2020, DOI: 10.1109/lra.2020.2972858

    work page doi:10.1109/lra.2020.2972858 2020

  25. [26]

    Online Verification of Automated Road Vehicles Using Reachability Analysis,

    M. Althoff and J. M. Dolan, “Online Verification of Automated Road Vehicles Using Reachability Analysis,”IEEE Trans. Robot., vol. 30, no. 4, pp. 903–918, 2014, DOI: 10.1109/TRO.2014.2312453

    work page doi:10.1109/tro.2014.2312453 2014

  26. [27]

    Hamilton-Jacobi reachability: A brief overview and recent advances,

    S. Bansal, M. Chen, S. Herbert, and C. J. Tomlin, “Hamilton-Jacobi reachability: A brief overview and recent advances,” inProc. Conf. Decis. Control, DOI: 10.1109/cdc.2017.8263977, 2017, pp. 2242–2253

    work page doi:10.1109/cdc.2017.8263977 2017

  27. [28]

    On infusing reachability-based safety assurance within probabilistic planning frameworks for human-robot vehicle interactions,

    K. Leung et al., “On infusing reachability-based safety assurance within probabilistic planning frameworks for human-robot vehicle interactions,” inProc. Int. Symp. Exp. Robot., DOI: 10.1007/978-3-030-33950-0 48, 2020, pp. 561–574

    work page doi:10.1007/978-3-030-33950-0 2020

  28. [29]

    High dimensional reachability analysis: Addressing the curse of dimensionality in formal verification,

    M. Chen, “High dimensional reachability analysis: Addressing the curse of dimensionality in formal verification,” Electrical Engineering and Computer Sciences, Uni. California, Berkeley, Technical Report, 2017

    work page 2017

  29. [30]

    A new strategy for verifying reach-avoid specifications in neural feedback systems,

    S. I. Akinwande, S. M. Katz, M. J. Kochenderfer, and C. Barrett, “A new strategy for verifying reach-avoid specifications in neural feedback systems,” 2026. arXiv: 2601.08065[cs.AI]

    work page arXiv 2026

  30. [31]

    Sampling-based reachability analysis: A random set theory approach with adversarial sampling,

    T. Lew and M. Pavone, “Sampling-based reachability analysis: A random set theory approach with adversarial sampling,” inProc. Conf. robot Learn., 2021, pp. 2055–2070

    work page 2021

  31. [32]

    Decomposition of Reachable Sets and Tubes for a Class of Nonlinear Systems,

    M. Chen et al., “Decomposition of Reachable Sets and Tubes for a Class of Nonlinear Systems,”IEEE Trans. Autom. Control, vol. 63, no. 11, pp. 3675–3688, 2018, DOI: 10.1109/tac.2018.2797194

    work page doi:10.1109/tac.2018.2797194 2018

  32. [33]

    Damaged Airplane Trajectory Planning Based on Flight Envelope and Motion Primitives,

    D. Asadi, M. Sabzehparvar, E. M. Atkins, and H. A. Talebi, “Damaged Airplane Trajectory Planning Based on Flight Envelope and Motion Primitives,”J. Aircr ., vol. 51, no. 6, pp. 1740–1757, 2014, DOI: 10.2514/1.C032422

    work page doi:10.2514/1.c032422 2014

  33. [34]

    Stochastic reachability of a target tube: Theory and computation,

    A. P. Vinod and M. M. Oishi, “Stochastic reachability of a target tube: Theory and computation,”Automatica, vol. 125, p. 109 458, 2021, DOI: 10.1016/j.automatica.2020.109458

    work page doi:10.1016/j.automatica.2020.109458 2021

  34. [35]

    CP–NCBF: A Conformal Prediction-based Approach to Synthesize Verified Neural Control Barrier Functions,

    M. Tayal, A. Singh, P. Jagtap, and S. Kolathaya, “CP–NCBF: A Conformal Prediction-based Approach to Synthesize Verified Neural Control Barrier Functions,”arXiv preprint, 2025, arXiv:2503.17395, DOI: 10.48550/arXiv.2503.17395

    work page doi:10.48550/arxiv.2503.17395 2025

  35. [36]

    Conformal Prediction in the Loop: Risk-Aware Control Barrier Functions for Stochastic Systems with Data-Driven State Estimators,

    J. Zhang, B. Hoxha, G. Fainekos, and D. Panagou, “Conformal Prediction in the Loop: Risk-Aware Control Barrier Functions for Stochastic Systems with Data-Driven State Estimators,”IEEE Control Syst. Lett., 2025, DOI: 10.1109/LCSYS.2025.3571828

    work page doi:10.1109/lcsys.2025.3571828 2025

  36. [37]

