pith. sign in

arxiv: 2512.03795 · v2 · submitted 2025-12-03 · 💻 cs.RO · cs.AI

MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving

Jia Hu , Zhexi Lian , Xuerun Yan , Ruiang Bi , Dou Shen , Yu Ruan , Chunlong Xia , Haoran Wang This is my paper

Pith reviewed 2026-05-17 02:33 UTC · model grok-4.3

classification 💻 cs.RO cs.AI
keywords autonomous drivingsocial interactionmodel predictive controlphysics-informedTransformertrajectory predictionlane changing
0
0 comments X

The pith

MPCFormer models multi-vehicle social interaction dynamics using physics priors and data to enable human-like autonomous driving.

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

Autonomous vehicles struggle to act naturally around other cars in busy traffic because they lack understanding of how vehicles influence each other socially. MPCFormer tackles this by putting social interaction dynamics into a discrete state-space form that includes physics rules for clearer explanations. The model learns the specific coefficients from real driving recordings with a Transformer encoder-decoder setup. This learned model then feeds into a model predictive control planner that produces varied and safe responses resembling human drivers. Tests on highway data and simulated intense maneuvers show stronger prediction accuracy and better overall planning outcomes than prior methods.

Core claim

MPCFormer formulates the dynamics of multi-vehicle social interactions into a discrete state-space representation that embeds physics priors to improve explainability, learns the dynamics coefficients from naturalistic driving data via a Transformer-based encoder-decoder architecture, and integrates the result into an MPC planner to generate manifold human-like behaviors while mitigating safety risks of purely learning-based approaches.

What carries the argument

Discrete state-space representation of coupled social interaction dynamics with embedded physics priors, whose coefficients are learned from data using a Transformer encoder-decoder architecture.

Load-bearing premise

The physics priors embedded in the discrete state-space representation correctly capture the underlying mechanisms of social interaction and coefficients learned from NGSIM data generalize to closed-loop interactive scenarios.

What would settle it

A closed-loop test in scenarios requiring consecutive lane changes to exit an off-ramp in which the planning success rate falls well below 94.67 percent or the collision rate exceeds 0.5 percent would falsify the performance claims.

read the original abstract

Autonomous Driving (AD) vehicles still struggle to exhibit human-like behavior in highly dynamic and interactive traffic scenarios. The key challenge lies in AD's limited ability to interact with surrounding vehicles, largely due to a lack of understanding the underlying mechanisms of social interaction. To address this issue, we introduce MPCFormer, an explainable socially-aware autonomous driving approach with physics-informed and data-driven coupled social interaction dynamics. In this model, the dynamics are formulated into a discrete space-state representation, which embeds physics priors to enhance modeling explainability. The dynamics coefficients are learned from naturalistic driving data via a Transformer-based encoder-decoder architecture. To the best of our knowledge, MPCFormer is the first approach to explicitly model the dynamics of multi-vehicle social interactions. The learned social interaction dynamics enable the planner to generate manifold, human-like behaviors when interacting with surrounding traffic. By leveraging the MPC framework, the approach mitigates the potential safety risks typically associated with purely learning-based methods. Open-looped evaluation on NGSIM dataset demonstrates that MPCFormer achieves superior social interaction awareness, yielding the lowest trajectory prediction errors compared with other state-of-the-art approaches. The prediction achieves an ADE as low as 0.86 m over a long prediction horizon of 5 seconds. Close-looped experiments in highly intense interaction scenarios, where consecutive lane changes are required to exit an off-ramp, further validate the effectiveness of MPCFormer. Results show that MPCFormer achieves the highest planning success rate of 94.67%, improves driving efficiency by 15.75%, and reduces the collision rate from 21.25% to 0.5%, outperforming a frontier Reinforcement Learning (RL) based planner.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper proposes MPCFormer, a hybrid physics-informed data-driven framework for socially-aware autonomous driving. It formulates multi-vehicle social interaction dynamics as a discrete state-space model with embedded physics priors, learns the dynamics coefficients from NGSIM naturalistic trajectories using a Transformer encoder-decoder, and incorporates the resulting model into a model predictive control (MPC) planner. The central claims are that this is the first explicit modeling of multi-vehicle social interaction dynamics, that the learned dynamics enable manifold human-like behaviors, and that the approach yields superior open-loop prediction (ADE 0.86 m over 5 s) on NGSIM plus strong closed-loop performance (94.67 % success rate, collision rate reduced to 0.5 %, 15.75 % efficiency gain) in intense off-ramp lane-change scenarios compared with a frontier RL planner.

