Pith. sign in

REVIEW 5 major objections 5 minor 1 cited by

Reduce Lap Time for Autonomous Racing with Curvature-Integrated MPCC Local Trajectory Planning Method

T0 review · 5 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read This paper claims that adding the racetrack centerline's smoothed curvature to the MPCC objective as a Gaussian reference-velocity profile reduces autonomous racing lap time by 11.4%–12.5%.

desk verdict Solid incremental MPCC extension with real vehicle experiments, but the headline lap-time gain is likely inflated by an asymmetric reference-velocity comparison. read the letter →

arxiv 2502.03695 v1 pith:77JM3HQ5 submitted 2025-02-06 cs.RO cs.SYeess.SY

Zhouheng Li , Lei Xie , Cheng Hu , Hongye Su This is my paper · ORCID
classification cs.ROcs.SYeess.SY
keywords autonomousracingmodelpredictivecontouringcontroltrajectoryplanningcurvature-awarevelocityprofileoptimizationlocal1:10scalevehicleGaussiancurvaturemapping
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

Traditional model predictive contouring control (MPCC) plans a racing trajectory online by maximizing progress along the racetrack centerline, but it ignores how sharply the centerline bends. This paper claims that feeding the centerline's smoothed, normalized curvature into the MPCC cost function—through a Gaussian mapping to a reference velocity profile—lets a 1:10 scale autonomous vehicle brake earlier into sharp corners and carry more speed out of them. On a physical track with four large curvature turns, the resulting CiMPCC method reports a mean lap time 11.8% shorter than the baseline MPCC and 11.7% shorter than a global path-velocity-decomposition planner, with mean velocity higher by 15.2% and 17.2%. If the claim transfers, curvature-aware online planning is a cheap, parameter-light way to close part of the gap between local and global racing trajectory methods.

What carries the argument

The central object is the normalized smooth curvature (NSC) of the centerline, $K_i^n=(K_i-K_{\min})/(K_{\max}-K_{\min})$, produced by moving-average smoothing of the discrete curvature and min-max normalization. A Gaussian map $g(K_i^n)=e^{-\alpha (K_i^n)^2}$ converts NSC into a reference velocity weight; the current value $\beta=g(K_{\text{cur}}^n)$ blends the cost between an aggressive velocity bound $\bar{v}$ and a safe bound $\underline{v}$. This continuous, bounded mapping is what lets the optimizer slow down in sharp bends without stalling and accelerate on straights, and it is the component the paper adds to the standard MPCC objective.

What would settle it

Run an ablation on the same physical track and vehicle with the curvature mapping disabled (set $\beta$ constant) while keeping the expert-derived velocity bounds in the cost; if lap time remains near the CiMPCC value, the curvature integration is not the cause of the reported gain. A second check is to repeat the comparison on a second track with different curvature statistics and see whether the 11–12% improvement persists.

Watch

Extended reading notes

Core claim

The paper's central claim is that integrating the racetrack centerline's normalized smooth curvature (NSC) into the MPCC objective improves racing performance. The curvature is smoothed with a moving average filter, normalized to $[0,1]$, and mapped through $g(K_i^n)=e^{-\alpha (K_i^n)^2}$ to a reference overall velocity. That reference enters the cost as a blending term: the planned velocity is pulled toward the aggressive upper bound when curvature is low and toward a safe lower bound when curvature is high, with a lower truncation coefficient that prevents over-braking. The authors assert that this mechanism, validated in seventeen laps on a physical 1:10 scale vehicle, reduces mean lap time by 11.8% versus MPCC and 11.7% versus RDM+OTG while raising mean velocity by 15.2% and 17.2%.

Load-bearing premise

The paper's lap-time gain rests on the assumption that the aggressive and safe velocity bounds, taken from the fastest expert-driven lap on this one track, are a fair and transferable benchmark rather than the main source of the improvement.

