An Intention-driven Lane Change Framework Considering Heterogeneous Dynamic Cooperation in Mixed-traffic Environment

Haichao Liu; Jun Ma; Xiaoyun Qiu; Xinhu Zheng; Yue Pan

REVIEW 4 major objections 4 minor 46 references

An Intention-driven Lane Change Framework Considering Heterogeneous Dynamic Cooperation in Mixed-traffic Environment

T0 review · 4 major / 4 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read This paper claims that explicitly modeling the heterogeneous cooperative intentions of surrounding drivers—through a learnable intrinsic score and an interactive dynamic score—enables autonomous vehicles to predict lane change decisions wit

desk verdict Competent integration of style clustering, cooperation scores, and BC-IRL, but the headline classification numbers are not yet trustworthy due to sample-level leakage and a latent cooperation score that is never validated against ground truth. read the letter →

arxiv 2509.22550 v4 pith:5JNPHDAB submitted 2025-09-26 cs.RO

Xiaoyun Qiu , Haichao Liu , Yue Pan , Jun Ma , Xinhu Zheng This is my paper · ORCID

classification cs.RO

keywords lanechangedecisionintentionpredictioninversereinforcementlearningbehaviorcloningcooperationscoredrivingstylemixedtrafficmodelpredictivecontrol

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

The paper proposes an intention-driven lane change framework for mixed traffic, where autonomous vehicles must share roads with human drivers of varied styles. Its central claim is that a dual-perspective cooperation score—one part learned from the driver's stable style, one part derived from the momentary interaction—captures how willing a nearby human driver is to yield, and that feeding this score into a behavior-cloning plus inverse-reinforcement-learning decision module improves lane change prediction over homogeneous-behavior baselines. On the NGSIM highway dataset, the framework reports 94.2% accuracy and 94.3% F1-score for classifying lane change versus lane keeping, a 4–15% F1 improvement over representative baselines for lane change recognition. A sympathetic reader would care because accurate anticipation of human yielding is a safety bottleneck for autonomous driving in mixed traffic.

What carries the argument

The central mechanism is the dual-perspective cooperation score c_final = α·c_intr + (1−α)·c_inter, where c_intr is a learnable function of ego-inner features plus the recognized driving style, c_inter is a learnable function of interaction features (relative distance and velocity), and α is a learned gate. This score is injected into an LSTM policy net (trained with reward-weighted behavior cloning) and a reward net (trained with an IRL-style preference loss). A second mechanism is the Max-Ent IRL trajectory predictor whose rewards are linear in efficiency, safety, and comfort features, feeding a sigmoid-based reference path tracked by an MPC controller. The cooperation score is what carrie

What would settle it

Train the full model with c_final replaced by a random vector of the same dimension, and compare classification accuracy. If accuracy remains at ~0.94, the cooperation score contributes no real signal. Alternatively, collect human annotations of yielding for NGSIM T-Rear vehicles and check whether c_final predicts them better than chance; a null result would refute the intention-driven interpretation.

Watch

Extended reading notes

Core claim

The paper's core discovery is that modeling inter-driver heterogeneity via a learned cooperation score improves lane change decision-making. The score fuses an intrinsic learnable component conditioned on a three-category driving style (aggressive, normal, conservative) with an interactive dynamic component built from relative distance and velocity, merged by a trainable gate. The authors train an LSTM policy with a reward-weighted behavior-cloning objective, an inverse-reinforcement-learning reward network, and a cooperation regularization, all end-to-end. The resulting model reaches 0.9418 accuracy and 0.9427 F1-score on NGSIM, with lane change precision 0.8911, recall 0.9239, and F1 0.907

Load-bearing premise

The learned cooperation score is assumed to represent a human driver's genuine willingness to cooperate, even though it is trained end-to-end to improve lane-change classification on the same NGSIM labels it is evaluated against, with no ground-truth cooperation labels.

Editorial extensions

If this is right

If the central claim is correct, lane change planners that condition on a predicted cooperation score will misclassify human yielding less often, directly addressing the roughly 17% of severe crashes attributed to sudden lane changes.
The dual-perspective score can be learned without hand-labeled cooperation, using unsupervised style clustering plus end-to-end training, so the approach scales to other trajectory datasets.
Combining reward-weighted behavior cloning with IRL stabilizes training in small demonstration sets, making it feasible to learn human-like policies from limited expert data.
The IRL-MPC motion planner produces faster and smoother lane changes than an IDM baseline, suggesting the same reward features can improve both decision and control.

