Pith. sign in

REVIEW 4 major objections 5 minor 82 references

Motion Forecasting for Autonomous Vehicles: A Survey

T0 review · 4 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash

Pith's one-line read This survey proposes a formal problem formulation for autonomous-vehicle motion forecasting and organizes recent work along two axes: scenario-based versus perception-based input, and supervised versus self-supervised learning.

desk verdict A useful broad survey of motion forecasting that is undermined by an internally inconsistent supervised/self-supervised classification and several factual errors. read the letter →

arxiv 2502.08664 v1 pith:E4WP3EMI submitted 2025-02-10 cs.RO cs.AI

classification cs.ROcs.AI
keywords motionforecastingautonomousdrivingtrajectorypredictionscenario-basedperception-basedsupervisedlearningself-supervisedevaluationmetrics
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

This survey tries to give autonomous-vehicle motion forecasting a common vocabulary and a workable map. It defines target agents, ego and surrounding agents, formalizes scenario-based input as past agent states plus HD maps, and formalizes perception-based input as raw sensor data mapped directly to future trajectories. It then sorts recent deep-learning methods by training paradigm: supervised learning, which dominates, and self-supervised learning, which is emerging. A sympathetic reader would care because a stable formulation and shared datasets and metrics are what let separate results be compared and progress measured.

What carries the argument

The load-bearing object is the formal problem formulation itself: scenario-based input is written as $T_{obs} = \{X_{TAs}, HDMap, X_{SAs}, Others\}$ with predicted trajectory $\hat{X}_{TA} = \{(x_{ta,t+1}, y_{ta,t+1}), ..., (x_{ta,t+T}, y_{ta,t+T})\}$, and perception-based forecasting is written as a learned map $f: Z_t \rightarrow X_{t+T}$ from raw sensor data to future states. The taxonomy axes carry the argument: scenario-based versus perception-based input, and supervised versus self-supervised training. The pipeline diagrams, an encoder-decoder structure for supervised learning and a pretrain-finetune structure for self-supervised learning, are the organizational device that lets every surveyed method be placed.

What would settle it

A concrete check would be to count, across a recent conference's motion-forecasting papers, how many methods the survey's two axes place without an arbitrary tie-breaker; if a large fraction (say, more than a third) must be assigned to both input categories or to both training paradigms, the proposed partition fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that motion forecasting can be stated as a clean mathematical problem with two standard pipelines. In scenario-based forecasting, the input is the historical states of target and surrounding agents plus an HD map, and the output is a set of future positions for each target agent over a prediction horizon; in perception-based forecasting, a learned function maps raw LiDAR, camera, and radar data directly to future trajectories. Recent methods then split into supervised-learning architectures, built from rasterized, graph, or attention encoders followed by anchor-conditioned or anchor-free decoders, and self-supervised-learning architectures, which pretrain encoders with contrastive learning, masked autoencoding, auxiliary tasks, or language-model-style token prediction before fine-tuning. The survey also organizes evaluation into geometry-level, probabilistic-level, and task-level metrics, and lines up the main public datasets behind them.

Load-bearing premise

The survey's organization depends on the assumption that every method falls cleanly into one of two input types and one of two training paradigms, rather than straddling both.

Editorial extensions

If this is right

  • A common problem formulation means results from scenario-based and perception-based papers can be described in the same terms, making cross-paper comparison less ambiguous.
  • The supervised/self-supervised split gives newcomers a direct route into the field: supervised methods supply the baseline architecture, while self-supervised methods supply the pretraining toolbox.
  • The three-level metric stack (geometry, probability, task) makes explicit that a forecast that is accurate in meters may still be poor for downstream planning, and vice versa.
  • Placing joint detection-and-prediction models inside the supervised section implies that perception-based forecasting is still understood primarily through its training objective rather than through a separate architectural theory.