Significance. If the generalization and modeling claims hold, the work would be significant as a concrete example of coupling interpretable physics priors with data-driven learning inside a safety-critical MPC loop, potentially improving both explainability and robustness over purely learning-based or purely kinematic planners. The reported numerical gains in prediction error and closed-loop safety metrics would be noteworthy for the socially-aware AD literature. However, the significance is currently limited by the absence of supporting details that would allow verification of whether the learned terms capture transferable social mechanisms rather than open-loop kinematic correlations.

major comments (3)
  1. [Abstract / closed-loop experiments] Abstract and closed-loop experiments description: the claim that the learned social interaction dynamics enable 'manifold, human-like behaviors' in closed-loop intense interactions rests on the unverified assumption that coefficients fitted to NGSIM open-loop trajectories generalize to consecutive forced lane-change scenarios. No distribution-shift analysis, out-of-distribution testing, or comparison of interaction terms versus pure kinematic baselines is provided, which is load-bearing for the central claim of explicit social-mechanism modeling.
  2. [Abstract / methodology] Abstract and presumed methodology section: the discrete state-space representation and embedded physics priors are described only at a high level with no equations, state definitions, or prior terms shown. Without these, it is impossible to assess whether the 'physics-informed' component is substantive or merely nominal, undermining the explainability and novelty assertions.
  3. [Experiments] Experiments: the manuscript reports strong numerical results (ADE 0.86 m, 94.67 % success, collision drop to 0.5 %) but supplies no training details, baseline implementations, ablation studies, or hyperparameter settings. This prevents reproduction and makes it impossible to isolate whether gains derive from the proposed social-dynamics modeling or from other implementation choices.
minor comments (2)
  1. [Abstract] Abstract contains minor grammatical issues: 'Open-looped' and 'Close-looped' should read 'Open-loop' and 'Closed-loop'.
  2. [Abstract] The efficiency improvement of 15.75 % and collision-rate reduction should explicitly state the baseline method and metric definition for each quantity.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback, which highlights important aspects for strengthening the manuscript's clarity, reproducibility, and claims regarding generalization. We have revised the paper to incorporate additional mathematical details, experimental settings, and analysis as suggested. Our point-by-point responses follow.

read point-by-point responses

  1. Referee: [Abstract / closed-loop experiments] Abstract and closed-loop experiments description: the claim that the learned social interaction dynamics enable 'manifold, human-like behaviors' in closed-loop intense interactions rests on the unverified assumption that coefficients fitted to NGSIM open-loop trajectories generalize to consecutive forced lane-change scenarios. No distribution-shift analysis, out-of-distribution testing, or comparison of interaction terms versus pure kinematic baselines is provided, which is load-bearing for the central claim of explicit social-mechanism modeling.

    Authors: We agree that explicit verification of generalization is important for substantiating the social-mechanism modeling claim. The closed-loop results in off-ramp scenarios demonstrate practical effectiveness, but we acknowledge the value of additional analysis. In the revised manuscript, we have added a dedicated subsection with qualitative inspection of learned interaction coefficients during closed-loop execution, a direct comparison against a pure kinematic baseline (e.g., constant-velocity model without social terms), and an out-of-distribution test using synthetic forced lane-change trajectories. This addresses the distribution-shift concern while preserving the original empirical results. revision: yes

  2. Referee: [Abstract / methodology] Abstract and presumed methodology section: the discrete state-space representation and embedded physics priors are described only at a high level with no equations, state definitions, or prior terms shown. Without these, it is impossible to assess whether the 'physics-informed' component is substantive or merely nominal, undermining the explainability and novelty assertions.