Reading between the lines

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

A natural next test is to compare the learned c_final against human-annotated yielding events, such as whether the T-Rear actually brakes or changes lane, which would confirm the score captures true cooperative intention rather than a statistical artifact of the LC/LK labels.
The discrete three-style taxonomy could be replaced by continuous driving-style embeddings, potentially capturing within-style variation the current model smooths over; the paper itself notes the three-category simplification as a limitation.
The decision module could be retrained on other maneuvers, such as merging or cut-in response, where cooperation inference is equally safety-critical; the paper explicitly lists multi-intention and multi-agent extensions as future work.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

You should know two things about this paper. First, it's a reasonable integration of existing ideas: driving style clustering, a dual-component cooperation score (intrinsic LCS and interactive DCS) fused by a learned gate, reward-weighted BC plus a pairwise IRL loss, and an MPC stage using IRL-predicted trajectories. Each piece is established, but the combination is new and the pipeline is clearly explained. The ablation study and the convergence analysis of the gate are sensibly done, and the lane-change duration validation against a log-normal fit is a nice touch.

Second, the central empirical claim—94.2% accuracy, 94.3% F1, and 4–15% F1 improvement over baselines—is not yet defensible. The train/validation split is on 2-second sliding windows, not on episodes or vehicles. Since adjacent windows from the same trajectory are nearly identical, the classifier can memorize trajectory-specific cues. The stress-test note about this is correct, and the paper's own numbers are inconsistent: the abstract says 96.98%, the full text says 94.2%, and Table II and Table III report different F1 values for the full model (0.9421 vs. 0.9379). No error bars, seeds, or significance tests appear anywhere. That alone makes the headline result unstable.

The bigger conceptual soft spot is the cooperation score. It is a latent variable trained end-to-end to improve classification on the same NGSIM labels used for evaluation. There is no ground-truth cooperation annotation, so the "intention-driven" framing is largely a re-description of the classification signal. The paper does not address this circularity, even though it does list other limitations honestly. The IRL loss (Eq. 10) is also a nonstandard pairwise preference objective, not the MaxEnt IRL the text implies; that's not a fatal flaw, but it needs justification.

The motion-planning section is only qualitatively evaluated with a single case study, so the claimed safety and efficiency benefits are not quantified. That said, the modules are standard and the direction is sensible.

Who is this for? Researchers working on lane-change decision-making who want to see how style clustering and cooperation scores can be wired into a BC-IRL pipeline. It deserves a serious referee, but it needs major revision: episode-level cross-validation, corrected numbers, significance tests, and either ground-truth cooperation annotation or a much more cautious interpretation of what the learned score means. I wouldn't cite the classification numbers as they stand.

Referee Report

4 major / 4 minor

Summary. The paper proposes an intention-driven lane change framework for mixed traffic, combining driving-style clustering/recognition, a dual-perspective cooperation score (LCS/DCS), a BC–IRL decision module, and an IRL–MPC motion planner. The authors claim superior lane-change classification on NGSIM (abstract: 96.98%; main text: 94.2% accuracy, 94.3% F1; Table II: 0.9418 accuracy and 0.9421–0.9427 F1 depending on the sentence), with 4–15% F1 improvement over baselines, plus qualitative evidence that IRL-based trajectory prediction enables faster, safer lane changes.

Significance. If the empirical claims were established, the framework would be a useful integration of several known ideas—style-conditional cooperation scoring, reward-weighted imitation, and IRL–MPC planning—into a single pipeline. The intended contributions are interpretability of cooperation and explicit modeling of inter-driver heterogeneity. However, the central classification result is not yet trustworthy because of a sample-level split that leaks temporally overlapping windows, internally inconsistent accuracy numbers, and the absence of error bars or significance testing. The cooperation-score interpretation is also under-validated. The paper's scope and topic are appropriate for the venue, and the framework is plausible, but the evidence provided in the current version is insufficient to support the headline claims.

major comments (4)