Editorial extensions

If this is right

  • On tracks with sharp curvature, CiMPCC should produce lap-time gains over plain MPCC whenever the expert-derived velocity bounds reflect the track's speed limits.
  • The curvature-to-reference-velocity term modifies only the objective, so it can be added to other model predictive contouring formulations without changing their constraints.
  • Because over 95% of CiMPCC solve times are under 0.0206 seconds, the method is compatible with real-time control on small-scale vehicles.
  • The planned velocity profile stays continuous, so it can be passed to lower-level controllers without a separate smoothing step.
  • CiMPCC's advantage over global planners is largest at corner entry and exit, where local curvature information lets it respond to the current vehicle state.

Reading between the lines

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

  • Editorial inference: the reported 11–12% gain likely depends on how much headroom the expert-driven aggressive velocity bound leaves; on a track where the expert baseline is already near the vehicle's limit, the curvature mapping would have less to add.
  • A testable extension would replace the fixed expert-derived bounds with an online-adapted bound, such as one updated from the achieved velocity of the previous lap, which would separate the mapping's contribution from the bound-tuning contribution.
  • The same normalized-curvature idea could transfer to overtaking by computing curvature of a reference race line rather than the centerline, which the paper lists as future work.
  • On full-scale vehicles, the kinematic model used here would need a dynamic model; the curvature-velocity mapping should still work, but the safe and aggressive bounds would need to encode tire friction limits.
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.

Referee Report

5 major / 5 minor

Summary. The paper proposes Curvature-Integrated MPCC (CiMPCC), a local trajectory planning method for autonomous racing that augments the MPCC cost function with a reference velocity profile obtained by mapping a smoothed, normalized curvature of the racetrack centerline through a Gaussian function. The method is validated on a 1:10 scale F1TENTH vehicle on a custom track over 17 laps. The authors report a 11.4%–12.5% reduction in lap time and a 14.9%–17.3% increase in mean velocity compared with a traditional MPCC baseline and an RDM+OTG global planning baseline. Code is made available.

Significance. If the claimed improvements are attributable to the proposed mechanism, the paper offers a simple, practical modification to MPCC that could benefit autonomous racing and related local planning problems. The use of a physical vehicle, the release of code, and the real-time computation analysis are concrete strengths. However, the central claim depends on a fair experimental comparison and on the reproducibility of the curvature-to-velocity mapping, both of which are currently not fully established.