    Authors: We appreciate this observation on presentation. The full manuscript (Section III) defines the discrete state-space model with states for relative positions, velocities, and accelerations, along with embedded priors such as repulsive potentials for collision avoidance and attractive terms for lane alignment. To make this accessible without requiring the reader to locate the equations, we have expanded the abstract with the core state-transition equation and added an illustrative figure plus explicit prior term definitions in the methodology section of the revision. revision: yes

  3. Referee: [Experiments] Experiments: the manuscript reports strong numerical results (ADE 0.86 m, 94.67 % success, collision drop to 0.5 %) but supplies no training details, baseline implementations, ablation studies, or hyperparameter settings. This prevents reproduction and makes it impossible to isolate whether gains derive from the proposed social-dynamics modeling or from other implementation choices.

    Authors: We agree that missing implementation details limit reproducibility and attribution of gains. The revised manuscript includes a new appendix with full training hyperparameters for the Transformer encoder-decoder, NGSIM preprocessing steps, optimizer settings, and baseline RL planner implementation details. We have also added ablation studies that isolate the contribution of the physics priors and the learned social coefficients separately, allowing readers to verify that performance improvements stem from the proposed modeling approach. revision: yes

Circularity Check

0 steps flagged

No significant circularity in MPCFormer derivation chain

full rationale

The paper defines social interaction dynamics as a discrete state-space model embedding physics priors, learns the coefficients directly from external NGSIM naturalistic trajectories via a Transformer encoder-decoder, and deploys the resulting model inside a standard MPC planner. Open-loop ADE evaluation on NGSIM and closed-loop success/collision metrics in off-ramp scenarios are reported against external baselines; neither the central claim of explicit multi-vehicle modeling nor the performance numbers reduce by construction to the fitted coefficients or to any self-citation. The derivation remains self-contained against the cited external dataset and conventional MPC framework.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that social vehicle interactions admit a discrete state-space representation that can embed useful physics priors and that coefficients learned from NGSIM data transfer to novel interactive planning tasks.

axioms (1)
  • domain assumption Social interaction dynamics can be usefully represented in a discrete state-space form that embeds physics priors for explainability.
    Stated in the abstract as the basis for the coupled physics-informed data-driven model.

pith-pipeline@v0.9.0 · 6297 in / 1280 out tokens · 159520 ms · 2026-05-17T02:33:03.440854+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

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

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

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

  1. DINO-VO: Learning Where to Focus for Enhanced State Estimation

    cs.CV 2026-04 unverdicted novelty 6.0

    DINO-VO achieves state-of-the-art monocular visual odometry accuracy and generalization by training a differentiable patch selector together with multi-task features and inverse-depth bundle adjustment.

Reference graph

Works this paper leans on

44 extracted references · 44 canonical work pages · cited by 1 Pith paper · 1 internal anchor

  1. [1]

    Review on the impacts of cooperative automated driving on transportation and environment,

    Z. Zhang, H. Liu, M. Lei, X. Yan, M. Wang, and J. Hu, "Review on the impacts of cooperative automated driving on transportation and environment," Transportation Research Part D: Transport and Environment, vol. 115, p. 103607, 2023

    work page 2023

  2. [2]

    Motion planning for autonomous driving: Th e state of the art and future perspectives,

    S. Teng et al., "Motion planning for autonomous driving: Th e state of the art and future perspectives," IEEE Transactions on Intelligent Vehicles, vol. 8, no. 6, pp. 3692-3711, 2023

    work page 2023

  3. [3]

    Recent advances in reinforcement learning-based autonomous driving behavior planning: A survey,

    J. Wu, C. Huang, H. Huang, C. Lv, Y. Wang, and F.-Y. Wang, "Recent advances in reinforcement learning-based autonomous driving behavior planning: A survey," Transportation Research Part C: Emerging Technologies, vol. 164, p. 104654, 2024

    work page 2024

  4. [4]

