TSCA-Net achieves state-of-the-art performance, with average ADE/FDE of 0.13/0.20 m on ETH/UCY and 6.95/10.43 pixels on SDD.
Abstract
Accurate pedestrian trajectory prediction in crowded environments remains challenging due to the multimodal uncertainty of human motion and the variable complexity of motion dynamics across different scene contexts. Existing goal-conditioned models rely on static displacement structures that assign equal weight to all historical time steps, standard graph attention mechanisms, and fixed-capacity motion decoders that cannot adapt to local prediction complexity. To address these limitations, we propose TSCA-Net, a trajectory prediction framework built upon three complementary modules. The Temporal-Spatial Clique Attention (TSCA) module introduces learnable temporal gating into clique-based goal-history interaction, enabling time-aware modulation of historical observations relative to each candidate goal. The Cross-Pedestrian Clique Potential (CPCP) module models asymmetric pairwise agent relationships through a dynamic clique potential framework with a time-varying social graph. The Adaptive KAN Grid Refinement (AKGR) mechanism dynamically adjusts the B-spline grid resolution of a Kolmogorov-Arnold Network-augmented LSTM decoder based on per-agent goal distribution entropy, balancing model expressiveness against overfitting across varying motion complexities. Extensive experiments on the ETH/UCY and Stanford Drone Dataset benchmarks demonstrate that TSCA-Net achieves state-of-the-art performance, with average ADE/FDE of 0.13/0.20 m on ETH/UCY and 6.95/10.43 pixels on SDD. Comprehensive ablation studies confirm the complementary contributions of all three proposed modules.
Forecasting pedestrian motion has always been fundamental for autonomous navigation in crowded environments. While attention-based methods achieve strong performance, they suffer from quadratic computational complexity in modeling social interactions, limiting scalability. Additionally, the existing methods often achieve high accuracy on prediction benchmarks at the individual level, but fail to fully capture the natural movement behaviors of crowds in real-world scenarios, particularly group structures. In this study, we propose Social Graph Mamba (SGM), a novel architecture that replaces attention-based social reasoning with Selective State Space Models (SSMs) operating on dynamically constructed interaction graphs. SGM introduces a dynamic interaction graph with social triplet factorization to decompose crowd interactions sequentially, and a community-aware module to effectively discover group structures via differentiable MinCut optimization and conditions both the embedding space and multi-modal decoder on group membership. Our experiments on standard benchmarks (ETH/UCY, SDD) demonstrate competitive performance with linear sequence complexity compared to quadratic attention-based methods. We further validate SGM in physical robot experiments by integrating predicted trajectories into a Social Force Model (SFM) for real-world implementation.
Human trajectory prediction has significant practical applications in various scenarios, such as autonomous driving, social robots and so on. Recently, it has been widely studied by diffusion models in order to model the inherent multi-modality of human motions. However, existing diffusion-based approaches only focus on modeling the social interactions
via
a single encoder and neglect the scene interactions, which results in producing unreasonable trajectories across obstacles or road boundaries. To address this issue, we propose the Interaction-Aware Diffusion Model (IADM), a novel diffusion-based framework considering both human motions and surrounding scene layout by treating the social and scene interactions as conditions in the parameterized reverse Markov chain. To implement IADM, we design two encoders,
i.e
., social encoder and scene encoder, where the social encoder models the social interactions
via
attention mechanism, and the scene encoder preserves spatial information of the scene when learning the scene interactions. Furthermore, we devise the dual-guidance decoder consisting of the motion-guided temporal module and the scene-guided spatial module to intensify the collaboratively guidance of the social and scene interactions. Extensive experiments on the ETH/UCY dataset, Stanford Drone Dataset and Intersection Drone Dataset validate the superiority of our method, achieving state-of-the-art results.
Zhong Zhang, Nuoran Wang, Song Gao et al.· PeerJ Computer Science· 0 citations
G-VTM, a generalized vision-trajectory model, is proposed, which captures global map semantics while modeling scenario-and direction-aware interaction based on intuitive visual perception and achieves strong generalized performance under heterogeneous traffic conditions.
Xinyue Zhang, Letian Gong, Yan Lin et al.· 0 citations
A comprehensive Explainable AI (XAI) evaluation framework is introduced, including temporal sensitivity analysis, interaction-aware perturbation studies, spatial influence analysis, and gradient-based feature attribution methods that provide insights into how the model captures temporal motion dependencies, neighboring vehicle interactions, and environmental context during trajectory prediction.
R. Megalingam, Naveen Prasaad Selvarajan, Pritty Vijay· Italian National Conference...· 0 citations
Accurate vehicle trajectory prediction is essential for the driving safety and efficiency of autonomous vehicles. However, this task remains challenging due to the complex spatial interactions among traffic participants and the wide range of temporal dependencies in motion sequences. To address these issues, this paper proposes a novel hybrid deep learning framework suitable for cloud-based control platforms, providing a foundational algorithmic solution for vehicle-infrastructure cooperative perception and decision-making. The proposed architecture employs an Adaptive Graph Convolutional Network (AGCN) to adaptively learn spatial relationships and interactions among vehicles. It also utilizes the Informer model, known for its efficient ProbSparse self-attention mechanism, to capture long-term temporal dependencies in trajectory sequences. Furthermore, a Temporal Convolutional Network (TCN) is integrated to enhance the model’s ability to learn fine-grained local temporal features. The proposed model is evaluated on the NGSIM dataset. The dataset is chronologically ordered and split into training (80%), validation (10%), and testing (10%) sets. Experimental results show that the proposed method achieves an average minADE of 2.032 m and minFDE of 2.866 m over a 5-second prediction horizon, outperforming several baseline models such as LSTM, CNN-LSTM, and Social-GAN. These results indicate the effectiveness of the AGCN-Informer-TCN combination for trajectory prediction. The study suggests potential for integration into intelligent transportation cloud control platforms.
Zi-Yan Liang, Rui Yuan, Peng-Ying Zhou et al.· SAE technical paper series· 0 citations
A Stochastic Gating Decoder for multimodal latent variable sampling, adaptively fusing kinematics and data-driven paths to capture driver intention uncertainty while maintaining kinematic consistency is introduced.