[Section III-B and V-A, Table II] The evaluation protocol does not support the claimed classification accuracy. The dataset is created by segmenting 307 lane-change episodes into 2-s sliding windows (Section III-B), yielding 5,295 samples with strong temporal overlap within each episode. Section V-A then splits the samples randomly 80/20 with stratified sampling. Consequently, windows from the same vehicle/episode appear in both training and validation, allowing the classifier to memorize episode-specific cues rather than generalize to new drivers. This directly undermines the 'heterogeneous driver' claim. The authors should use an episode-level or vehicle-level split, report metrics across multiple seeds or folds, and provide standard deviations or confidence intervals. Without this, the 0.9418 vs 0.9207 advantage over VWC and the LC F1 advantage are not established.
[Abstract, Section V-B, Table II, Table III, Fig. 5(c)] The reported numerical results are internally inconsistent. The abstract says 96.98% accuracy; the conclusion and abstract text say 94.2% accuracy / 94.3% F1; Section V-B reports accuracy 0.9418 and F1-score 0.9427, but Table II's overall F1 column reads 0.9421 for 'Ours'; Table III gives the full model as 0.9367 accuracy and 0.9379 F1; and the confusion-matrix discussion in Section V-B says there are 171 false positives and 99 false negatives, which would be 270 errors on 1,059 validation samples—an accuracy around 0.745, not 0.9418. These discrepancies must be reconciled and a single consistent evaluation protocol reported before the classification claim can be accepted.
[Section IV-B, Eq. (3), and Section V-D] The 'cooperation score' is a latent variable trained end-to-end on the same lane-change labels used for evaluation. There is no ground truth for cooperative intention, so the claim that LCS/DCS quantify actual willingness to cooperate is not supported. Section V-D only shows that the training-time value of c_final converges and that the gating weight changes; it does not demonstrate that the score corresponds to observable cooperative behavior. A concrete test is needed, e.g., validating the score against independent yielding/gap-acceptance annotations, or showing that predicted cooperation scores correlate with human judgments or with actual target-vehicle braking/acceleration in held-out interaction episodes. Without such validation, the 'intention prediction' contribution may be an interpretive label on a classification feature rather than a measured human quantity.
[Section IV-C, Eq. (10)] The IRL loss in Eq. (10) is a pairwise preference objective with an L2 penalty and a margin regularizer, not the maximum-entropy IRL objective described in Eq. (12). The paper calls this 'IRL' and claims it recovers latent rewards, but the connection between the reward network's pairwise log-likelihood and the trajectory-level MaxEnt IRL used in the motion-planning module is not established. The authors should either formalize the decision-module objective as a proper IRL method (and specify how it relates to Eq. 12) or rename it and avoid the IRL claim. Additionally, the hyperparameters beta, lambda_2, and lambda_s in Eqs. (7) and (10) are never specified, so the results are not reproducible.

minor comments (4)

[Section IV-A] The number of PCA components and the choice of K=3 in K-means are not justified; these affect the style labels that propagate through the rest of the framework. Please report these values and, ideally, a sensitivity analysis.
[Section IV-D, Eq. (19)] The sigmoid-based reference path is introduced without justification and is called 'the expected longitudinal distance' with no derivation. Since the motion-planning results are only qualitative, this is not blocking, but the assumption should be stated more explicitly and its effect on trajectory quality discussed.
[Section V-C, Table III] The ablation table header contains both 'Best F1' and 'F1-score' columns; the meaning of 'Best F1' is unclear. The full model's accuracy/F1 in Table III also differs from Table II for what appears to be the same configuration; this should be explained (e.g., different split or initialization).
[General] There are several formatting issues: inconsistent spacing in 'A Vs', duplicated reference-like text in the abstract footnote, and the confusion-matrix description in Fig. 5(c) mentioning counts that do not match the reported accuracy. A careful proofread and consistent notation for superscripts such as r^(a_i) vs r^(1-a_i) would improve readability.

Circularity Check

1 steps flagged · score 5.0 of 10

The cooperation score is a latent variable fitted end-to-end on the same lane-change labels, so its interpretation as 'intention prediction' is partially circular; the classification results are additionally undermined by sample-level leakage and internal inconsistencies.

fitted input called prediction [Section IV-B (Eqs. 1-3, 5-11) and Section V-D]
"The parameters θ_intr are trainable and receive gradients both from the cooperation regularization loss and from the downstream decision-making objective... First, the LCS steadily increases throughout training and converges around 0.60, indicating that the model increasingly relies on intrinsic driver style features to capture cooperative intentions."

The cooperation score c_final is presented as 'intention prediction' (Fig. 1), but no ground-truth cooperation label exists. It is a differentiable function of state/style features optimized end-to-end with L_BC and L_IRL on the same NGSIM LC/LK labels used for evaluation. The learned values of LCS/DCS are therefore shaped by the classification objective itself. Section V-D then interprets these fitted values as evidence that the model 'captures cooperative intentions,' which is a post-hoc semantic label on a discriminant trained to separate LC from LK. The 'prediction' of cooperation is thus equivalent to the classification signal by construction, not an independently validated behavioral quantity. The ablation improvement from adding LCS/DCS is expected when adding flexible parameters an

full rationale

The central classification result is a legitimate supervised-learning claim, but the paper's distinctive 'intention-driven' contribution rests on c_final being a meaningful measure of human cooperative intent. That meaning is not derived from any external ground truth; it is imposed after training on the same binary labels used for evaluation. This is partial circularity: the LC/LK accuracy numbers have independent content, but the 'intention prediction' semantics reduce to the fitted classifier. Separate non-circular validity concerns further weaken the empirical claims: the 80/20 split is performed on 2-s sliding-window samples (5,295 samples from only 307 episodes), so temporally adjacent windows from the same trajectory likely appear in both training and validation; no error bars or significance tests are reported; and the abstract's 96.98% accuracy conflicts with the 94.18% in Table II and 93.67% in Table III. Self-citations to [36] and [39] are used for standard duration-distribution and reward-formulation support and are not load-bearing.