    Dense reinforcement learning for safety validation of autonomous vehicles,

    S. Feng et al., "Dense reinforcement learning for safety validation of autonomous vehicles," Nature, vol. 615, no. 7953, pp. 620-627, 2023

    work page 2023

  5. [5]

    Social interactions for autonomous driving: A review and perspectives,

    W. Wang, L. Wang, C. Zhang, C. Liu, and L. Sun, "Social interactions for autonomous driving: A review and perspectives," Foundations and Trends® in Robotics, vol. 10, no. 3-4, pp. 198-376, 2022

    work page 2022

  6. [6]

    Differentiable Integrated Motion Prediction and Planning With Learnable Cost Function for Autonomous Driving,

    Z. Huang, H. Liu, J. Wu, and C. Lv, "Differentiable Integrated Motion Prediction and Planning With Learnable Cost Function for Autonomous Driving," IEEE Transactions on Neural Networks and Learning Systems, pp. 1-15, 2023, doi: 10.1109/TNNLS.2023.3283542

    work page doi:10.1109/tnnls.2023.3283542 2023

  7. [7]

    Predictive risk estimation for intelligent ADAS functions,

    J. Eggert, "Predictive risk estimation for intelligent ADAS functions, " in 17th International IEEE Conference on Intelligent Transportation Systems (ITSC), 8-11 Oct. 2014 2014, pp. 711 -718, doi: 10.1109/ITSC.2014.6957773

    work page doi:10.1109/itsc.2014.6957773 2014

  8. [8]

    Online trajectory replanning for sudden environmental changes during automated parking: A parallel stitching method,

    B. Li, Z. Yin, Y. Ouyang, Y. Zhang, X. Zhong, and S. Tang, "Online trajectory replanning for sudden environmental changes during automated parking: A parallel stitching method," IEEE Transactions on Intelligent Vehicles, vol. 7, no. 3, pp. 748-757, 2022

    work page 2022

  9. [9]

    Human -like motion planning of autonomous vehicle based on probabilistic trajectory prediction,

    P. Li, X. Pei, Z. Chen, X. Zhou, and J. Xu, "Human -like motion planning of autonomous vehicle based on probabilistic trajectory prediction," Applied Soft Computing, vol. 118, p. 108499, 2022

    work page 2022

  10. [10]

    Optimal traject ory generation for dynamic street scenarios in a Frené t Frame,

    M. Werling, J. Ziegler, S. Kammel, and S. Thrun, "Optimal traject ory generation for dynamic street scenarios in a Frené t Frame," in 2010 IEEE International Conference on Robotics and Automation , 3-7 May 2010 2010, pp. 987-993, doi: 10.1109/ROBOT.2010.5509799

    work page doi:10.1109/robot.2010.5509799 2010

  11. [11]

    Saf e POMDP Online Planning Among Dynamic Agents via Adaptive Conformal Prediction,

    S. Sheng, P. Yu, D. Parker, M. Kwiatkowska, and L. Feng, "Saf e POMDP Online Planning Among Dynamic Agents via Adaptive Conformal Prediction," IEEE Robotics and Automation Letters, vol. 9, no. 11, pp. 9946-9953, 2024, doi: 10.1109/LRA.2024.3468092

    work page doi:10.1109/lra.2024.3468092 2024

  12. [12]

    A Belief State Planner for Interactive Merge Maneuvers in Congested Traffic,

    C. Hubmann, J. Schulz, G. Xu, D. Althoff, and C. Stiller, "A Belief State Planner for Interactive Merge Maneuvers in Congested Traffic," in 2018 21st International Conference on Intelligent Transportation Systems (ITSC), 4-7 Nov. 2018 2018, pp. 1617 -1624, doi: 10.1109/ITSC.2018.8569729

    work page doi:10.1109/itsc.2018.8569729 2018

  13. [13]

    Interaction-aware motion planning for autonomous vehicles with multi -modal obstacle uncertainty predictions,