major comments (5)
  1. [Section V, Table II and Eq. (10)] The comparison between MPCC and CiMPCC is confounded by asymmetric reference-velocity tuning. MPCC uses uref = [3.3, 0, 3] m/s, while CiMPCC is given vbar = [4.18, 3.8] m/s, where 3.8 m/s is 'the fastest velocity at which the expert drives the vehicle.' Since the new term in Eq. (10) explicitly penalizes deviations from vbar, CiMPCC is effectively handed a more aggressive speed reference than MPCC. The paper does not report an experiment in which MPCC is given the same vbar (or any reference above 3.3 m/s). Without such a control, the lap-time gain may reflect the expert-derived reference speed rather than the curvature-integration mechanism itself.
  2. [Section IV-B and IV-C] The two key tuning parameters of the proposed mapping are not reported: the MAF window width w in Eq. (7) and the exponent coefficient alpha in Eq. (9). Both directly determine the shape of the reference velocity profile and thus the planned velocity. Without these values, the method is not reproducible and it is impossible to assess whether the chosen mapping is a reasonable or an unusually well-tuned instance.
  3. [Section V, Table I] The experimental results are presented as max/min/mean over 17 laps with no standard deviations, confidence intervals, or statistical tests. The claimed 11.4%–12.5% lap-time improvement therefore lacks an assessment of run-to-run variability. Given that 17 laps is a modest sample, error bars or a paired significance test are needed to support the central quantitative claim.
  4. [Section II and Section V] The paper does not compare CiMPCC against the closest prior work, the curvature-aware MPCC of Lyons and Ferranti (reference [9]), which is discussed in the introduction and related work. Since that method also incorporates curvature into MPCC, a direct comparison is necessary to support the novelty and practical advantage of CiMPCC. Without it, the contribution over the existing curvature-aware MPCC is not demonstrated.
  5. [Section V, Eq. (10) and Table II] The reference velocity bounds vbar and v are derived from a single expert lap on the same test track and then used as fixed parameters in the cost function. This raises a transferability concern: the method's performance may depend on having an expert demonstration on every new track, and the paper provides no analysis of sensitivity to these bounds or to the choice of alpha and w. A small sensitivity study (e.g., varying alpha and vbar) would materially strengthen the claim that the method generalizes beyond the single reported track.
minor comments (5)
  1. [Section III, Eq. (5)] The discrete curvature formula in Eq. (5) is not the standard finite-difference curvature of a plane curve; the authors should clarify the approximation used and whether the denominator uses the chord length or something else.
  2. [Throughout] The baseline method is inconsistently named: 'RDM+OTG' in Table I and Fig. 7(c) appears as 'RMD+OTG' in Fig. 7(c) caption and in the text near Fig. 7 ('RMD+OTG'). Please unify the notation.
  3. [Section V] The sentence 'The trajectory planned by RDM+OTG is tracked by the proposed controller in [28]' is ambiguous; it should read 'the controller proposed in [28]'.
  4. [Fig. 6] Fig. 6(a) is referenced as showing 'control inputs of MPCC and CiMPCC in a complete lap,' but the y-axis label is 'Velocity [m/s]' and the legend indicates vl. Consider clarifying that this is the planned longitudinal velocity, not the steering input.
  5. [Section IV-C, Eq. (10)] The notation v with underbar and overbar is overloaded: v is used both as a general velocity vector and as the lower bound, and the boldface is not used consistently. Please introduce distinct symbols (e.g., v_low and v_high) to avoid confusion.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the lap-time gain is an experimental measurement, not a quantity derived from the fitted reference bounds.

full rationale

The claimed lap-time improvement is an empirical result from 17 physical laps (Table I), not a quantity derived from the proposed equations, so it cannot be circular in the derivation-chain sense. The curvature-dependent objective in Eq. (10) is an explicit design choice: the authors choose the Gaussian mapping of Eq. (9) and the weight beta = g(K_n^cur) so that the planned velocity tends toward the higher bound v_bar on low-curvature sections and toward the lower bound v on high-curvature sections. This is an ansatz, not a hidden identification of output with input. The bounds v_bar and v are adjustable parameters; although v_bar_p is taken from an expert lap on the same track, the measured mean velocity (3.351 m/s) and lap times are not equal to those bounds, so no fitted parameter is renamed as a prediction. The comparison with MPCC is possibly confounded by asymmetric reference-velocity settings (MPCC u_ref = [3.3, 0, 3] m/s versus CiMPCC v_bar_p = 3.8 m/s), but an unfair benchmark is a validity concern, not circularity. Self-citations appear for vehicle models and unrelated controllers and are not load-bearing for the central curvature-velocity claim. No equation in the paper reduces to its own input.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The method introduces no new physical entities. The central design choices are free parameters: the curvature smoothing window w, mapping sensitivity alpha, and the velocity bounds vbar and v_underline, the latter derived from an expert lap on the test track. The axioms are standard modeling and solver assumptions plus the ad hoc assumption that a hand-built curvature-velocity objective is appropriate for lap-time minimization.