Assumptions & free parameters 6 free parameters · 6 assumptions · 3 invented entities

The central claim rests on a latent cooperation score that is never measured against real cooperative behavior. All the learned entities are trained on the same NGSIM labels used for evaluation, so the 'intention prediction' component is not independently grounded.

free parameters (6)

Number of driving style clusters K = 3
Chosen by hand to represent aggressive, normal, conservative; no model selection or external validation.
Reward temperature beta = not reported
In Eq. (7), controls how much reward differences influence BC weights; value not specified.
Regularization weights lambda2, lambda_s = not reported
In Eq. (10), L2 and margin penalties; values not reported.
PCA components = not reported
Features transformed via PCA; number of retained components not specified.
Neutral cooperation target = 0.5
Lcoop penalizes deviations from 0.5; the interpretation that 0.5 means neutral is imposed.
Lane change detection threshold delta = not reported
In Alg. 2, threshold on lateral speed for start/end detection; value not given.

assumptions (6)

domain assumption Six statistical features (mean/max/std of velocity and acceleration) suffice to characterize driving style.
Section IV-A; the entire style recognition rests on this feature set.
domain assumption The state variables {vE, aE, vf, dEf, vtr, dEtr} are sufficient for lane change decision-making.
Section III-A; other factors like lane geometry, traffic density, and driver state are ignored.
domain assumption Linear reward combination with efficiency, safety, comfort features captures human driving preferences.
Eq. (14)-(18); the paper relies on [39] for this formulation.
domain assumption NGSIM I-80 trajectories are accurate enough for training and evaluation.
Section III-B; known noise issues in NGSIM are addressed with filtering but not otherwise validated.
ad hoc to paper The sigmoid function provides an appropriate reference path for lane changes.
Eq. (19); no kinematic comparison to other path geometries.
domain assumption Markov decision process assumption for the lane change decision problem.
Section III-A; though the LSTM policy suggests non-Markovian state.

invented entities (3)

Cooperation score c_final (composed of LCS and DCS)
purpose: Quantify surrounding driver willingness to cooperate with the ego lane change.
Learned latent variable without ground truth labels; interpretability is asserted from convergence values, not external validation.
Driving style categories (aggressive, normal, conservative)
purpose: Discrete labels to condition cooperation and decision modules.
Result of unsupervised K-means on six features; no validation against psychological or behavioral ground truth.
Intrinsic and interactive cooperation scores (LCS/DCS) as separate signals
purpose: Decompose style-dependent and context-dependent cooperation.
They are components of the same learned latent; their 'complementary roles' are inferred post hoc.

how reviews work

0 comments

Cite this review

Pith. "Pith review of An Intention-driven Lane Change Framework Considering Heterogeneous Dynamic Cooperation in Mixed-traffic Environment." pith.science (2026). https://pith.science/paper/5JNPHDAB

@misc{pith2026250922550,
  author       = {Pith},
  title        = {Pith review of: An Intention-driven Lane Change Framework Considering Heterogeneous Dynamic Cooperation in Mixed-traffic Environment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5JNPHDAB}},
  note         = {Machine review of arXiv:2509.22550}
}

read the original abstract

In mixed-traffic environments, autonomous vehicles (AVs) must interact with heterogeneous human-driven vehicles (HVs) whose intentions and driving styles vary across individuals and scenarios. Such variability introduces uncertainty into lane change interactions, where safety and efficiency critically depend on accurately anticipating surrounding drivers' cooperative responses. Existing methods often oversimplify these interactions by assuming uniform or fixed behavioral patterns. To address this limitation, we propose an intention-driven lane change framework that integrates driving-style recognition with cooperation-aware decision-making and motion-planning. A deep learning-based classifier identifies distinct human driving styles in real time. We then introduce a dual-perspective cooperation score composed of intrinsic style-dependent tendencies and interactive dynamic components, enabling interpretable and adaptive intention prediction and quantitative inference. A decision-making module combines behavior cloning (BC) and inverse reinforcement learning (IRL) to determine lane change feasibility. Later, a coordinated motion-planning architecture integrating IRL-based intention inference with model predictive control (MPC) is established to generate collision-free and socially compliant trajectories. Experiments on the NGSIM dataset show that the proposed decision-making model outperforms representative rule-based and learning-based baselines, achieving 96.98% accuracy in lane change classification. Motion-planning evaluations further demonstrate improved maneuver success and execution stability in mixed-traffic environments. These results validate the effectiveness of structured cooperation modeling for intention-driven autonomous lane changes.