    J. Zhou, B. Olofsson, and E. Frisk, "Interaction-aware motion planning for autonomous vehicles with multi -modal obstacle uncertainty predictions," IEEE Transactions on Intelligent Vehicles, 2023

    work page 2023

  14. [14]

    Game Theoretic Merging Behavior Control for Autonomous Vehicle at Highway On -Ramp,

    C. Wei, Y. He, H. Tian, and Y. Lv, "Game Theoretic Merging Behavior Control for Autonomous Vehicl e at Highway On -Ramp," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 11, pp. 21127-21136, 2022, doi: 10.1109/TITS.2022.3174659

    work page doi:10.1109/tits.2022.3174659 2022

  15. [15]

    Social Interaction ‐Aware Dynamica l Models and Decision ‐Making for Autonomous Vehicles,

    L. Crosato, K. Tian, H. P. H. Shum, E. S. L. Ho, Y. Wang, and C. Wei, "Social Interaction ‐Aware Dynamica l Models and Decision ‐Making for Autonomous Vehicles," Advanced Intelligent Systems, vol. 6, no. 3, 2023, doi: 10.1002/aisy.202300575

    work page doi:10.1002/aisy.202300575 2023

  16. [16]

    A survey on imitation learning techniques for end -to-end autonomous ve hicles,

    L. Le Mero, D. Yi, M. Dianati, and A. Mouzakitis, "A survey on imitation learning techniques for end -to-end autonomous ve hicles," IEEE Transactions on Intelligent Transportation Systems , vol. 23, no. 9, pp. 14128-14147, 2022

    work page 2022

  17. [17]

    TransFuser: Imitation With Transformer -Based Sensor Fusion for Autonomous Driving,

    K. Chitta, A. Prakash, B. Jaeger, Z. Yu, K. Renz, and A. Geiger, "TransFuser: Imitation With Transformer -Based Sensor Fusion for Autonomous Driving," IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 11, pp. 12878 -12895, 2023, doi: 10.1109/TPAMI.2022.3200245

    work page doi:10.1109/tpami.2022.3200245 2023

  18. [18]

    Hierarchical model -based imitation learning for planning in autonomous driving,

    E. Bronstein et al., "Hierarchical model -based imitation learning for planning in autonomous driving," in 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022: IEEE, pp. 8652-8659

    work page 2022

  19. [20]

    Automated Driving with Evolution Capability: A Reinforcement Learning Method with Monotonic Performance Enhancement,

    J. Hu, X. Yan, T. Xu, and H. Wang, "Automated Driving with Evolution Capability: A Reinforcement Learning Method with Monotonic Performance Enhancement," arXiv preprint arXiv:2412.10822, 2024

    work page arXiv 2024

  20. [21]

    Safe, efficient, and comfortable veloc ity control based on reinforcement learning for autonomous driving,

    M. Zhu, Y. Wang, Z. Pu, J. Hu, X. Wan g, and R. Ke, "Safe, efficient, and comfortable velocity control based on reinforcement learning for autonomous driving," Transportation Research Part C: Emerging Technologies, vol. 117, p. 102662, 2020/08/01/ 2020, doi: https://doi.org/10.1016/j.trc.2020.102662

    work page doi:10.1016/j.trc.2020.102662 2020

  21. [22]

    A reduction of imitation learning and structured prediction to no -regret online learning,

    S. Ross, G. Gordon, and D. Bagnell, "A reduction of imitation learning and structured prediction to no -regret online learning," in Proceedings of the fourteenth international conference on artificial intelligence and statistics, 2011: JMLR Workshop and Conference Proceedings , pp. 627-635

    work page 2011

  22. [23]

    Imitation learning decision with driving style tuning for personalized autonomous driving,

    R. Hu et al., "Imitation learning decision with driving style tuning for personalized autonomous driving," in International Conference on Database Systems for Advanced Applications, 2024: Springer, pp. 220- 231

    work page 2024

  23. [24]