free parameters (7)
  • alpha (curvature mapping sensitivity) = not reported
    Coefficient in g(K^n) = exp(-alpha*(K^n)^2) (Eq. 9) controls how strongly curvature reduces reference velocity; no value or tuning procedure given.
  • w (MAF window width) = not reported
    Window width in Eq. (7) for smoothing curvature; affects the normalized curvature and hence the velocity profile; no value given.
  • vbar (aggressive velocity vector) = [4.18, 3.8] m/s
    Upper bound in Eq. (10); set from the fastest expert-driven lap on the same track (Section V), so it is fitted to the test course and encodes the desired performance.
  • v_underline (safe velocity vector) = [2.72, 2.47] m/s
    Lower bound in Eq. (10); derived from vbar with a 0.65 discount factor, another hand-chosen tuning choice.
  • R3 (velocity cost weight) = diag([40, 40])
    Weight of the new curvature-velocity cost; hand-tuned; central to how strongly the reference profile is enforced.
  • R2 (control reference weight for CiMPCC) = diag([0, 10, 0])
    The reference velocity component of uref is omitted for CiMPCC; this asymmetry versus MPCC's R2=diag([40,10,40]) is a tuning choice that affects the comparison.
  • Shared MPCC tuning parameters = Q=diag([800,800]), gamma=40, R1=diag([10,3500,0]), Np=Nc=10
    Chosen by hand and not varied in any sensitivity analysis; they affect both baselines and the proposed method.
assumptions (5)
  • domain assumption The kinematic bicycle model (Eq. 11) adequately represents the vehicle's motion at racing speeds.
    Used in both MPCC and CiMPCC; no dynamics model or validation of model mismatch is provided.
  • standard math The contouring and lag error approximation used in MPCC (from [8]) is valid for the considered track and speeds.
    Borrowed from [8]; the paper does not re-derive or verify the approximation quality.
  • ad hoc to paper A curvature-dependent reference velocity bound constructed from smoothed normalized centerline curvature is an appropriate objective for minimizing lap time.
    Core design choice in Eq. (10); no proof that this objective yields time-optimal or near-time-optimal trajectories.
  • domain assumption The racetrack centerline, built offline with Cartographer, is accurate enough for the planner.
    Localization via particle filter; localization error is not quantified.
  • domain assumption The NLP solver (CasADi with multiple shooting) returns a sufficiently good local optimum in real time.
    Algorithm performance depends on warm-started local solutions; no optimality gap is reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reduce Lap Time for Autonomous Racing with Curvature-Integrated MPCC Local Trajectory Planning Method." pith.science (2026). https://pith.science/paper/77JM3HQ5

@misc{pith2026250203695,
  author       = {Pith},
  title        = {Pith review of: Reduce Lap Time for Autonomous Racing with Curvature-Integrated MPCC Local Trajectory Planning Method},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/77JM3HQ5}},
  note         = {Machine review of arXiv:2502.03695}
}
read the original abstract

The widespread application of autonomous driving technology has significantly advanced the field of autonomous racing. Model Predictive Contouring Control (MPCC) is a highly effective local trajectory planning method for autonomous racing. However, the traditional MPCC method struggles with racetracks that have significant curvature changes, limiting the performance of the vehicle during autonomous racing. To address this issue, we propose a curvature-integrated MPCC (CiMPCC) local trajectory planning method for autonomous racing. This method optimizes the velocity of the local trajectory based on the curvature of the racetrack centerline. The specific implementation involves mapping the curvature of the racetrack centerline to a reference velocity profile, which is then incorporated into the cost function for optimizing the velocity of the local trajectory. This reference velocity profile is created by normalizing and mapping the curvature of the racetrack centerline, thereby ensuring efficient and performance-oriented local trajectory planning in racetracks with significant curvature. The proposed CiMPCC method has been experimented on a self-built 1:10 scale F1TENTH racing vehicle deployed with ROS platform. The experimental results demonstrate that the proposed method achieves outstanding results on a challenging racetrack with sharp curvature, improving the overall lap time by 11.4%-12.5% compared to other autonomous racing trajectory planning methods. Our code is available at https://github.com/zhouhengli/CiMPCC.

Figures

Figures reproduced from arXiv: 2502.03695 by the authors.