Figures

Figures reproduced from arXiv: 2509.22550 by the authors.

**Figure 2.** Analysis of lane change duration times. (a) The log-normal distribution [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Performance of the lane change start and end point detection algorithm. Spatial distribution of lane change points along vehicle trajectories is illustrated. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: Analysis of driving-style representation: (a) clustering results show [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

**Figure 5.** Figure 5: Performance evaluation of the proposed intention-driven lane change decision-making model, including training dynamics, validation metrics, and [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

**Figure 6.** Figure 6: Convergence analysis of the cooperation-aware intention module: (a) intrinsic and interactive cooperation scores (LCS/DCS); (b) gating weight [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]

**Figure 7.** Figure 7: Qualitative results. Comparison of MPC-based lane change performance with predicted trajectories of the T-Rear vehicle. The black vehicle represents [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]

**Figure 8.** Figure 8: The control input for the ego vehicle when executing lane change [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 1 linked inside Pith

[1]

A survey of autonomous driving: Common practices and emerging technologies,

E. Yurtsever, J. Lambert, A. Carballo, and K. Takeda, “A survey of autonomous driving: Common practices and emerging technologies,” IEEE Access, vol. 8, pp. 58 443–58 469, 2020

2020
[2]

Toward human-vehicle col- laboration: Review and perspectives on human-centered collaborative automated driving,

Y . Xing, C. Lv, D. Cao, and P. Hang, “Toward human-vehicle col- laboration: Review and perspectives on human-centered collaborative automated driving,”Transportation Research Part C: Emerging Tech- nologies, vol. 128, p. 103199, 2021

2021
[3]

Milestones in autonomous driving and intelligent vehicles: Survey of surveys,

L. Chen, Y . Li, C. Huang, B. Li, Y . Xing, D. Tian, L. Li, Z. Hu, X. Na, Z. Liet al., “Milestones in autonomous driving and intelligent vehicles: Survey of surveys,”IEEE Transactions on Intelligent Vehicles, vol. 8, no. 2, pp. 1046–1056, 2022

2022
[4]

Planning and decision- making for autonomous vehicles,

W. Schwarting, J. Alonso-Mora, and D. Rus, “Planning and decision- making for autonomous vehicles,”Annual Review of Control, Robotics, and Autonomous Systems, vol. 1, pp. 187–210, 2018

2018
[5]

A model for the structure of lane-changing decisions,

P. G. Gipps, “A model for the structure of lane-changing decisions,” Transportation Research Part B: Methodological, vol. 20, no. 5, pp. 403–414, 1986

1986
[6]

Recent developments and research needs in modeling lane changing,

Z. Zheng, “Recent developments and research needs in modeling lane changing,”Transportation Research Part B: Methodological, vol. 60, pp. 16–32, 2014

2014
[7]

Review of lane-changing maneuvers of connected and automated vehicles: models, algorithms and traffic impact analyses,

Z. Wang, X. Shi, and X. Li, “Review of lane-changing maneuvers of connected and automated vehicles: models, algorithms and traffic impact analyses,”Journal of the Indian Institute of Science, vol. 99, pp. 589– 599, 2019

2019
[8]

Inverse reinforcement learning based: Segmented lane-change trajectory planning with consideration of interactive driving intention,

Y . Sun, Y . Chu, T. Xu, J. Li, and X. Ji, “Inverse reinforcement learning based: Segmented lane-change trajectory planning with consideration of interactive driving intention,”IEEE Transactions on Vehicular Technol- ogy, vol. 71, no. 11, pp. 11 395–11 407, 2022

2022

Show all 46 references

[9]

Probabilistic prediction of inter- active driving behavior via hierarchical inverse reinforcement learning,

L. Sun, W. Zhan, and M. Tomizuka, “Probabilistic prediction of inter- active driving behavior via hierarchical inverse reinforcement learning,” in2018 21st International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2018, pp. 2111–2117

2018
[10]

A cooperation- aware lane change method for automated vehicles,

Z. Sheng, L. Liu, S. Xue, D. Zhao, M. Jiang, and D. Li, “A cooperation- aware lane change method for automated vehicles,”IEEE Transactions on Intelligent Transportation Systems, 2022

2022
[11]

Factors affecting lane change crashes,

M. Shawky, “Factors affecting lane change crashes,”IATSS research, vol. 44, no. 2, pp. 155–161, 2020

2020
[12]