    No more road bullying: An integrated behavioral and motion planner with proactive right-of-way acquisition capability,

    Z. Zhang, X. Yan, H. Wang, C. Ding, L. Xiong, and J. Hu, "No more road bullying: An integrated behavioral and motion planner with proactive right -of-way acquisition capability," Transportation Research Part C: Emerging Technologies, vol. 156, p. 104363, 2023/11/01/ 2023, doi: https://doi.org/10.1016/j.trc.2023.104363

    work page doi:10.1016/j.trc.2023.104363 2023

  24. [25]

    Proactive emergency collision avoidance for automated driving in highway scenarios,

    L. Gharavi, A. Dabiri, J. Verkuijlen, B. De Schutter, and S. Baldi , "Proactive emergency collision avoidance for automated driving in highway scenarios," IEEE Transactions on Control Systems Technology, 2024

    work page 2024

  25. [26]

    Anti-bullying Adaptive Cruise Control: A proactive right-of-way protection approach

    J. Hu et al., "Anti-bullying Adaptive Cruise Control: A proactive right- of-way protection approach," arXiv preprint arXiv:2412.12197, 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  26. [27]

    Toward Proactive -Aware Autonomous Driving: A Reinforcement Learning Approach Utilizing Expert Priors During Unprotected Turns,

    J. Fan, Y. Ni, D. Zhao, P. Hang, and J. Sun, "Toward Proactive -Aware Autonomous Driving: A Reinforcement Learning Approach Utilizing Expert Priors During Unprotected Turns," IEEE Transactions on Intelligent Transportation Systems, 2024

    work page 2024

  27. [28]

    Legible and proactive robot planning for prosocial human-robot interactions,

    J. Geldenbott and K. Leung, "Legible and proactive robot planning for prosocial human-robot interactions," arXiv preprint arXiv:2404.03734, 2024

    work page arXiv 2024

  28. [29]

    Lateral Vehicle Dynamics,

    R. Rajamani, "Lateral Vehicle Dynamics," in Vehicle Dynamics and Control, R. Rajamani Ed. Boston, MA: Springer US, 2012, pp. 15-46

    work page 2012

  29. [30]

    Attention is all you need,

    A. Vaswani, "Attention is all you need," Advances in Neural Information Processing Systems, 2017

    work page 2017

  30. [31]

    Stackelberg Competition with Endogenous Entry,

    F. Etro, "Stackelberg Competition with Endogenous Entry," The Economic Journal , vol. 118, no. 532, pp. 1670 -1697, 2008, doi: 10.1111/j.1468-0297.2008.02185.x

    work page doi:10.1111/j.1468-0297.2008.02185.x 2008

  31. [32]

    The Big -M method with the numerical infinite M,

    M. Cococcioni and L. Fiaschi, "The Big -M method with the numerical infinite M," Optimization Letters, vol. 15, no. 7, pp. 2455-2468, 2021

    work page 2021

  32. [33]

    A critical evaluation of the Next Generation Simulation (NGSIM) vehicle trajectory dataset,

    B. Coifman and L. Li, "A critical evaluation of the Next Generation Simulation (NGSIM) vehicle trajectory dataset," Transportation Research Part B: Methodological, vol. 105, pp. 362 -377, 2017/11/01/ 2017, doi: https://doi.org/10.1016/j.trb.2017.09.018

    work page doi:10.1016/j.trb.2017.09.018 2017

  33. [34]

    Social gan: Socially acceptable trajectories with generative adversarial networks,

    A. Gupta, J. Johnson, L. Fei-Fei, S. Savarese, and A. Alahi, "Social gan: Socially acceptable trajectories with generative adversarial networks," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 2255-2264

    work page 2018

  34. [35]

    Social-stgcnn: A social spatio-temporal graph convolutional neural network for human trajectory prediction,

    A. Mohamed, K. Qian, M. Elhoseiny, and C. Claudel, "Social-stgcnn: A social spatio-temporal graph convolutional neural network for human trajectory prediction," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 14424-14432

    work page 2020

  35. [36]