Reading between the lines

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

  • Editorial inference: a natural extension the paper leaves implicit is a third axis of input representation granularity (rasterized, vectorized, BEV, raw), since several surveyed methods differ more in representation than in training paradigm.
  • Editorial inference: the taxonomy predicts that self-supervised pretraining will become the default initialization for supervised forecasting models, a prediction that is testable by tracking whether new forecasting papers report gains from pretrained encoders.
  • Editorial inference: the formal distinction between marginal prediction $p(s_1)p(s_2)...p(s_n)$ and joint prediction $p(s_1,s_2,...,s_n)$ could be pushed further, because a joint model that factorizes poorly in practice would sit at the boundary and suggest a separate axis for interaction modeling.
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

4 major / 5 minor

Summary. This manuscript is a survey of motion forecasting for autonomous vehicles. It proposes a formal problem formulation, contrasts scenario-based and perception-based forecasting, reviews public datasets and evaluation metrics, and organizes recent deep-learning methods into supervised and self-supervised learning architectures. The survey covers representative work in temporal-spatial encoding, trajectory decoding, masked autoencoders, data augmentation, and language-modeling approaches, and it closes with future research directions.

Significance. A well-organized survey of this rapidly moving area is potentially valuable as a reference, and the paper does bring together a broad set of methods, datasets, and metrics, including recent self-supervised and language-modeling lines. The proposed scenario/perception dichotomy is a useful organizing axis, and the detailed enumeration of encoder and decoder designs gives the reader a compact entry point into the literature. However, the paper's central claim to classify methods into supervised and self-supervised learning is not internally consistent, and several technical definitions in the metrics section are imprecise. These issues must be corrected before the survey can serve as a reliable reference.

major comments (4)
  1. [Section 5.2.5] The subsection 'Language Modeling Methods' is placed under 'Self-Supervised Learning-based Architecture' (Section 5.2), but the paper's own definition of self-supervised learning in Section 5.2 is representation learning from large-scale unlabeled data via a pretrain/finetune pipeline (Figure 7). MotionLM, STR, and AMP are autoregressive sequence models trained with maximum-likelihood objectives on ground-truth trajectory tokens; they do not have an unlabeled pretraining phase or a finetuning phase in which pretrained encoders are loaded. Grouping them under SSL contradicts the stated definition and collapses the supervised/self-supervised dichotomy promised in the abstract. The authors should either move these methods to the supervised-learning section, or revise the definition and explain how next-token prediction qualifies as self-supervised under that definition.
  2. [Section 4.1 and Table 1] The Waymo Open Motion Dataset is described in Section 4.1 as containing scenes that are '20 seconds long and sampled at 10 Hz,' but Table 1 lists the Waymo duration as '9s.' These two statements are inconsistent. Since Table 1 is intended as a compact reference for dataset properties, the authors should reconcile the discrepancy or state explicitly whether '9s' refers to a different quantity (for example, the forecasting horizon).
  3. [Section 4.2, Miss Rate] Miss Rate (MR) is defined as 'the number of scenarios in which all predicted trajectories deviate by more than 2.0 meters from the ground truth, as measured by endpoint error.' As written, this is a count, not a rate. In the motion-forecasting literature, MR is conventionally a fraction of scenarios, and a count would not be comparable across datasets of different sizes. The definition should be corrected to a normalized rate, or the metric should be renamed if a raw count is intended.
  4. [Section 4.2, Eq. (16)] In the task-level metric, the function f(a, |∇ˆs(t:T)c|) is not defined, and the notation |∇ˆs(t:T)c| is unclear without specifying over which variables the gradient is taken and how the norm is computed. Additionally, the sentence introducing the metric says s_a^{(t:T)} and \hat{s}_a^{(t:T)} are the 'predicted positions' and 'ground truth positions,' respectively, while the formula places them in the opposite order inside Metric(·). The equation and surrounding text should be revised so that the weighting scheme, the gradient notation, and the argument order are unambiguous and reproducible.
minor comments (5)
  1. [Figure 1] The vertical-axis label 'Y ear' contains a typo and should read 'Year.'
  2. [Eq. (13) and Eq. (14)] The summation limits and indices in minADE and minFDE are not motivated: the sum runs from t = T_obs to T_pred - 1, but the relationship between T_obs, T_pred, and the forecast horizon is not stated in the text. A brief explanation of these indices would help readers apply the formulas.
  3. [Section 4.1] The paragraph on the Interaction dataset reports 'over 41,000' trajectories and 'more than 990 minutes' of driving, but the table lists no length or scene count for Interaction. Adding these values to Table 1 would make the comparison complete.
  4. [References] Several citations in the text, such as 'Liu et al (2021)' in Table 1, do not have a matching full reference entry in the bibliography; the authors should check that all cited works appear in the reference list with complete author information.
  5. [Section 3] The challenge titled 'Fusion of road information' is described at a high level but does not cite specific methods that address HDMap alignment and fusion; a pointer to one or two representative approaches would make the discussion more actionable.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the survey's problem formulation and taxonomy are stipulative, and no derivation or fitted prediction is reduced to its own inputs.