Learning driving styles for autonomous vehicles from demonstration,

M. Kuderer, S. Gulati, and W. Burgard, “Learning driving styles for autonomous vehicles from demonstration,” in2015 IEEE International Conference on Robotics and Automation (ICRA), 2015, pp. 2641–2646

2015
[13]

General lane-changing model mobil for car-following models,

A. Kesting, M. Treiber, and D. Helbing, “General lane-changing model mobil for car-following models,”Transportation Research Record, vol. 1999, no. 1, pp. 86–94, 2007

1999
[14]

An intelli- gent lane-changing behavior prediction and decision-making strategy for an autonomous vehicle,

W. Wang, T. Qie, C. Yang, W. Liu, C. Xiang, and K. Huang, “An intelli- gent lane-changing behavior prediction and decision-making strategy for an autonomous vehicle,”IEEE Transactions on Industrial Electronics, vol. 69, no. 3, pp. 2927–2937, 2021

2021
[15]

Lane change decision- making through deep reinforcement learning with rule-based con- straints,

J. Wang, Q. Zhang, D. Zhao, and Y . Chen, “Lane change decision- making through deep reinforcement learning with rule-based con- straints,” in2019 International Joint Conference on Neural Networks (IJCNN). IEEE, 2019, pp. 1–6

2019
[16]

Rule-based safety-critical control design using control barrier functions with application to au- tonomous lane change,

S. He, J. Zeng, B. Zhang, and K. Sreenath, “Rule-based safety-critical control design using control barrier functions with application to au- tonomous lane change,” in2021 American Control Conference (ACC). IEEE, 2021, pp. 178–185

2021
[17]

Using support vector machines for lane-change detection,

H. M. Mandalia and M. D. D. Salvucci, “Using support vector machines for lane-change detection,” inProceedings of the Human Factors and Ergonomics Society Annual Meeting, vol. 49, no. 22. SAGE Publica- tions Sage CA: Los Angeles, CA, 2005, pp. 1965–1969

2005
[18]

Lane changing prediction at highway lane drops using support vector machine and artificial neural network classi- fiers,

Y . Dou, F. Yan, and D. Feng, “Lane changing prediction at highway lane drops using support vector machine and artificial neural network classi- fiers,” in2016 IEEE International Conference on Advanced Intelligent Mechatronics (AIM). IEEE, 2016, pp. 901–906

2016
[19]

A deep learning method for lane changing situation assessment and decision making,

X. Liu, J. Liang, and B. Xu, “A deep learning method for lane changing situation assessment and decision making,”IEEE Access, vol. 7, pp. 133 749–133 759, 2019. 13

2019
[20]

A vehicle driving intention prediction method based on gated dual tower transformer model for autonomous driving,

X. Guo, H. Jia, Q. Huang, Q. Luo, N. Wang, and Z. Mao, “A vehicle driving intention prediction method based on gated dual tower transformer model for autonomous driving,”Expert Systems with Appli- cations, p. 128000, 2025

2025
[21]

Combining planning and deep reinforcement learning in tactical decision making for autonomous driving,

C.-J. Hoel, K. Driggs-Campbell, K. Wolff, L. Laine, and M. J. Kochen- derfer, “Combining planning and deep reinforcement learning in tactical decision making for autonomous driving,”IEEE Transactions on Intel- ligent Vehicles, vol. 5, no. 2, pp. 294–305, 2019

2019
[22]

Human-like autonomous car-following model with deep reinforcement learning,

M. Zhu, X. Wang, and Y . Wang, “Human-like autonomous car-following model with deep reinforcement learning,”Transportation Research Part C: Emerging Technologies, vol. 97, pp. 348–368, 2018

2018
[23]

A survey of inverse reinforcement learning: Challenges, methods and progress,

S. Arora and P. Doshi, “A survey of inverse reinforcement learning: Challenges, methods and progress,”Artificial Intelligence, vol. 297, p. 103500, 2021

2021
[24]

Driving behavior modeling using natu- ralistic human driving data with inverse reinforcement learning,

Z. Huang, J. Wu, and C. Lv, “Driving behavior modeling using natu- ralistic human driving data with inverse reinforcement learning,”IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 8, pp. 10 239–10 251, 2021

2021
[25]

Improved car-following strategy based on merging behavior prediction of adjacent vehicle from natural- istic driving data,

Y . Guo, Q. Sun, R. Fu, and C. Wang, “Improved car-following strategy based on merging behavior prediction of adjacent vehicle from natural- istic driving data,”IEEE Access, vol. 7, pp. 44 258–44 268, 2019

2019
[26]

Dual transformer based prediction for lane change intentions and trajectories in mixed traffic environment,