Figure 1
Figure 1. The curvature-integrated MPCC (CiMPCC) trajectory planning method is illustrated in comparison with the traditional MPCC method. The CiMPCC method maps the curvature of the racetrack centerline into the optimization problem to optimize the velocity of the planned local trajectory, thereby reducing lap time. method do not consider the effect of significant changes in the curvature of the racetrack centerline on the h… view at source ↗
Figure 2
Figure 2. Schematic representation of the approximation of the contour error [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. The schematic diagram illustrates the implementation of the CiMPCC method, which is divided into two main sections: offline and online. The [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Fig. 4a and Fig. 4b visualize the process of computing the NSC. [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: The hardware and software architecture of the DDRA vehicle are used to validate the CiMPCC method, and the racetrack setting is demonstrated. ing to the planned trajectory of CiMPCC maintains a high level of continuity. The implementation of CiMPCC is described in deta…
Figure 6
Figure 6. Figure 6: The background color in the left figure corresponds to the NSC, corresponding to the point where the actual pose of the vehicle is projected onto [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: Results of three different trajectory planning methods applied to DDRA for autonomous racing. The trajectory points are obtained by Particle [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Mean velocity of a single lap, lap time, and mean lap time(MLT) [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: Histogram of CiMPCC computation time. The brown curve [PITH_FULL_IMAGE:figures/full_fig_p007_9.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Rapid and Safe Trajectory Planning over Diverse Scenes through Diffusion Composition

    cs.RO 2025-07 conditional novelty 4.0 of 10

    Diffusion models trained separately on static and dynamic scenes can be composed at test time to plan collision-free, kinematically feasible trajectories in unseen scenes, with real-time performance on an F1TENTH vehicle.

Reference graph

Works this paper leans on

28 extracted references · 21 canonical work pages · cited by 1 Pith paper

  1. [9]

    Curvature-aware model predictive contour- ing control,

    L. Lyons and L. Ferranti, “Curvature-aware model predictive contour- ing control,” in 2023 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2023, pp. 3204–3210

    work page 2023

  2. [1]

    Autonomous vehicles on the edge: A survey on autonomous vehicle racing,

    J. Betz, H. Zheng, A. Liniger, U. Rosolia, P. Karle, M. Behl, V . Krovi, and R. Mangharam, “Autonomous vehicles on the edge: A survey on autonomous vehicle racing,” IEEE Open Journal of Intelligent Transportation Systems, vol. 3, pp. 458–488, 2022

    2022

  3. [2]

    Minimum curvature trajectory planning and control for an autonomous race car,

    A. Heilmeier, A. Wischnewski, L. Hermansdorfer, J. Betz, M. Lienkamp, and B. Lohmann, “Minimum curvature trajectory planning and control for an autonomous race car,” Vehicle System Dynamics, 2019

    work page 2019

  4. [3]

    A sequential two-step algorithm for fast generation of vehicle racing trajectories,

    N. R. Kapania, J. Subosits, and J. Christian Gerdes, “A sequential two-step algorithm for fast generation of vehicle racing trajectories,” Journal of Dynamic Systems, Measurement, and Control , vol. 138, no. 9, p. 091005, 2016

    work page 2016

  5. [4]

    Manoeuvre gener- ation and control for automated highway driving,

    J. Nilsson, Y . Gao, A. Carvalho, and F. Borrelli, “Manoeuvre gener- ation and control for automated highway driving,” IFAC Proceedings Volumes, vol. 47, no. 3, pp. 6301–6306, 2014

    work page 2014

  6. [5]

    A hierarchical approach for strategic motion planning in autonomous racing,

    R. Reiter, J. Hoffmann, J. Boedecker, and M. Diehl, “A hierarchical approach for strategic motion planning in autonomous racing,” in 2023 European Control Conference (ECC) . IEEE, 2023, pp. 1–8

    work page 2023

  7. [6]