full rationale

This is a survey, not a derivation or prediction paper. The problem formulation in Sections 2.1 and 2.2 defines inputs, outputs, and pipelines (Equations 1 through 12) as formal definitions rather than as results derived from a fitted model; no parameter is fitted and no quantity is 'predicted' from data that was used to define it. The supervised/self-supervised division in Sections 5 and 6 is a proposed organizational scheme, and a taxonomy is stipulative by nature: organizing papers by a chosen axis does not reduce a derived claim to its inputs. The placement of MotionLM, STR, and AMP in Section 5.2.5 ('Language Modeling Methods') is hard to reconcile with Section 5.2's own pretrain-finetune definition of self-supervised learning, since those methods are trained with maximum-likelihood objectives on ground-truth token sequences rather than through a separate unlabeled-data pretraining phase; however, that is an internal consistency and correctness concern about the taxonomy, not a circular step in a derivation. I found no load-bearing self-citation: the cited MTR and MTR++ works share a surname with an author but are distinct prior papers by other researchers, and no 'uniqueness theorem' or prior-work assumption is invoked to force the survey's choices. No fitted input is renamed as a prediction, and no equation is equivalent by construction to a claimed output. The survey is therefore self-contained as a literature organization, and the appropriate circularity verdict is 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

No free parameters or new entities appear because this is a survey. The assumptions above are structural premises of the taxonomy and of the reference material, not independent evidence.

assumptions (3)
  • domain assumption Motion forecasting methods can be cleanly partitioned into scenario-based and perception-based pipelines.
    Introduced in Section 2.2 and Figure 5; the entire organization depends on this partition, but some perception-based joint detection and prediction models are later discussed inside the supervised-learning section, so the partition is not shown to be exclusive.
  • domain assumption The supervised-learning versus self-supervised-learning distinction is the most useful second taxonomy.
    Section 1 and Section 5 use this split without comparative evidence; the paper does not justify why this axis is more informative than alternatives such as encoder type or output representation.
  • domain assumption The listed datasets and metrics are representative and correctly characterized.
    Section 4 and Table 1: the review relies on these summaries, but Table 1 has an inconsistent duration for Waymo and Miss Rate is described as a count, so the reader must trust the summaries without verification.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Motion Forecasting for Autonomous Vehicles: A Survey." pith.science (2026). https://pith.science/paper/E4WP3EMI

@misc{pith2026250208664,
  author       = {Pith},
  title        = {Pith review of: Motion Forecasting for Autonomous Vehicles: A Survey},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/E4WP3EMI}},
  note         = {Machine review of arXiv:2502.08664}
}
read the original abstract

In recent years, the field of autonomous driving has attracted increasingly significant public interest. Accurately forecasting the future behavior of various traffic participants is essential for the decision-making of Autonomous Vehicles (AVs). In this paper, we focus on both scenario-based and perception-based motion forecasting for AVs. We propose a formal problem formulation for motion forecasting and summarize the main challenges confronting this area of research. We also detail representative datasets and evaluation metrics pertinent to this field. Furthermore, this study classifies recent research into two main categories: supervised learning and self-supervised learning, reflecting the evolving paradigms in both scenario-based and perception-based motion forecasting. In the context of supervised learning, we thoroughly examine and analyze each key element of the methodology. For self-supervised learning, we summarize commonly adopted techniques. The paper concludes and discusses potential research directions, aiming to propel progress in this vital area of AV technology.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

