基于自然驾驶数据的交互轨迹基元表征与提取

doi:10.19562/j.chinasae.qcgc.2024.08.005

摘要/Abstract

摘要：

在共享道路空间中不同流向道路使用者间存在通行路径冲突，为规避碰撞风险，道路使用者须通过驾驶交互进行路权协商，从而消解潜在冲突。对交互行为的表述和建模，对于准确理解和预测动态环境具有重要意义。为此，本文提出一种以交互基元为分析单元的多车驾驶交互行为语义级表征和提取方法。首先，利用非参数贝叶斯方法对交互驾驶行为进行分割，得到具有显著行为模式的驾驶交互片段。然后，利用黏性层次狄利克雷-隐马尔可夫模型，从驾驶交互片段中提取得到交互基元。最后，对规范化处理后的交互基元进行无监督聚类，以获得驾驶交互场景的语义级行为特征。基于NGSIM高速公路数据集中20 797组多车交互数据的实证研究，本文提出的方法可提取并分析多个体参与的复杂交互场景，突破了已有研究中只针对两车交互场景构建交互基元的局限性，可支撑对多交通参与者交互的本质进行分析。实验结果表明，本文所提出的方法可以将连续的驾驶行为划分为离散的交互基元。且聚类划分结果可以与实际交互场景相对应，用于不同交互轨迹基元中车辆之间的交互行为特性分析。同时，该方法对于复杂场景下游驾驶任务具有提升作用。在车辆多步轨迹预测任务中，相比于基线方法，本文所提出的交互基元提取方法在与基线预测方法融合后可以将平均预测误差和终点预测误差分别降低19.3%和14.6%。

关键词: 交互行为, 轨迹基元, 非参数化贝叶斯方法, 无监督聚类

Abstract:

In shared road space， there are path conflicts between different road users moving in various directions. Road users must negotiate right-of-way through driving interactions to avoid collision risks， thus resolving potential conflicts. The description and modeling of interactive behaviors is crucial for accurately understanding and predicting the dynamic environment. Therefore， a semantic-level representation and extraction method for multi-vehicle interactive behaviors is proposed in this paper， taking interactive trajectory primitives as analysis units. Firstly， a nonparametric Bayesian method is utilized to segment interactive behaviors， obtaining interaction segments with significant behavior patterns. Then， the sticky hierarchical Dirichlet-Hidden Markov Model is employed to extract interaction primitives from these interaction segments. Finally， unsupervised clustering is applied to the normalized interaction primitives to obtain semantic-level behavioral features of interaction scenarios. An empirical study based on 20 797 pairs of multi-vehicle interaction data from the NGSIM highway dataset shows that the method proposed in this paper can extract and analyze complex interactive scenarios involving multiple participants， breaking through the limitation of existing research that only constructs interaction primitives for two vehicle interaction scenarios， and supporting the analysis of interaction among multiple traffic participants. The experimental results show that the proposed method can segment continuous driving behaviors into discrete interaction primitives. The clustering results correspond to actual interaction scenarios and can be used to characterize the interaction behaviors among vehicles in different interactive trajectory primitives. Furthermore， the method can enhance performance of downstream driving tasks in complex scenarios. In multi-step vehicle trajectory prediction， by integrating with baseline prediction methods， the proposed method can reduce the average prediction error and final position error by 19.3% and 14.6%， respectively， compared to baseline methods.

Key words: interactive behaviors, trajectory primitive, non-parametric Bayesian method, unsupervised clustering

李子睿,王浩闻,龚建伟,吕超,赵晓聪,王猛. 基于自然驾驶数据的交互轨迹基元表征与提取[J]. 汽车工程, 2024, 46(8): 1382-1393.

Zirui Li,Haowen Wang,Jianwei Gong,Lü Chao,Xiaocong Zhao,Meng Wang. Interactive Trajectory Primitives Representation and Extraction Based on Naturalistic Driving Data[J]. Automotive Engineering, 2024, 46(8): 1382-1393.