    Autonomous high-speed overtaking of intelligent chassis using fast iterative model predictive control,

    H. Chu, D. Meng, S. Huang, M. Tian, J. Zhang, B. Gao, and H. Chen, “Autonomous high-speed overtaking of intelligent chassis using fast iterative model predictive control,” IEEE Transactions on Transportation Electrification, vol. 10, no. 1, pp. 1244–1256, 2023

    work page 2023

  8. [7]

    A nonlinear model- predictive contouring controller for shared control driving assistance in high-performance scenarios,

    E. Picotti, M. Bruschetta, E. Mion, and A. Beghi, “A nonlinear model- predictive contouring controller for shared control driving assistance in high-performance scenarios,” IEEE Transactions on Systems, Man, and Cybernetics: Systems , vol. 53, no. 1, pp. 204–215, 2022

    work page 2022

Show all 28 references
  1. [8]

    Optimization-based au- tonomous racing of 1: 43 scale rc cars,

    A. Liniger, A. Domahidi, and M. Morari, “Optimization-based au- tonomous racing of 1: 43 scale rc cars,” Optimal Control Applications and Methods, vol. 36, no. 5, pp. 628–647, 2015

    2015

  2. [10]

    Race driver model,

    F. Braghin, F. Cheli, S. Melzi, and E. Sabbioni, “Race driver model,” Computers & Structures , vol. 86, no. 13-14, pp. 1503–1516, 2008

    2008

  3. [11]

    Minimum curvature trajectory planning and control for an autonomous race car,

    A. Heilmeier, A. Wischnewski, L. Hermansdorfer, J. Betz, M. Lienkamp, and B. Lohmann, “Minimum curvature trajectory planning and control for an autonomous race car,” Vehicle System Dynamics, 2020

    2020

  4. [12]

    An aggressive cornering framework for autonomous vehicles combining trajectory planning and drift control,

    W. Weng, C. Hu, Z. Li, H. Su, and L. Xie, “An aggressive cornering framework for autonomous vehicles combining trajectory planning and drift control,” in 2024 IEEE Intelligent Vehicles Symposium (IV) . IEEE, 2024, pp. 2749–2755

    2024

  5. [13]

    Nonlinear model predictive control for optimal motion planning in autonomous race cars,

    S. Sivashangaran, D. Patel, and A. Eskandarian, “Nonlinear model predictive control for optimal motion planning in autonomous race cars,” IFAC-PapersOnLine, vol. 55, no. 37, pp. 645–650, 2022

    2022

  6. [14]

    Multilayer graph-based trajectory planning for race vehicles in dynamic sce- narios,

    T. Stahl, A. Wischnewski, J. Betz, and M. Lienkamp, “Multilayer graph-based trajectory planning for race vehicles in dynamic sce- narios,” in 2019 IEEE intelligent transportation systems conference (ITSC). IEEE, 2019, pp. 3149–3154

    2019

  7. [15]

    Optimization-based colli- sion avoidance,

    X. Zhang, A. Liniger, and F. Borrelli, “Optimization-based colli- sion avoidance,” IEEE Transactions on Control Systems Technology , vol. 29, no. 3, pp. 972–983, 2020

    2020

  8. [16]

    Trajectory tracking control for au- tonomous parking using reduced-horizon model predictive control,

    Z. Zhang, L. Xie, and H. Su, “Trajectory tracking control for au- tonomous parking using reduced-horizon model predictive control,” in 2020 4th CAA International Conference on Vehicular Control and Intelligence (CVCI). IEEE, pp. 401–405

    2020

  9. [17]

    Learning-based mpc controller for drift control of autonomous vehicles,

    X. Zhou, C. Hu, R. Duo, H. Xiong, Y . Qi, Z. Zhang, H. Su, and L. Xie, “Learning-based mpc controller for drift control of autonomous vehicles,” in 2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC) . IEEE, 2022, pp. 322–328

    2022

  10. [18]