82 extracted references · 70 canonical work pages

  1. [1]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 8295--8305

    Aydemir G, Akan AK, G \"u ney F (2023) Adapt: Efficient multi-agent trajectory prediction with adaptation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 8295--8305

  2. [2]

    arXiv preprint arXiv:221004672

    Azevedo C, Gilles T, Sabatini S, et al (2022) Exploiting map information for self-supervised learning in motion forecasting. arXiv preprint arXiv:221004672

  3. [3]

    Barth A, Franke U (2008) Where will the oncoming vehicle be the next second? In: 2008 IEEE Intelligent Vehicles Symposium, IEEE, pp 1068--1073

  4. [4]

    In: Conference on Robot Learning, PMLR, pp 1793--1805

    Bhattacharyya P, Huang C, Czarnecki K (2023) Ssl-lanes: Self-supervised learning for motion forecasting in autonomous driving. In: Conference on Robot Learning, PMLR, pp 1793--1805

  5. [5]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 11621--11631

    Caesar H, Bankiti V, Lang AH, et al (2020) nuscenes: A multimodal dataset for autonomous driving. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 11621--11631

  6. [6]

    In: European conference on computer vision, Springer, pp 213--229

    Carion N, Massa F, Synnaeve G, et al (2020) End-to-end object detection with transformers. In: European conference on computer vision, Springer, pp 213--229

  7. [7]

    In: Conference on Robot Learning, PMLR, pp 947--956

    Casas S, Luo W, Urtasun R (2018) Intentnet: Learning to predict intention from raw sensor data. In: Conference on Robot Learning, PMLR, pp 947--956

  8. [8]

    In: 2020 IEEE International Conference on Robotics and Automation (ICRA), IEEE, pp 9491--9497

    Casas S, Gulino C, Liao R, et al (2020) Spagnn: Spatially-aware graph neural networks for relational behavior forecasting from sensor data. In: 2020 IEEE International Conference on Robotics and Automation (ICRA), IEEE, pp 9491--9497

Show all 82 references
  1. [9]

    arXiv preprint arXiv:191005449

    Chai Y, Sapp B, Bansal M, et al (2019) Multipath: Multiple probabilistic anchor trajectory hypotheses for behavior prediction. arXiv preprint arXiv:191005449

  2. [10]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 8748--8757

    Chang MF, Lambert J, Sangkloy P, et al (2019) Argoverse: 3d tracking and forecasting with rich maps. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 8748--8757

  3. [11]

    arXiv preprint arXiv:230306697

    Chen H, Wang J, Shao K, et al (2023) Traj-mae: Masked autoencoders for trajectory prediction. arXiv preprint arXiv:230306697

  4. [12]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 8679--8689

    Cheng J, Mei X, Liu M (2023) Forecast-mae: Self-supervised pre-training for motion forecasting with masked autoencoders. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 8679--8689

  5. [13]

    Transportation Research Part B: Methodological 105:362--377

    Coifman B, Li L (2017) A critical evaluation of the next generation simulation (ngsim) vehicle trajectory dataset. Transportation Research Part B: Methodological 105:362--377

  6. [14]

    ://arxiv.org/abs/1809.10732, 1809.10732

    Cui H, Radosavljevic V, Chou FC, et al (2019) Multimodal trajectory predictions for autonomous driving using deep convolutional networks. ://arxiv.org/abs/1809.10732, 1809.10732

  7. [15]

    In: Conference on Robot Learning, PMLR, pp 203--212

    Deo N, Wolff E, Beijbom O (2022) Multimodal trajectory prediction conditioned on lane-graph traversals. In: Conference on Robot Learning, PMLR, pp 203--212

  8. [16]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 9710--9719

    Ettinger S, Cheng S, Caine B, et al (2021) Large scale interactive motion forecasting for autonomous driving: The waymo open motion dataset. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 9710--9719

  9. [17]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 11525--11533

    Gao J, Sun C, Zhao H, et al (2020) Vectornet: Encoding hd maps and agent dynamics from vectorized representation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 11525--11533