K. Gao, X. Li, B. Chen, L. Hu, J. Liu, R. Du, and Y . Li, “Dual transformer based prediction for lane change intentions and trajectories in mixed traffic environment,”IEEE Transactions on Intelligent Trans- portation Systems, 2023

2023
[27]

Lc-rss: A lane- change responsibility-sensitive safety framework based on data-driven lane-change prediction,

N. Zhao, B. Wang, K. Zhang, Y . Lu, R. Luo, and R. Su, “Lc-rss: A lane- change responsibility-sensitive safety framework based on data-driven lane-change prediction,”IEEE Transactions on Intelligent Vehicles, 2023

2023
[28]

Socially-aware autonomous driving: Inferring yielding intentions for safer interactions,

J. Wang, Y . Jin, H. Taghavifar, F. Ding, and C. Wei, “Socially-aware autonomous driving: Inferring yielding intentions for safer interactions,” arXiv preprint arXiv:2504.20004, 2025

2025 arXiv
[29]

A novel lane change decision-making model of autonomous vehicle based on support vector machine,

Y . Liu, X. Wang, L. Li, S. Cheng, and Z. Chen, “A novel lane change decision-making model of autonomous vehicle based on support vector machine,”IEEE Access, vol. 7, pp. 26 543–26 550, 2019

2019
[30]

Personalized driver/vehicle lane change models for adas,

V . A. Butakov and P. Ioannou, “Personalized driver/vehicle lane change models for adas,”IEEE Transactions on Vehicular Technology, vol. 64, no. 10, pp. 4422–4431, 2014

2014
[31]

An inverse reinforcement learning approach for customizing automated lane change systems,

J. Liu, L. N. Boyle, and A. G. Banerjee, “An inverse reinforcement learning approach for customizing automated lane change systems,” IEEE Transactions on Vehicular Technology, vol. 71, no. 9, pp. 9261– 9271, 2022

2022
[32]

Driving with style: Inverse reinforcement learning in general-purpose planning for automated driving,

S. Rosbach, V . James, S. Großjohann, S. Homoceanu, and S. Roth, “Driving with style: Inverse reinforcement learning in general-purpose planning for automated driving,” in2019 IEEE/RSJ International Con- ference on Intelligent Robots and Systems (IROS). IEEE, 2019, pp. 2658–2665

2019
[33]

Personalized lane-change as- sistance system with driver behavior identification,

B. Zhu, S. Yan, J. Zhao, and W. Deng, “Personalized lane-change as- sistance system with driver behavior identification,”IEEE Transactions on Vehicular Technology, vol. 67, no. 11, pp. 10 293–10 306, 2018

2018
[34]

Human-like lane-change control strategy for connected and autonomous vehicles to improve interactions with human- driven vehicles,

Y . Liu and S. Peeta, “Human-like lane-change control strategy for connected and autonomous vehicles to improve interactions with human- driven vehicles,”Transportation Research Part C: Emerging Technolo- gies, vol. 177, p. 105211, 2025

2025
[35]

Interstate 80 freeway dataset,

J. Halkia and J. Colyar, “Interstate 80 freeway dataset,”Federal Highway Administration, US Department of Transportation, 2006

2006
[36]

Analysing and modelling of discretionary lane change duration considering driver heterogeneity,

G. Li, Z. Yang, Y . Pan, and J. Ma, “Analysing and modelling of discretionary lane change duration considering driver heterogeneity,” Transportmetrica B: Transport Dynamics, vol. 11, no. 1, pp. 343–360, 2023

2023
[37]

Modeling duration of lane changes,

T. Toledo and D. Zohar, “Modeling duration of lane changes,”Trans- portation Research Record, vol. 1999, no. 1, pp. 71–78, 2007

1999
[38]

Maximum entropy inverse reinforcement learning

B. D. Ziebart, A. L. Maas, J. A. Bagnell, A. K. Deyet al., “Maximum entropy inverse reinforcement learning.” inAAAI, vol. 8. Chicago, IL, USA, 2008, pp. 1433–1438

2008
[39]

Driving style- aware car-following considering cut-in tendencies of adjacent vehicles with inverse reinforcement learning,

X. Qiu, Y . Pan, M. Zhu, L. Yang, and X. Zheng, “Driving style- aware car-following considering cut-in tendencies of adjacent vehicles with inverse reinforcement learning,” in2024 IEEE Intelligent Vehicles Symposium (IV). IEEE, 2024, pp. 1329–1336

2024
[40]

Trajectory reference generation and guidance control for autonomous vehicle lane change maneuver,

M. Ammour, R. Orjuela, and M. Basset, “Trajectory reference generation and guidance control for autonomous vehicle lane change maneuver,” in2020 28th Mediterranean Conference on Control and Automation (MED). IEEE, 2020, pp. 13–18

2020
[41]

