A Double-Layer LSTM Model Based on Driving Style and Adaptive Grid for Intention-Trajectory Prediction

In the evolution of autonomous vehicles (AVs), ensuring safety is of the utmost significance. Precise trajectory prediction is indispensable for augmenting vehicle safety and system performance in intricate environments. This study introduces a novel double-layer long short-term memory (LSTM) model...

Full description

Saved in:
Bibliographic Details
Main Authors: Yikun Fan, Wei Zhang, Wenting Zhang, Dejin Zhang, Li He
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/7/2059
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850213126736183296
author Yikun Fan
Wei Zhang
Wenting Zhang
Dejin Zhang
Li He
author_facet Yikun Fan
Wei Zhang
Wenting Zhang
Dejin Zhang
Li He
author_sort Yikun Fan
collection DOAJ
description In the evolution of autonomous vehicles (AVs), ensuring safety is of the utmost significance. Precise trajectory prediction is indispensable for augmenting vehicle safety and system performance in intricate environments. This study introduces a novel double-layer long short-term memory (LSTM) model to surmount the limitations of conventional prediction methods, which frequently overlook predicted vehicle behavior and interactions. By incorporating driving-style category values and an improved adaptive grid generation method, this model achieves more accurate predictions of vehicle intentions and trajectories. The proposed approach fuses multi-sensor data collected by perception modules to extract vehicle trajectories. By leveraging historical trajectory coordinates and driving style, and by dynamically adjusting grid sizes according to vehicle dimensions and lane markings, this method significantly enhances the representation of vehicle motion features and interactions. The double-layer LSTM module, in conjunction with convolutional layers and a max-pooling layer, effectively extracts temporal and spatial features. Experiments conducted using the Next Generation Simulation (NGSIM) US-101 and I-80 datasets reveal that the proposed model outperforms existing benchmarks, with higher intention accuracy and lower root mean square error (RMSE) over 5 s. The impact of varying sliding window lengths and grid sizes is examined, thereby verifying the model’s stability and effectiveness.
format Article
id doaj-art-8cceb0bef02b4967b00bb7eab1cb7d0d
institution OA Journals
issn 1424-8220
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-8cceb0bef02b4967b00bb7eab1cb7d0d2025-08-20T02:09:11ZengMDPI AGSensors1424-82202025-03-01257205910.3390/s25072059A Double-Layer LSTM Model Based on Driving Style and Adaptive Grid for Intention-Trajectory PredictionYikun Fan0Wei Zhang1Wenting Zhang2Dejin Zhang3Li He4College of Electronics and Information Engineering, Shenzhen University, Shenzhen 518000, ChinaInstitute of Applied Artificial Intelligence of the Guangdong-Hong Kong-Macao Greater Bay Area, Shenzhen Polytechnic University, Shenzhen 518000, ChinaCollege of Electronics and Information Engineering, Shenzhen University, Shenzhen 518000, ChinaSchool of Architecture & Urban Planning, Shenzhen University, Shenzhen 518000, ChinaCollege of Mechatronics and Control Engineering, Shenzhen University, Shenzhen 518000, ChinaIn the evolution of autonomous vehicles (AVs), ensuring safety is of the utmost significance. Precise trajectory prediction is indispensable for augmenting vehicle safety and system performance in intricate environments. This study introduces a novel double-layer long short-term memory (LSTM) model to surmount the limitations of conventional prediction methods, which frequently overlook predicted vehicle behavior and interactions. By incorporating driving-style category values and an improved adaptive grid generation method, this model achieves more accurate predictions of vehicle intentions and trajectories. The proposed approach fuses multi-sensor data collected by perception modules to extract vehicle trajectories. By leveraging historical trajectory coordinates and driving style, and by dynamically adjusting grid sizes according to vehicle dimensions and lane markings, this method significantly enhances the representation of vehicle motion features and interactions. The double-layer LSTM module, in conjunction with convolutional layers and a max-pooling layer, effectively extracts temporal and spatial features. Experiments conducted using the Next Generation Simulation (NGSIM) US-101 and I-80 datasets reveal that the proposed model outperforms existing benchmarks, with higher intention accuracy and lower root mean square error (RMSE) over 5 s. The impact of varying sliding window lengths and grid sizes is examined, thereby verifying the model’s stability and effectiveness.https://www.mdpi.com/1424-8220/25/7/2059trajectory predictionLSTMdriving stylegrids in interactionautonomous vehicle
spellingShingle Yikun Fan
Wei Zhang
Wenting Zhang
Dejin Zhang
Li He
A Double-Layer LSTM Model Based on Driving Style and Adaptive Grid for Intention-Trajectory Prediction
Sensors
trajectory prediction
LSTM
driving style
grids in interaction
autonomous vehicle
title A Double-Layer LSTM Model Based on Driving Style and Adaptive Grid for Intention-Trajectory Prediction
title_full A Double-Layer LSTM Model Based on Driving Style and Adaptive Grid for Intention-Trajectory Prediction
title_fullStr A Double-Layer LSTM Model Based on Driving Style and Adaptive Grid for Intention-Trajectory Prediction
title_full_unstemmed A Double-Layer LSTM Model Based on Driving Style and Adaptive Grid for Intention-Trajectory Prediction
title_short A Double-Layer LSTM Model Based on Driving Style and Adaptive Grid for Intention-Trajectory Prediction
title_sort double layer lstm model based on driving style and adaptive grid for intention trajectory prediction
topic trajectory prediction
LSTM
driving style
grids in interaction
autonomous vehicle
url https://www.mdpi.com/1424-8220/25/7/2059
work_keys_str_mv AT yikunfan adoublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT weizhang adoublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT wentingzhang adoublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT dejinzhang adoublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT lihe adoublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT yikunfan doublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT weizhang doublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT wentingzhang doublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT dejinzhang doublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction
AT lihe doublelayerlstmmodelbasedondrivingstyleandadaptivegridforintentiontrajectoryprediction