    A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification

    A. N. Angelopoulos and S. Bates, “A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification,” inProc. arXiv preprint, arXiv:2107.07511 [cs, math, stat], arXiv, 2022

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  37. [38]

    Formal Verification and Control with Conformal Prediction: Practical Safety Guarantees for Autonomous Systems,

    L. Lindemann et al., “Formal Verification and Control with Conformal Prediction: Practical Safety Guarantees for Autonomous Systems,”IEEE Control Syst., 2025, DOI: 10.1109/MCS.2025.3611545

    work page doi:10.1109/mcs.2025.3611545 2025

  38. [39]

    Sensitivity Analysis for Vehicle Dynamics Models – An Approach to Model Quality Assessment for Automated Vehicles,

    M. Nolte, R. Schubert, C. Reisch, and M. Maurer, “Sensitivity Analysis for Vehicle Dynamics Models – An Approach to Model Quality Assessment for Automated Vehicles,” inProc. Intell. V ehicles Symp. (IV), DOI: 10.1109/iv47402.2020.9304801, 2020, pp. 1162–1169

    work page doi:10.1109/iv47402.2020.9304801 2020

  39. [40]

    A survey of deep learning applications to autonomous vehicle control,

    S. Kuutti et al., “A survey of deep learning applications to autonomous vehicle control,”IEEE Trans. Intell. Transp. Syst., vol. 22, no. 2, pp. 712–733, 2021, DOI: 10.1109/TITS.2019.2962338

    work page doi:10.1109/tits.2019.2962338 2021

  40. [41]

    ReachNN: Reachability Analysis of Neural-Network Controlled Systems,

    C. Huang et al., “ReachNN: Reachability Analysis of Neural-Network Controlled Systems,”ACM Trans. Embed. Comput. Syst., vol. 18, no. 5s, pp. 1–22, 2019

    work page 2019

  41. [42]

    Reachability Analysis of Neural Network Control Systems,

    C. Zhang, W. Ruan, and P. Xu, “Reachability Analysis of Neural Network Control Systems,” inProc. AAAI Conf. Artif. Intell., vol. 37, 2023, pp. 15 218–15 226

    work page 2023

  42. [43]

    Making the relationship between uncertainty estimation and safety less uncertain,

    V . Aravantinos and P. Schlicht, “Making the relationship between uncertainty estimation and safety less uncertain,” inProc. Des., Automat. & Test Europe Conf. & Exhib., DOI: 10.23919/date48585.2020.9116541, IEEE, 2020, pp. 1139–1144

    work page doi:10.23919/date48585.2020.9116541 2020

  43. [44]

    Online scalable gaussian processes with conformal prediction for guaranteed coverage,

    J. Xu, Q. Lu, and G. B. Giannakis, “Online scalable gaussian processes with conformal prediction for guaranteed coverage,” inProc. ICASSP 2025-2025 IEEE Int. Conf. Acoust., Speech Signal Process. (ICASSP), IEEE, 2025, pp. 1–5

    work page 2025

  44. [45]

    Toward Fault-Tolerant Vehicle Motion Control for Over-Actuated Automated Vehicles: A Non-Linear Model Predictive Approach,

    T. Stolte et al., “Toward Fault-Tolerant Vehicle Motion Control for Over-Actuated Automated Vehicles: A Non-Linear Model Predictive Approach,”IEEE Access, vol. 11, pp. 10 499–10 519, 2023, DOI: 10.1109/access.2023.3239518

    work page doi:10.1109/access.2023.3239518 2023

  45. [46]

    Towards Functional Safety in Drive-by-Wire Vehicles,

    P. Bergmiller, “Towards Functional Safety in Drive-by-Wire Vehicles,” PhD Thesis, TU Braunschweig, Braunschweig, 2014

    work page 2014

  46. [47]

    Model Predictive Control Based Trajectory Generation for Autonomous Vehicles – An Architectural Approach,

    M. Nolte, M. Rose, T. Stolte, and M. Maurer, “Model Predictive Control Based Trajectory Generation for Autonomous Vehicles – An Architectural Approach,” inProc. Intell. V ehicles Symp. (IV), DOI: 10.1109/ivs.2017.7995814, 2017, pp. 798–805

    work page doi:10.1109/ivs.2017.7995814 2017

  47. [48]

    A Controller Framework for Autonomous Drifting: Design, Stability, and Experimental Validation,