图/表 10

图1

图2

图3

图4

图5

图6

图7

图8

图9

表1

参考文献 33

1	WAGNER C. Asymptotic solutions for a multi-anticipative car-following model ［J］. Physica A： Statistical Mechanics its Applications， 2012， 260（1-2）： 218-224.
2	LI Z， GONG C， LU C， et al. Transferable driver behavior learning via distribution adaption in the lane change scenario [C]. 2019 IEEE Intelligent Vehicles Symposium (IV). IEEE， 2019: 193-200.
3	XING Y， LV C， WANG H， et al. An ensemble deep learning approach for driver lane change intention inference ［J］. Transportation Research Part C： Emerging Technologies， 2020， 115： 102615.
4	LU C， WANG H， LV C， et al. Learning driver-specific behavior for overtaking： a combined learning framework ［J］. IEEE Transactions on Vehicular Technology， 2018， 67（8）： 6788-6802.
5	GADEPALLY V， KRISHNAMURTHY A， OZGUNER U. A framework for estimating driver decisions near intersections ［J］. IEEE Transactions on Intelligent Transportation Systems， 2013， 15（2）： 637-646.
6	HAVLAK F， CAMPBELL M. Discrete and continuous， probabilistic anticipation for autonomous robots in urban environments ［J］. IEEE Transactions on Robotics， 2013， 30（2）： 461-474.
7	ZHANG W， WANG W. Learning V2V interactive driving patterns at signalized intersections ［J］. Transportation Research Part C： Emerging Technologies， 2019， 108： 151-166.
8	XING Y， LV C， WANG H， et al. Driver lane change intention inference for intelligent vehicles： framework， survey， and challenges ［J］. IEEE Transactions on Vehicular Technology， 2019， 68（5）： 4377-4390.
9	朱冰，蒋渊德，赵健，等. 基于深度强化学习的车辆跟驰控制［J］. 中国公路学报， 2019， 32（6）： 53-60.
	ZHU B， JIANG Y， ZHAO J， et al. A car-following control algorithm based on deep reinforcement learning［J］. China Journal of Highway and Transport， 2019， 32（6）： 53-60.
10	赵健，宋东鉴，朱冰，等. 基于自学习和监督学习混合驱动的智能汽车跟驰控制策略［J］. 中国公路学报， 2022， 35（3）： 55-65.
	ZHAO J， SONG D， ZHU B， et al. Intelligent vehicle-following control strategy based on self-learning and supervised-learning hybrid-driven framework［J］. China Journal of Highway and Transport， 2022， 35（3）： 55-65.
11	KOBER J, PETERS J. Policy search for motor primitives in robotics [C]. Advances in Neural Information Processing Systems, 2008: 849-856.
12	GOMEZ-GONZALEZ S， NEUMANN G， SCHöLKOPF B， et al. Adaptation and robust learning of probabilistic movement primitives ［J］. IEEE Transactions on Robotics， 2020， 36（2）： 366-379.
13	KOERT D， PAJARINEN J， SCHOTSCHNEIDER A， et al. Learning intention aware online adaptation of movement primitives ［J］. IEEE Robotics， 2019， 4（4）： 3719-3726.
14	LIOUTIKOV R， NEUMANN G， MAEDA G， et al. Learning movement primitive libraries through probabilistic segmentation ［J］. The International Journal of Robotics Research， 2017， 36（8）： 879-894.
15	WANG B， GONG J， CHEN H. Motion primitives representation， extraction and connection for automated vehicle motion planning applications ［J］. IEEE Transactions on Intelligent Transportation Systems， 2019， 21（9）： 3931-3945.
16	WANG W， ZHANG W， ZHU J， et al. Understanding V2V driving scenarios through traffic primitives ［J］. IEEE Transactions on Intelligent Transportation Systems， 2020， 23（1）： 610-619.
17	ZHANG C， ZHU J， WANG W， et al. Spatiotemporal learning of multivehicle interaction patterns in lane-change scenarios ［J］. IEEE Transactions on Intelligent Transportation Systems， 2021， 23（7）： 6446-6459.