    Vehicle Trajectory Prediction Based on Spatial -aware Transformer,

    T. Yu, F. Hui, X. Shuoshuo, W. Mengyuan, and X. Lingyang, "Vehicle Trajectory Prediction Based on Spatial -aware Transformer," in 2023 IEEE 11th International Conference on Information, Communication and Networks (ICICN), 2023: IEEE, pp. 884-889

    work page 2023

  36. [37]

    Hierarchical motion encoder-decoder network for trajectory forecasting,

    Q. Xue, S. Li, X. Li, J. Zhao, and W. Zhang, "Hierarchical motion encoder-decoder network for trajectory forecasting," arXiv preprint arXiv:2111.13324, 2021

    work page arXiv 2021

  37. [38]

    It is not the journey but the destination: Endpoint conditioned trajectory prediction,

    K. Mangalam et al., "It is not the journey but the destination: Endpoint conditioned trajectory prediction," in Computer Vision –ECCV 2020: 16th European Conference, Glasgow, UK, August 23 –28, 2020, Proceedings, Part II 16, 2020: Springer, pp. 759-776

    work page 2020

  38. [39]

    C2f-tp: A coarse-to-fine denoising framework for uncertainty -aware trajectory prediction,

    Z. Wang, H. Miao, S. Wang, R. Wang, J. Wang, and J. Zhang, "C2f-tp: A coarse-to-fine denoising framework for uncertainty -aware trajectory prediction," in Proceedings of the AAAI Conference on Artificial Intelligence, 2025, vol. 39, no. 12, pp. 12810-12817

    work page 2025

  39. [40]

    Interaction -Aware Trajectory Prediction Method Based on Sparse Spatial -Temporal Transformer for Internet of Vehicles,

    X. Li, J. Zhang, J. Chen, and P. Qian, "Interaction -Aware Trajectory Prediction Method Based on Sparse Spatial -Temporal Transformer for Internet of Vehicles," IEEE Transactions on Intelligent Transportation Systems, 2025

    work page 2025

  40. [41]

    A Simulation Platform for Truck Platooning Evaluation in an Interactive Traffic Environment,

    J. Hu et al., "A Simulation Platform for Truck Platooning Evaluation in an Interactive Traffic Environment," IEEE Transactions on Intelligent Transportation Systems, 2024

    work page 2024

  41. [42]

    Driver intent inference at urban intersections using the intelligent driver model,

    M. Liebner, M. Baumann, F. Klanner, and C. Stiller, "Driver intent inference at urban intersections using the intellig ent driver model," in 2012 IEEE Intelligent Vehicles Symposium , 3-7 June 2012 2012, pp. 1162-1167, doi: 10.1109/IVS.2012.6232131

    work page doi:10.1109/ivs.2012.6232131 2012

  42. [43]

    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, 2007/01/01 2007, doi: 10.3141/1999-10

    work page doi:10.3141/1999-10 1999

  43. [44]

    RSS-Net: Weakly-Supervised Multi-Class Seman- tic Segmentation with FMCW Radar

    F. Ye, X. Cheng, P. Wang, C. Y. Chan, and J. Zhang, "Automated Lane Change Strategy using Proximal Policy Optimization -based Deep Reinforcement Learning," in 2020 IEEE Intelligent Vehicles Symposium (IV) , 19 Oct. -13 Nov. 2020 2020, pp. 1746 -1752, doi: 10.1109/IV47402.2020.9304668

    work page doi:10.1109/iv47402.2020.9304668 2020

  44. [45]

    FFTW: An adaptive software architecture for the FFT,

    M. Frigo and S. G. Johnson, "FFTW: An adaptive software architecture for the FFT," in Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing , ICASSP'98 (Cat. No. 98CH36181), 1998, vol. 3: IEEE, pp. 1381-1384. Jia Hu (Senior Member, IEEE) is currently working as a Zhongte Distinguished Chair of Cooperative Automat...

    work page 1998