    An mpc-based controller framework for agile maneuvering of autonomous vehicles,

    Y . Qi, Z. Zhang, C. Hu, X. Zhou, L. Xie, and H. Su, “An mpc-based controller framework for agile maneuvering of autonomous vehicles,” in 2021 IEEE Intelligent Vehicles Symposium (IV) . IEEE, 2021, pp. 1228–1234

    2021

  11. [19]

    Combined fast control of drifting state and trajectory tracking for autonomous vehicles based on mpc controller,

    C. Hu, X. Zhou, R. Duo, H. Xiong, Y . Qi, Z. Zhang, and L. Xie, “Combined fast control of drifting state and trajectory tracking for autonomous vehicles based on mpc controller,” in 2022 International Conference on Robotics and Automation (ICRA) . IEEE, 2022, pp. 1373–1379

    2022

  12. [20]

    A guaranteed collision-free trajectory planning method for autonomous parking,

    Z. Zhang, S. Lu, L. Xie, H. Su, D. Li, Q. Wang, and W. Xu, “A guaranteed collision-free trajectory planning method for autonomous parking,” IET Intelligent Transport Systems , vol. 15, no. 2, pp. 331– 343, 2021

    2021

  13. [21]

    The scientist and engineer’s guide to digital signal processing,

    S. W. Smith et al., “The scientist and engineer’s guide to digital signal processing,” 1997

    1997

  14. [22]

    Deep drifting: Autonomous drifting of arbitrary trajectories using deep reinforcement learning,

    F. Domberg, C. C. Wembers, H. Patel, and G. Schildbach, “Deep drifting: Autonomous drifting of arbitrary trajectories using deep reinforcement learning,” in 2022 International Conference on Robotics and Automation (ICRA) . IEEE, 2022, pp. 7753–7759

    2022

  15. [23]

    F1tenth: An open-source evaluation environment for continuous control and reinforcement learning,

    M. O’Kelly, H. Zheng, D. Karthik, and R. Mangharam, “F1tenth: An open-source evaluation environment for continuous control and reinforcement learning,” Proceedings of Machine Learning Research , vol. 123, 2020

    2020

  16. [24]

    CasADi – A software framework for nonlinear optimization and optimal control,

    J. A. E. Andersson, J. Gillis, G. Horn, J. B. Rawlings, and M. Diehl, “CasADi – A software framework for nonlinear optimization and optimal control,” Mathematical Programming Computation , vol. 11, no. 1, pp. 1–36, 2019

    2019

  17. [25]

    Real-time loop closure in 2d lidar slam,

    W. Hess, D. Kohler, H. Rapp, and D. Andor, “Real-time loop closure in 2d lidar slam,” in 2016 IEEE International Conference on Robotics and Automation (ICRA) , 2016, pp. 1271–1278

    2016

  18. [26]

    Cddt: Fast approximate 2d ray casting for accelerated localization,

    C. H. Walsh and S. Karaman, “Cddt: Fast approximate 2d ray casting for accelerated localization,” in 2018 IEEE International Conference on Robotics and Automation (ICRA) , 2018, pp. 3677–3684

    2018

  19. [27]

    Optimal trajectory generation for dynamic street scenarios in a frenet frame,

    M. Werling, J. Ziegler, S. Kammel, and S. Thrun, “Optimal trajectory generation for dynamic street scenarios in a frenet frame,” in 2010 IEEE international conference on robotics and automation . IEEE, 2010, pp. 987–993

    2010

  20. [28]

    Model-and acceleration-based pursuit controller for high-performance autonomous racing,

    J. Becker, N. Imholz, L. Schwarzenbach, E. Ghignone, N. Baumann, and M. Magno, “Model-and acceleration-based pursuit controller for high-performance autonomous racing,” in 2023 IEEE International Conference on Robotics and Automation (ICRA) . IEEE, 2023, pp. 5276–5283

    2023

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.