18	LI Z， LU C， YI Y， et al. A hierarchical framework for interactive behaviour prediction of heterogeneous traffic participants based on graph neural network ［J］. IEEE Transactions on Intelligent Transportation Systems， 2021， 23（7）： 9102-9114.
19	田洪清，丁峰，郑讯佳，等. 基于势能场虚拟力的智能网联车辆运动规划［J］. 汽车工程， 2021， 43（4）： 518-526.
	TIAN H， DING F， ZHENG J， et al. Motion planning based on virtual force of potential field for intelligent connected vehicles［J］. Automotive Engineering， 2021， 43（4）： 518-526.
20	PUNZO V， BORZACCHIELLO M T， CIUFFO B. On the assessment of vehicle trajectory data accuracy and application to the Next Generation SIMulation （NGSIM） program data ［J］. Transportation Research Part C： Emerging Technologies， 2011， 19（6）： 1243-1262.
21	FOX E， SUDDERTH E， JORDAN M， et al. A sticky HDP-HMM with application to speaker diarization ［J］. The Annals of Applied Statistics， 2011： 1020-1056.
22	DEO N, TRIVEDI M M. Convolutional social pooling for vehicle trajectory prediction [C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops， 2018: 1468-1476.
23	KRAJEWSKI R, BOCK J, KLOEKER L, et al. The highD Dataset: a drone dataset of naturalistic vehicle trajectories on German highways for validation of highly automated driving systems [C]. 2018 21st International Conference on Intelligent Transportation Systems (ITSC)， 2018: 2118-2125.
24	ZHAN W， SUN L， WANG D， et al. INTERACTION Dataset： an INTERnational， adversarial and cooperative moTION dataset in interactive driving scenarios with semantic maps ［J］. arXiv preprint arXiv：， 2019.
25	ETTINGER S, CHENG S, CAINE B, et al. Large scale interactive motion forecasting for autonomous driving: the waymo open motion dataset [C]. Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 9710-9719.
26	CAESAR H, BANKITI V, LANG A H, et al. nuScenes: a multimodal dataset for autonomous driving [C]. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020: 11618-11628.
27	朱冰，孙宇航，赵健，等. 自动驾驶汽车测试场景基元自动提取方法［J］. 汽车工程， 2022， 44（11）： 1647-1655.
	ZHU B， SUN Y， ZHAO J， et al. Automatic extraction method for autonomous vehicle test scene primitives［J］. Automotive Engineering， 2022， 44（11）： 1647-1655.
28	FORNEY G D. The viterbi algorithm ［J］. Proceedings of the IEEE， 1973， 61（3）： 268-278.
29	TEH Y， JORDAN M， BEAL M， et al. Sharing clusters among related groups： hierarchical dirichlet processes ［J］. Advances in Neural Information Processing Systems， 2004， 17： 1385-1392.
30	WANG W， XI J， ZHAO D. Driving style analysis using primitive driving patterns with Bayesian nonparametric approaches ［J］. IEEE Transactions on Intelligent Transportation Systems， 2018， 20（8）： 2986-2998.
31	SIMONYAN K， ZISSERMAN A. Very deep convolutional networks for large-scale image recognition ［J］. arXiv preprint arXiv：， 2014.
32	HE K， ZHANG X， REN S， et al. Deep residual learning for image recognition [C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition， 2016: 770-778.
33	ALTCHé F， DE LA FORTELLE A. An LSTM network for highway trajectory prediction [C]. 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC). IEEE， 2017: 353-359.

方法	ADE/m	FDE/m
LSTM-1	3.31	3.80
LSTM-2	2.84	3.21
LSTM-3	2.29	2.74