Freeway Short-Term Travel Time Prediction Based on LightGBM Algorithm

Wang, Fahui; Cheng, Hui-Ling; Dai, H. L.; Han, Haihang

doi:10.1088/1755-1315/638/1/012029

Cited by 8 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the LightGBM algorithm selects 'Leaf-wise' leaf growth strategy with depth restriction. This leaf node expansion method can reduce the training error and get better accuracy [38]. Simultaneously, the performance of LightGBM has been significantly improved by technology optimizations such as support for parallel learning, support of the histogram algorithm, and so on.…”

Section: Use Lightgbm To Establish a Retrieval Modelmentioning

confidence: 99%

An Improved Method of Soil Moisture Retrieval Using Multi-Frequency SNR Data

et al. 2021

View full text Add to dashboard Cite

Soil moisture monitoring using Global Navigation Satellite System (GNSS) multipath signals has gained continuous interests in recent years. However, traditional GNSS-interferometric reflectometry (GNSS-IR) soil moisture retrieval methods generally utilize a single frequency or single satellite, which fail to take full advantage of different and complementary of satellite signals with different frequencies. An improved algorithm for soil moisture retrieval based on principal component analysis (PCA) and entropy method using multi-frequency amplitude and phase offset fusion data was proposed in this research. The performance of the proposed soil moisture retrieval method was evaluated using data recorded by Plate Boundary Observatory (PBO) H2O networks and a self-built site in Henan, China. The results from GPS and BeiDou both showed that the retrieved soil moisture has a stronger correlation with in situ soil moisture, which can better reflect the fluctuation of ground truth measurements. Compared with the traditional method, the retrieval accuracy of the proposed method in terms of root-mean-square error (RMSE) was improved by 50.93%, and the average correlation coefficient were increased by 11.71%. This research proved that the proposed method could effectively improve retrieval accuracy due to the increasing number of frequencies and tracks clustering. Moreover, this study has illustrated the feasibility of BeiDou signals to precisely estimate surface soil moisture.

show abstract

Section: Use Lightgbm To Establish a Retrieval Modelmentioning

confidence: 99%

An Improved Method of Soil Moisture Retrieval Using Multi-Frequency SNR Data

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Xia et al [29] proposed a new traffic prediction model based on an integrated framework of bagging and LightGBM. Wang et al [30] proposed a short-term driving time prediction model based on LightGBM. The experiment showed good advantages in prediction accuracy and training speed compared with a KNN model and gradient boosting decision tree (GBDT) model.…”

Section: Related Workmentioning

confidence: 99%

Dynamic Identification Method for Potential Threat Vehicles beyond Line of Sight in Expressway Scenarios

Zou,

Xia,

Guo

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Due to the challenge of limited line of sight in the perception system of intelligent driving vehicles (cameras, radar, body sensors, etc.), which can only perceive threats within a limited range, potential threats outside the line of sight cannot be fed back to the driver. Therefore, this article proposes a safety perception detection method for beyond the line of sight for intelligent driving. This method can improve driving safety, enabling drivers to perceive potential threats to vehicles in the rear areas beyond the line of sight earlier and make decisions in advance. Firstly, the electronic toll collection (ETC) transaction data are preprocessed to construct the vehicle trajectory speed dataset; then, wavelet transform (WT) is used to decompose and reconstruct the speed dataset, and lightweight gradient noosting machine learning (LightGBM) is adopted to train and learn the features of the vehicle section speed. On this basis, we also consider the features of vehicle type, traffic flow, and other characteristics, and construct a quantitative method to identify potential threat vehicles (PTVs) based on a fuzzy set to realize the dynamic safety assessment of vehicles, so as to effectively detect PTVs within the over-the-horizon range behind the driver. We simulated an expressway scenario using an ETC simulation platform to evaluate the detection of over-the-horizon PTVs. The simulation results indicate that the method can accurately detect PTVs of different types and under different road scenarios with an identification accuracy of 97.66%, which verifies the effectiveness of the method in this study. This result provides important theoretical and practical support for intelligent driving safety assistance in vehicle–road collaboration scenarios.

show abstract

“…Concurrently, predicting the arrival time remains a predominant challenge for other means of PT [8], i.e bus itineraries, or demand-responsive transport (DRT), whether taxi routes [16] [17] or private trips [18], with applications even in the field of logistics supply chains [19]. Considering DRT, the authors of [16] collected GPS taxi data from New York and developed prediction models based on both regression and classification.…”

Section: Related Workmentioning

confidence: 99%

“…Furthermore, the studies based on freeway private trips are equally important. Indeed, the focus of [18] is ETA prediction in the Shanghai-Hangzhou-Ningbo freeway, comparing LightGBM to GB Decision Trees and K-Nearest Neighbor algorithm. Using MAE and MAPE as evaluation metrics, LightGBM was superior to the other prediction models.…”

Section: Related Workmentioning

confidence: 99%

Estimated Time of Arrival in Autonomous Vehicles Using Gradient Boosting: Real-life case study in public transportation

Antypas

Spanos

Lalas

et al. 2022

2022 IEEE International Smart Cities Conference (ISC2)

View full text Add to dashboard Cite

Autonomous Vehicles (AVs) are expected to revolutionise the methods of transportation. Research and innovation in this field is making huge leaps in the last few years, whether it considers vehicles used for private or public transport. Predicting the Estimated Time of Arrival (ETA) is a very important attribute associated with Public Transport (PT). Especially with the rise of AVs' adoption, PT is expected to follow this trend. Therefore, ETA prediction is deemed to be a service that interests the majority of PT stakeholders. PT is a field that automation benefits both stakeholders and commuters, and this research aims to provide a benchmark considering AVs in PT. Within this work, Gradient Boosting (GB) techniques for ETA prediction were employed, namely eXtreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost) and Light Gradient Boosting Machines (LightGBM). This study proposes competitive ETA prediction methods in Autonomous Buses, while the results of this research are very encouraging and aim to contribute to the overall investigations in the field of autonomous and automated PT.

show abstract

Freeway Short-Term Travel Time Prediction Based on LightGBM Algorithm

Cited by 8 publications

References 7 publications

An Improved Method of Soil Moisture Retrieval Using Multi-Frequency SNR Data

An Improved Method of Soil Moisture Retrieval Using Multi-Frequency SNR Data

Dynamic Identification Method for Potential Threat Vehicles beyond Line of Sight in Expressway Scenarios

Estimated Time of Arrival in Autonomous Vehicles Using Gradient Boosting: Real-life case study in public transportation

Contact Info

Product

Resources

About