摘要
复杂环境下行人轨迹短时预测在自动驾驶、社交机器人控制、智能安防及智慧城市等领域有着广泛用途.行人与行人、行人与环境之间的交互具有多尺度复杂性和不确定性,具有挑战.现有深度学习模型虽然有助于挖掘行人的复杂交互关系,但都假设行人轨迹在不同场景遵循相同运动模式,未考虑场景间存在的潜在分布差异;域适应模型虽然考虑了这一点,但仍未考虑行人间和行人环境间的多层次特性.为了解决上述问题,本文提出了一种基于分级包络域适应的行人轨迹预测模型.通过构造局部层次行人邻接关系设计局部层次包络样本,通过个体层次行人关系设计个体层次包络样本,并将两者融合形成双级包络样本.基于双级包络样本构造模块,求得行人轨迹的时空特征分布,从而构造全局层次包络样本.基于注意力机制和跨域分布对齐,分别设计了局部层次包络域适应模块和全局层次包络域适应模块,构建加权预测损失函数将两者融合一体,并联合优化.实验部分选取了2个有代表性的公共数据集,并与5个相关代表性算法模型进行对比.通过消融实验、参数分析、方法对比和轨迹可视化等来进行综合验证.在ETH和UCY的实验结果表明,相比于T-GNN,本文方法的平均位移误差降低了22.7%,终点位移误差降低了19.8%.文章完整版参见链接:https://github.com/LWZ9910/MESC-HEDA.git.
In complex environments,short-term pedestrian trajectory prediction finds extensive applications in auton⁃omous driving,social robotics,intelligent security,and smart city infrastructures.Interactions among pedestrians and be⁃tween pedestrians and their environment exhibit multi-scale complexities and uncertainties,posing substantial challenges.Although current deep learning models are effective in uncovering complex pedestrian interactions,they typically assume uniform motion patterns across various scenes,thereby neglecting potential distributional discrepancies.While domain adap⁃tation models partially address this issue,they often overlook the multi-level characteristics of pedestrian interactions and environmental influences.To address these challenges,this study proposes a pedestrian trajectory prediction model founded on hierarchical envelope domain adaptation.We design a local-level envelope sample construction module by establishing local-level pedestrian adjacency relationships.An individual-level envelope sample construction module is devised based on individual pedestrian relationships.These two modules are subsequently integrated to form a bi-level envelope sample con⁃struction module.Leveraging the bi-level envelope sample construction module,we compute the spatio-temporal feature dis⁃tribution of all pedestrian trajectories to construct global-level envelope samples.Employing the attention mechanism and cross-domain distribution alignment,we respectively design the local-level envelope domain adaptation and global-level en⁃velope domain adaptation modules.These modules are then integrated into a unified framework using a weighted prediction loss function,which is jointly optimized.The experimental section utilizes two representative public datasets and compares them with five representative algorithm models.Comprehensive validation is conducted through ablation studies,parameter analysis,method comparison,and trajectory visualization.The experimental results in the ETH and UCY datasets show that compared with T-GNN,the average displacement error is reduced by 22.7%and the final displacement error is reduced by 19.8%.For the full version of the article,please refer to the link:https://github.com/LWZ9910/MESC-HEDA.git.
作者
李勇明
李文正
张小恒
王品
胡杰
LI Yong-ming;LI Wen-zheng;ZHANG Xiao-heng;WANG Pin;HU Jie(School of Microelectronics and Communication Engineering,Chongqing University,Chongqing 400044,China)
出处
《电子学报》
北大核心
2025年第4期1308-1321,共14页
Acta Electronica Sinica
基金
国家自然科学基金(No.U21A20448,No.61771080)。
关键词
行人轨迹预测
分级包络
域适应
多层次
包络样本
pedestrian trajectory prediction
hierarchical envelopes
domain adaptation
multilevel
envelope samples