    R. Y . Hindiyeh and J. Christian Gerdes, “A Controller Framework for Autonomous Drifting: Design, Stability, and Experimental Validation,” J. Dynamic Syst., Meas., Control, vol. 136, no. 051015, 2014, DOI: 10.1115/1.4027471

    work page doi:10.1115/1.4027471 2014

  48. [49]

    Regression quantiles,

    R. Koenker and G. Bassett Jr, “Regression quantiles,”Econometrica: J. Econometric Soc., pp. 33–50, 1978, Publisher: JSTOR, DOI: 10.2307/1913643

    work page doi:10.2307/1913643 1978

  49. [50]

    An introduction to feature extraction,

    I. Guyon and A. Elisseeff, “An introduction to feature extraction,” inFeature extraction: foundations and applications, DOI: 10.1007/978-3-540-35488-8 1, Springer, 2006, pp. 1–25

    work page doi:10.1007/978-3-540-35488-8 2006

  50. [51]

    ODD-Centric Contextual Sensitivity Analysis Applied To A Non-Linear Vehicle Dynamics Model,

    R. Schubert, M. Nolte, A. de La Fortelle, and M. Maurer, “ODD-Centric Contextual Sensitivity Analysis Applied To A Non-Linear Vehicle Dynamics Model,” inProc. IV, DOI: 10.1109/iv55152.2023.10186729, 2023, pp. 1–8

    work page doi:10.1109/iv55152.2023.10186729 2023

  51. [52]

    Maximum relevance and minimum redundancy feature selection methods for a marketing machine learning platform,

    Z. Zhao, R. Anand, and M. Wang, “Maximum relevance and minimum redundancy feature selection methods for a marketing machine learning platform,”arXiv preprint arXiv:1908.05376, 2019. arXiv: 1908.05376 [stat.ML]

    work page arXiv 1908

  52. [53]

    Mazzanti,mrmr: Minimum redundancy maximum relevance feature selection, https://github.com/smazzanti/mrmr, GitHub repository, version 0.2.9, accessed Feb

    S. Mazzanti,mrmr: Minimum redundancy maximum relevance feature selection, https://github.com/smazzanti/mrmr, GitHub repository, version 0.2.9, accessed Feb. 2026, 2024, Accessed: Feb. 22 2026

    work page 2026

  53. [54]

    A simple test for heteroscedasticity and random coefficient variation,

    T. S. Breusch and A. R. Pagan, “A simple test for heteroscedasticity and random coefficient variation,”Econometrica, vol. 47, no. 5, pp. 1287–1294, 1979, DOI: 10.2307/1911963

    work page doi:10.2307/1911963 1979

  54. [55]

    Robust tests for the equality of variances,

    M. B. Brown and A. B. Forsythe, “Robust tests for the equality of variances,”J. Amer . statistical Assoc., vol. 69, no. 346, pp. 364–367, 1974, DOI: 10.1080/01621459.1974.10482955

    work page doi:10.1080/01621459.1974.10482955 1974

  55. [56]

    In: 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), pp

    R. Graubohm, N. F. Salem, M. Nolte, and M. Maurer, “On assumptions with respect to occlusions in urban environments for automated vehicle speed decisions,” inProc. Int. Conf. Intell. Transp. Syst., DOI: 10.1109/1TSC57777.2023.10422457, IEEE, 2023, pp. 738–745

    work page doi:10.1109/1tsc57777.2023.10422457 2023

  56. [57]

    Conformal prediction for semantically-aware autonomous perception in urban environments,

    A. Doula, T. G ¨udelh¨ofer, M. M ¨uhlh¨auser, and A. S. Guinea, “Conformal prediction for semantically-aware autonomous perception in urban environments,” inProc. Conf. Robot Learn., 2023

    work page 2023

  57. [58]

    Safe planning in dynamic environments using conformal prediction,

    L. Lindemann, M. Cleaveland, G. Shim, and G. J. Pappas, “Safe planning in dynamic environments using conformal prediction,” IEEE Robot. Automat. Lett., vol. 8, no. 8, pp. 5116–5123, 2023, DOI: 10.1109/LRA.2023.3292071

    work page doi:10.1109/lra.2023.3292071 2023

  58. [59]

    Safe-enhanced autonomous driving technology using conformal prediction results,

    J. Liang et al., “Safe-enhanced autonomous driving technology using conformal prediction results,”J. Phys.: Conf. Ser ., vol. 2861, no. 1, 2024, DOI: 10.1088/1742-6596/2861/1/012002

    work page doi:10.1088/1742-6596/2861/1/012002 2024