  10. [18]

    The International Journal of Robotics Research 32(11):1231--1237

    Geiger A, Lenz P, Stiller C, et al (2013) Vision meets robotics: The kitti dataset. The International Journal of Robotics Research 32(11):1231--1237

  11. [19]

    arXiv preprint arXiv:211006607

    Gilles T, Sabatini S, Tsishkou D, et al (2021 a ) Thomas: Trajectory heatmap output with learned multi-agent sampling. arXiv preprint arXiv:211006607

  12. [20]

    In: 2021 IEEE International Intelligent Transportation Systems Conference (ITSC), IEEE, pp 500--507

    Gilles T, Sabatini S, Tsishkou D, et al (2021 b ) Home: Heatmap output for future motion estimation. In: 2021 IEEE International Intelligent Transportation Systems Conference (ITSC), IEEE, pp 500--507

  13. [21]

    In: 2022 international conference on robotics and automation (ICRA), IEEE, pp 9107--9114

    Gilles T, Sabatini S, Tsishkou D, et al (2022) Gohome: Graph-oriented heatmap output for future motion estimation. In: 2022 international conference on robotics and automation (ICRA), IEEE, pp 9107--9114

  14. [22]

    arXiv preprint arXiv:210400563

    Girgis R, Golemo F, Codevilla F, et al (2021) Latent variable sequential set transformers for joint multi-agent motion prediction. arXiv preprint arXiv:210400563

  15. [23]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 15303--15312

    Gu J, Sun C, Zhao H (2021) Densetnt: End-to-end trajectory prediction from dense goal sets. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 15303--15312

  16. [24]

    ://arxiv.org/abs/2208.01582, 2208.01582

    Gu J, Hu C, Zhang T, et al (2023) Vip3d: End-to-end visual trajectory prediction via 3d agent queries. ://arxiv.org/abs/2208.01582, 2208.01582

  17. [25]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 16000--16009

    He K, Chen X, Xie S, et al (2022) Masked autoencoders are scalable vision learners. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 16000--16009

  18. [26]

    In: Conference on Robot Learning, PMLR, pp 409--418

    Houston J, Zuidhof G, Bergamini L, et al (2021) One thousand and one hours: Self-driving motion prediction dataset. In: Conference on Robot Learning, PMLR, pp 409--418

  19. [27]

    ://arxiv.org/abs/2212.10156, 2212.10156

    Hu Y, Yang J, Chen L, et al (2023) Planning-oriented autonomous driving. ://arxiv.org/abs/2212.10156, 2212.10156

  20. [28]

    In: 2022 International Conference on Robotics and Automation (ICRA), IEEE, pp 2605--2611

    Huang Z, Mo X, Lv C (2022) Multi-modal motion prediction with transformer-based neural network for autonomous driving. In: 2022 International Conference on Robotics and Automation (ICRA), IEEE, pp 2605--2611

  21. [29]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 3903--3913

    Huang Z, Liu H, Lv C (2023) Gameformer: Game-theoretic modeling and learning of transformer-based interactive prediction and planning for autonomous driving. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 3903--3913

  22. [30]

    arXiv preprint arXiv:210710297

    Ivanovic B, Pavone M (2021) Rethinking trajectory forecasting evaluation. arXiv preprint arXiv:210710297

  23. [31]

    IEEE Robotics and Automation Letters 6(2):3065--3072

    Jia X, Sun L, Tomizuka M, et al (2021) Ide-net: Interactive driving event and pattern extraction from human data. IEEE Robotics and Automation Letters 6(2):3065--3072

  24. [32]

    In: Conference on Robot Learning, PMLR, pp 1434--1443

    Jia X, Sun L, Zhao H, et al (2022) Multi-agent trajectory prediction by combining egocentric and allocentric views. In: Conference on Robot Learning, PMLR, pp 1434--1443

  25. [33]

    IEEE transactions on pattern analysis and machine intelligence

    Jia X, Wu P, Chen L, et al (2023) Hdgt: Heterogeneous driving graph transformer for multi-agent trajectory prediction via scene encoding. IEEE transactions on pattern analysis and machine intelligence

