A trajectory prediction method based on bayonet importance encoding and bidirectional LSTM

Guan, Lechen; Shi, Jingtian; Wang, Dongle; Shao, Hu; Chen, Zilong; Chu, Danhua

doi:10.1016/j.eswa.2023.119888

Cited by 8 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given the rapid development of deep learning models in recent years, various individual mobility prediction methods based on deep learning have emerged. Recurrent neural networks and their variants have been widely used for next-location prediction (Al-Molegi et al, 2016;Guan et al, 2023;Liu et al, 2019;Zhang et al, 2022). Such models can capture high-order spatiotemporal dependencies and periodic features, demonstrating better performance than statistical models.…”

Section: Literature Reviewmentioning

confidence: 99%

Real-time Prediction of Construction Waste Hauling Trucks’ Transportation Activities: An Input‒Output Hidden Markov Modelling Approach

Yang,

Lei,

Han

et al. 2023

Preprint

View full text Add to dashboard Cite

Construction waste hauling trucks (CWHTs), as one of the most commonly seen heavy-duty vehicles in major cities around the globe, are usually subject to a series of regulations and access restrictions because they not only produce significant NOx and PM emissions but also causes on-road fugitive dust. To limit their environmental impact, specific spatial-temporal access restrictions are implemented, and timely and accurate detection of possible infringements has become a key challenge facing many municipal managers. To address this challenge, we propose a prediction method based on an interpretable activity-based model, input-output hidden Markov model (IOHMM), and apply it to trajectories of 300 CWHTs in Chengdu, China for evaluation. Contextual factors including weather conditions, recent transportation activities, and historical work statistics are considered in the model to improve its prediction power. Results show that the IOHMM has an average percentage of 64% for correctly predicting the destination and an average \({R}^{2}\) value of 69% for predicting the duration of the transportation activity. These values are higher than the results of the baseline models, including Markov chains, linear regression, and long short-term memory. To investigate the factors that influence the predictability of CWHTs’ transportation activities, linear regression models are constructed using the percentage of correct destination predictions and \({R}^{2}\) of the duration prediction model as the dependent variables, respectively. Our findings indicate that the number of active days and the proportion of days with nighttime activities are positively correlated with the prediction accuracy of both models. The average number of daily transportation activities and the number of different staying activities have positive and negative effects on the prediction accuracy of the destination model, respectively. In conclusion, the proposed model holds promise in assisting government agencies by predicting the next transportation activities of CWHTs and detecting infringements in a timely and effective manner.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Real-time Prediction of Construction Waste Hauling Trucks’ Transportation Activities: An Input‒Output Hidden Markov Modelling Approach

Yang,

Lei,

Han

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Zhan et al 15 through mining checkpoint data, combined with machine learning algorithm and traffic flow theory, proposed a model for estimating real-time lane queue length. Guan et al 16 proposed a deep learning model utilizing the Bi-LSTM self-attention mechanism to predict the next position of a vehicle based on the historical vehicle trajectory data collected by the checkpoint system. Huang et al 17 constructed a traffic network based on the context of the traffic checkpoint context and used a bi-directional gated recurrent unit (Bi-GRU) for vehicle trajectory prediction.…”

Section: Introductionmentioning

confidence: 99%

Checkpoint Data-Driven GCN-GRU Vehicle Trajectory and Traffic Flow Prediction

Guan,

Ren,

Wang

et al. 2024

Preprint

View full text Add to dashboard Cite

In order to accurately assess the road traffic flow, improve the efficiency and safety of the transportation system, and provide technical support for vehicle path planning and road congestion early warning, this paper proposes a method for accurately forecast the traffic flow on the urban road network by using trajectory prediction technology. The method uses a combination model of graph convolutional neural network (GCN) and gated recurrent unit (GRU) for vehicle trajectory prediction, and uses the output of trajectory prediction to more accurately forecast the traffic flow. Firstly, transforming the checkpoint data into daily vehicle trajectories with time series characteristics, realizing the division of vehicle trajectory travel chain. Secondly, the adjacency matrix is established by using the spatial relationship of each checkpoint, and the feature matrix of the vehicle's driving trajectory over time is established, which are used as the input of GCN to learn the spatial characteristics of the vehicle during driving on the road network, and then GRU is added to further process the data after GCN training, constructing a GCN-GRU vehicle trajectory prediction model for vehicle trajectory prediction. Finally, the traffic flow of each checkpoint is calculated based on the prediction result of vehicle trajectory, and compared with the real checkpoint flow. In this paper, Qingdao city checkpoint data is used as an example for experiments, and the results show that compared with the traditional GCN model, the GCN-GRU vehicle trajectory prediction model can obtain stronger temporal and spatial correlation characteristics and higher prediction accuracy. The forecasted traffic flow is highly consistent with the real checkpoint traffic flow, which verifies the reliability of the proposed method.

show abstract

An Improved Machine Learning Method to Speed up the Trajectory Prediction: Taking Melbourne Airport as a Study Case

Xi,

Ma,

Wang

et al. 2024

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

A trajectory prediction method based on bayonet importance encoding and bidirectional LSTM

Cited by 8 publications

References 19 publications

Real-time Prediction of Construction Waste Hauling Trucks’ Transportation Activities: An Input‒Output Hidden Markov Modelling Approach

Real-time Prediction of Construction Waste Hauling Trucks’ Transportation Activities: An Input‒Output Hidden Markov Modelling Approach

Checkpoint Data-Driven GCN-GRU Vehicle Trajectory and Traffic Flow Prediction

An Improved Machine Learning Method to Speed up the Trajectory Prediction: Taking Melbourne Airport as a Study Case

Contact Info

Product

Resources

About