Optimized Long Short-Term Memory Network for LiDAR-Based Vehicle Trajectory Prediction Through Bayesian Optimization

In vehicle trajectory prediction, traditional methods like Kalman filtering often rely heavily on user expertise and prior knowledge, while newer deep learning approaches, such as Long Short-Term Memory (LSTM) networks, also face challenges related to human intervention and subjective hyperparameter selection. This study proposes a systematic approach for Light Detection and Ranging (LiDAR)-based vehicle trajectory prediction, leveraging LSTM networks to predict vehicle trajectories and employing Bayesian optimization to automatically search for optimal hyperparameter values related to both the training scheme and LSTM architectures. In the experimental study, a custom vehicle trajectory dataset extracted from roadside LiDAR data, along with the V2X-Seq-TFD dataset, was utilized for network training and testing. The optimal LSTM network obtained through Bayesian optimization was compared against two benchmark models: a handcrafted LSTM network and a Kalman filter with a 2D constant velocity motion model. The results demonstrate that the proposed deep learning-based framework, with robust hyperparameter selection through Bayesian optimization, yields more accurate and consistent prediction performance than the benchmark models.

I. Introduction

Over the past decade, Light Detection and Ranging (LiDAR) technology has rapidly emerged as one of the most prominent and advanced sensing technologies for transportation research and applications. Its uses span a wide range of domains, such as Traffic Big Data (TBD) analytics, surveying, transportation system modeling and optimization, Connected Autonomous Vehicles (CAV), traffic object recognition and tracking, and sensor fusion technologies [1], [2], [3]. Compared to traditional sensors such as cameras, LiDAR offers distinct advantages: insensitivity to changing illumination conditions, high-resolution distance measurement, wide-range detection, and high-speed data acquisition [1], [4], [5], [6]. These features make LiDAR sensors ideal for deployment on mobile platforms or transportation infrastructure, providing distance measurements in traffic scenarios that require high-speed acquisition of high-quality three-dimensional (3D) data under varying lighting conditions, such as in autonomous driving.