  26. [34]

    arXiv preprint arXiv:240313331

    Jia X, Shi S, Chen Z, et al (2024) Amp: Autoregressive motion prediction revisited with next token prediction for autonomous driving. arXiv preprint arXiv:240313331

  27. [35]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 9644--9653

    Jiang C, Cornman A, Park C, et al (2023) Motiondiffuser: Controllable multi-agent motion prediction using diffusion. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 9644--9653

  28. [36]

    In: 2018 21st international conference on intelligent transportation systems (ITSC), IEEE, pp 2118--2125

    Krajewski R, Bock J, Kloeker L, et al (2018) The highd dataset: A drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems. In: 2018 21st international conference on intelligent transportation systems (ITSC), IEEE...

  29. [37]

    arXiv preprint arXiv:230915289

    Lan Z, Jiang Y, Mu Y, et al (2023) Sept: Towards efficient scene representation learning for motion prediction. arXiv preprint arXiv:230915289

  30. [38]

    arXiv preprint arXiv:230910121

    Li Y, Zhao SZ, Xu C, et al (2023) Pre-training on synthetic driving data for trajectory prediction. arXiv preprint arXiv:230910121

  31. [39]

    In: Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part II 16, Springer, pp 541--556

    Liang M, Yang B, Hu R, et al (2020 a ) Learning lane graph representations for motion forecasting. In: Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part II 16, Springer, pp 541--556

  32. [40]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 11553--11562

    Liang M, Yang B, Zeng W, et al (2020 b ) Pnpnet: End-to-end perception and prediction with tracking in the loop. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 11553--11562

  33. [41]

    Engineering 19:228--239

    Liu J, Luo Y, Zhong Z, et al (2022) A probabilistic architecture of long-term vehicle trajectory prediction for autonomous driving. Engineering 19:228--239

  34. [42]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 7577--7586

    Liu Y, Zhang J, Fang L, et al (2021) Multimodal motion prediction with stacked transformers. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 7577--7586

  35. [43]

    In: Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, pp 3569--3577

    Luo W, Yang B, Urtasun R (2018) Fast and furious: Real time end-to-end 3d detection, tracking and motion forecasting with a single convolutional net. In: Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, pp 3569--3577

  36. [44]

    In: Conference on Robot Learning, PMLR, pp 1457--1467

    Luo W, Park C, Cornman A, et al (2023) Jfp: Joint future prediction with interactive multi-agent modeling for autonomous driving. In: Conference on Robot Learning, PMLR, pp 1457--1467

  37. [45]

    In: 2008 11th International IEEE Conference on Intelligent Transportation Systems, IEEE, pp 803--808

    Lytrivis P, Thomaidis G, Amditis A (2008) Cooperative path prediction in vehicular environments. In: 2008 11th International IEEE Conference on Intelligent Transportation Systems, IEEE, pp 803--808

  38. [46]

    In: 2021 IEEE Intelligent Vehicles Symposium (IV), IEEE, pp 165--170

    Messaoud K, Deo N, Trivedi MM, et al (2021) Trajectory prediction for autonomous driving based on multi-head attention with joint agent-map representation. In: 2021 IEEE Intelligent Vehicles Symposium (IV), IEEE, pp 165--170

  39. [47]

    In: 2023 IEEE International Conference on Robotics and Automation (ICRA), IEEE, pp 2980--2987

    Nayakanti N, Al-Rfou R, Zhou A, et al (2023) Wayformer: Motion forecasting via simple & efficient attention networks. In: 2023 IEEE International Conference on Robotics and Automation (ICRA), IEEE, pp 2980--2987

  40. [48]

    arXiv preprint arXiv:210608417 2(7)

    Ngiam J, Caine B, Vasudevan V, et al (2021) Scene transformer: A unified multi-task model for behavior prediction and planning. arXiv preprint arXiv:210608417 2(7)

  41. [49]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 15065--15076

    Park D, Jeong J, Yoon SH, et al (2024) T4p: Test-time training of trajectory prediction via masked autoencoder and actor-specific token memory. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 15065--15076

  42. [50]