UDMC: Unified decision-making and control framework for urban autonomous driving with motion prediction of traffic participants,

H. Liu, K. Chen, Y . Li, Z. Huang, M. Liu, and J. Ma, “UDMC: Unified decision-making and control framework for urban autonomous driving with motion prediction of traffic participants,”IEEE Transactions on Intelligent Transportation Systems, vol. 26, no. 5, pp. 1–16, 2025

2025
[42]

Enhanced intelligent driver model to access the impact of driving strategies on traffic capacity,

A. Kesting, M. Treiber, and D. Helbing, “Enhanced intelligent driver model to access the impact of driving strategies on traffic capacity,” Philosophical Transactions of the Royal Society A: Mathematical, Phys- ical and Engineering Sciences, vol. 368, no. 1928, pp. 4585–4605, 2010

1928
[43]

Online prediction of lane change with a hierarchical learning- based approach,

X. Liao, Z. Wang, X. Zhao, Z. Zhao, K. Han, P. Tiwari, M. J. Barth, and G. Wu, “Online prediction of lane change with a hierarchical learning- based approach,” in2022 International Conference on Robotics and Automation (ICRA). IEEE, 2022, pp. 948–954

2022
[44]

High-level decision making for automated highway driving via behavior cloning,

L. Wang, C. Fernandez, and C. Stiller, “High-level decision making for automated highway driving via behavior cloning,”IEEE Transactions on Intelligent Vehicles, vol. 8, no. 1, 2023

2023
[45]

Variable weight combination model for lane-changing prediction of human-driven vehicle,

X. Wang, Y . Zan, J. Zheng, Y . Tao, Y . Xiao, T. Liet al., “Variable weight combination model for lane-changing prediction of human-driven vehicle,”IEEE Transactions on Intelligent Vehicles, 2024

2024
[46]

Deepsignals: Predicting intent of drivers through visual signals,

D. Frossard, E. Kee, and R. Urtasun, “Deepsignals: Predicting intent of drivers through visual signals,” in2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019, pp. 9697–9703

2019

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [1]

A survey of autonomous driving: Common practices and emerging technologies,

E. Yurtsever, J. Lambert, A. Carballo, and K. Takeda, “A survey of autonomous driving: Common practices and emerging technologies,” IEEE Access, vol. 8, pp. 58 443–58 469, 2020

2020

[2] [2]

Toward human-vehicle col- laboration: Review and perspectives on human-centered collaborative automated driving,

Y . Xing, C. Lv, D. Cao, and P. Hang, “Toward human-vehicle col- laboration: Review and perspectives on human-centered collaborative automated driving,”Transportation Research Part C: Emerging Tech- nologies, vol. 128, p. 103199, 2021

2021

[3] [3]

Milestones in autonomous driving and intelligent vehicles: Survey of surveys,

L. Chen, Y . Li, C. Huang, B. Li, Y . Xing, D. Tian, L. Li, Z. Hu, X. Na, Z. Liet al., “Milestones in autonomous driving and intelligent vehicles: Survey of surveys,”IEEE Transactions on Intelligent Vehicles, vol. 8, no. 2, pp. 1046–1056, 2022

2022

[4] [4]

Planning and decision- making for autonomous vehicles,

W. Schwarting, J. Alonso-Mora, and D. Rus, “Planning and decision- making for autonomous vehicles,”Annual Review of Control, Robotics, and Autonomous Systems, vol. 1, pp. 187–210, 2018

2018

[5] [5]

A model for the structure of lane-changing decisions,

P. G. Gipps, “A model for the structure of lane-changing decisions,” Transportation Research Part B: Methodological, vol. 20, no. 5, pp. 403–414, 1986

1986

[6] [6]

Recent developments and research needs in modeling lane changing,

Z. Zheng, “Recent developments and research needs in modeling lane changing,”Transportation Research Part B: Methodological, vol. 60, pp. 16–32, 2014

2014

[7] [7]

Review of lane-changing maneuvers of connected and automated vehicles: models, algorithms and traffic impact analyses,

Z. Wang, X. Shi, and X. Li, “Review of lane-changing maneuvers of connected and automated vehicles: models, algorithms and traffic impact analyses,”Journal of the Indian Institute of Science, vol. 99, pp. 589– 599, 2019

2019

[8] [8]

Inverse reinforcement learning based: Segmented lane-change trajectory planning with consideration of interactive driving intention,

Y . Sun, Y . Chu, T. Xu, J. Li, and X. Ji, “Inverse reinforcement learning based: Segmented lane-change trajectory planning with consideration of interactive driving intention,”IEEE Transactions on Vehicular Technol- ogy, vol. 71, no. 11, pp. 11 395–11 407, 2022

2022