ITSC 2025 Paper Abstract

Paper TH-EA-T30.6

Vora, Jay Umesh (Wilfrid Laurier University), Sehra, sukhjit (wilfrid laurier university), sehra, sumeet kaur (Conestoga College), Singh, Jaiteg (chitkara University)

Efficacy of Early and Late Spatio-Temporal Fusion in LSTM Models for Vehicle Trajectory Prediction

Scheduled for presentation during the Regular Session "S30b-Intelligent Modeling and Prediction of Traffic Dynamics" (TH-EA-T30), Thursday, November 20, 2025, 14:50−15:30, Gold Coast

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on October 18, 2025

Keywords AI, Machine Learning for Real-time Traffic Flow Prediction and Management, Data Analytics and Real-time Decision Making for Autonomous Traffic Management, Model-based Validation of Traffic Flow Prediction Algorithms

Abstract

Vehicle trajectory prediction is crucial for optimizing transportation systems, reducing traffic congestion, minimizing travel time, and enhancing overall safety. Accurately forecasting trajectories in complex traffic remains challenging due to both temporal vehicle dynamics and spatial interactions among neighboring vehicles. In this paper, we present a head‑to‑head comparison of two long short‑term memory (LSTM) architectures that fuse spatio‑temporal data at different stages: an early‑fusion model that aggregates neighbor features at every time step, and a late‑fusion model that encodes target and neighbor sequences separately and merges them only at the final prediction layer. A vanilla LSTM baseline that uses only the target vehicle’s own history (no neighbor information) is also included to isolate the fusion benefit. All models are trained and evaluated on the NGSIM US‑101 highway dataset using mean squared error (MSE), mean absolute error (MAE), and R² metrics. The results show that both fusion strategies yield modest error reductions compared to the baseline (vanilla LSTM MSE = 1.65 → early‑fusion MSE = 0.9874, late‑fusion MSE = 0.9910), with only a slight advantage for early fusion. We provide qualitative analyses of lane‑change versus lane‑keep scenarios, discuss the limitations of our one‑step prediction and simple neighbor pooling, and outline directions for future work, including attention mechanisms and graph‑based models. The data and implementation are available at https://github.com/sukhjitsehra/FVTP.