    ://arxiv.org/abs/2203.16297, 2203.16297

    Peri N, Luiten J, Li M, et al (2022) Forecasting from lidar via future object detection. ://arxiv.org/abs/2203.16297, 2203.16297

  43. [51]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 14074--14083

    Phan-Minh T, Grigore EC, Boulton FA, et al (2020) Covernet: Multimodal behavior prediction using trajectory sets. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 14074--14083

  44. [52]

    IEEE Transactions on Intelligent Transportation Systems 8(3):549--562

    Polychronopoulos A, Tsogas M, Amditis AJ, et al (2007) Sensor fusion for predicting vehicles' path for collision avoidance systems. IEEE Transactions on Intelligent Transportation Systems 8(3):549--562

  45. [53]

    Qingkai W, Manjiang H, Guotao X, et al (2020) Lightweight hd map construction for autonomous vehicles in non-paved roads. Tech. rep., SAE Technical Paper

  46. [54]

    In: Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part XVIII 16, Springer, pp 683--700

    Salzmann T, Ivanovic B, Chakravarty P, et al (2020) Trajectron++: Dynamically-feasible trajectory forecasting with heterogeneous data. In: Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part XVIII 16, Springer, pp 683--700

  47. [55]

    IEEE Robotics and Automation Letters 5(2):1696--1703

    Sch \"o ller C, Aravantinos V, Lay F, et al (2020) What the constant velocity model can teach us about pedestrian motion prediction. IEEE Robotics and Automation Letters 5(2):1696--1703

  48. [56]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 8579--8590

    Seff A, Cera B, Chen D, et al (2023) Motionlm: Multi-agent motion forecasting as language modeling. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 8579--8590

  49. [57]

    Advances in Neural Information Processing Systems 35:6531--6543

    Shi S, Jiang L, Dai D, et al (2022) Motion transformer with global intention localization and local movement refinement. Advances in Neural Information Processing Systems 35:6531--6543

  50. [58]

    IEEE Transactions on Pattern Analysis and Machine Intelligence

    Shi S, Jiang L, Dai D, et al (2024) Mtr++: Multi-agent motion prediction with symmetric scene modeling and guided intention querying. IEEE Transactions on Pattern Analysis and Machine Intelligence

  51. [59]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 6543--6552

    Sun Q, Huang X, Gu J, et al (2022) M2i: From factored marginal trajectory prediction to interactive prediction. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 6543--6552

  52. [60]

    ://arxiv.org/abs/2310.19620, 2310.19620

    Sun Q, Zhang S, Ma D, et al (2024) Large trajectory models are scalable motion predictors and planners. ://arxiv.org/abs/2310.19620, 2310.19620

  53. [61]

    In: Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI-22, Lud De Raedt, Ed, pp 5630--5637

    Teeti I, Khan S, Shahbaz A, et al (2022) Vision-based intention and trajectory prediction in autonomous vehicles: A survey. In: Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI-22, Lud De Raedt, Ed, pp 5630--5637

  54. [62]

    In: 2022 International Conference on Robotics and Automation (ICRA), IEEE, pp 7814--7821

    Varadarajan B, Hefny A, Srivastava A, et al (2022) Multipath++: Efficient information fusion and trajectory aggregation for behavior prediction. In: 2022 International Conference on Robotics and Automation (ICRA), IEEE, pp 7814--7821

  55. [63]

    In: ICRA2023 Workshop on Pretraining for Robotics (PT4R)

    Wagner R, Klemp M, Lopez CF, et al (2023) Road barlow twins: Redundancy reduction for motion prediction. In: ICRA2023 Workshop on Pretraining for Robotics (PT4R)

  56. [64]

    arXiv preprint arXiv:240719564

    Wang J, Messaoud K, Liu Y, et al (2024) Forecast-peft: Parameter-efficient fine-tuning for pre-trained motion forecasting models. arXiv preprint arXiv:240719564

  57. [65]

    arXiv preprint arXiv:230915685

    Wang S, Chen Y, Cheng J, et al (2023) Improving autonomous driving safety with pop: A framework for accurate partially observed trajectory predictions. arXiv preprint arXiv:230915685

  58. [66]

    In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 6563--6572, doi:10.1109/CVPR52688.2022.00646

    Weng X, Ivanovic B, Kitani K, et al (2022) Whose track is it anyway? improving robustness to tracking errors with affinity-based trajectory prediction. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 6563--6572, doi:10.1109/CVPR52688.2022.00646

  59. [67]

    arXiv preprint arXiv:230100493

    Wilson B, Qi W, Agarwal T, et al (2023) Argoverse 2: Next generation datasets for self-driving perception and forecasting. arXiv preprint arXiv:230100493

  60. [68]

    IEEE Transactions on Industrial Electronics 65(7):5999--6008

    Xie G, Gao H, Qian L, et al (2017) Vehicle trajectory prediction by integrating physics-and maneuver-based approaches using interactive multiple models. IEEE Transactions on Industrial Electronics 65(7):5999--6008

  61. [69]

    In: European Conference on Computer Vision, Springer, pp 34--50

    Xu C, Li T, Tang C, et al (2022) Pretram: Self-supervised pre-training via connecting trajectory and map. In: European Conference on Computer Vision, Springer, pp 34--50

  62. [70]

    Xu Y, Chambon L, Éloi Zablocki, et al (2024) Towards motion forecasting with real-world perception inputs: Are end-to-end approaches competitive? ://arxiv.org/abs/2306.09281, 2306.09281

  63. [71]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 9813--9823

    Yuan Y, Weng X, Ou Y, et al (2021) Agentformer: Agent-aware transformers for socio-temporal multi-agent forecasting. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 9813--9823

  64. [72]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 8660--8669

    Zeng W, Luo W, Suo S, et al (2019) End-to-end interpretable neural motion planner. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 8660--8669

  65. [73]

    In: 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, pp 532--539

    Zeng W, Liang M, Liao R, et al (2021) Lanercnn: Distributed representations for graph-centric motion forecasting. In: 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, pp 532--539

  66. [74]

    1910.03088

    Zhan W, Sun L, Wang D, et al (2019) Interaction dataset: An international, adversarial and cooperative motion dataset in interactive driving scenarios with semantic maps. 1910.03088

  67. [75]

    ://arxiv.org/abs/2202.01478, 2202.01478

    Zhang P, Bai L, Xue J, et al (2022) Trajectory forecasting from detection with uncertainty-aware motion encoding. ://arxiv.org/abs/2202.01478, 2202.01478

  68. [76]

    In: Conference on Robot Learning, PMLR, pp 895--904

    Zhao H, Gao J, Lan T, et al (2021) Tnt: Target-driven trajectory prediction. In: Conference on Robot Learning, PMLR, pp 895--904

  69. [77]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 15281--15290

    Zhou Y, Shao H, Wang L, et al (2024) Smartrefine: A scenario-adaptive refinement framework for efficient motion prediction. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 15281--15290

  70. [78]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 8823--8833

    Zhou Z, Ye L, Wang J, et al (2022) Hivt: Hierarchical vector transformer for multi-agent motion prediction. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 8823--8833

  71. [79]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 17863--17873

    Zhou Z, Wang J, Li YH, et al (2023 a ) Query-centric trajectory prediction. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 17863--17873

  72. [80]

    arXiv preprint arXiv:230610508

    Zhou Z, Wen Z, Wang J, et al (2023 b ) Qcnext: A next-generation framework for joint multi-agent trajectory prediction. arXiv preprint arXiv:230610508

  73. [81]

    , " * write output.state after.block = add.period write newline

    ENTRY address archive author booktitle chapter doi edition editor eid eprint howpublished institution journal key keywords month note number organization pages publisher school series title type url volume year archivePrefix primaryClass adsurl adsnote version label extra.labe...

  74. [82]

    write newline

    " write newline "" before.all 'output.state := FUNCTION add.period duplicate empty 'skip "." * add.blank if FUNCTION if.digit duplicate "0" = swap duplicate "1" = swap duplicate "2" = swap duplicate "3" = swap duplicate "4" = swap duplicate "5" = swap duplicate "6" = swap dupl...

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.