Structured behaviour prediction of on‐road vehicles via deep forest

Mou, Luntian; Mao, Shasha; Xie, Hualong; Chen, Yanyan

doi:10.1049/el.2019.0472

Cited by 9 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This algorithm has also been implemented for the study of prediction of structured behaviour of road vehicles. Mou et al [4] proposed a structured gcForest modeling approach for the prediction of road vehicle behavior. According to Liu et al [5], deep neural networks (DNNs) for intelligent machine failure diagnosis require a large amount of training data, powerful computational resources, and many hyperparameters, which must be carefully tuned to ensure maximum performance.…”

Section: Literature Reviewmentioning

confidence: 99%

Failure Prediction and Intelligent Maintenance of a Transportation Company’s Urban Fleet

Foké¹,

Kenné²,

Diego³

2023

JTTs

View full text Add to dashboard Cite

The present work deals with intelligent vehicle fleet maintenance and prediction. We propose an approach based primarily on the history of failures data and on the geographical data system. The objective here is to predict the date of failures for a fleet of vehicles in order to allow the maintenance department to efficiently deploy the proper resources; we further provide specific details regarding the origins of failures, and finally, give recommendations. This study used the Société de transport de Montréal (STM)'s historical bus failure data as well as weather data from Environment Canada. We thank Facebook's Prophet, Simple Feed-forward, and Beats algorithms (Uber), we proposed a set of computer codes that allow us to identify the 20% of buses that are responsible for the 80% of failures by mean of the failure history. Then, we deepened our study on the unreliable equipments identified during the diffusion of our computer code This allowed us to propose probable predictions of the dates of future failures. To ensure the validity of the proposed algorithm, we carried out simulations with more than 250,000 data. The results obtained are similar to the predicted theoretical values.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Failure Prediction and Intelligent Maintenance of a Transportation Company’s Urban Fleet

Foké¹,

Kenné²,

Diego³

2023

JTTs

View full text Add to dashboard Cite

show abstract

“…Yang et al (2018) [21] combined a simple region proposal network and a gcForest for ship detection based on thermal remote sensing image, where the region proposal network is designed to pre-process images so as to extract candidate regions from complex backgrounds. Mou et al (2019) [22] proposed a structured gcForest modelling approach for on-road vehicles behavior prediction. The images of on road automobiles were used as input.…”

Section: B the Studies Of Gcforestmentioning

confidence: 99%

Automobile Maintenance Modelling Using gcForest

Chen

Liu

Sun

et al. 2020

2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)

View full text Add to dashboard Cite

Automobile maintenance has gained increasing attention in recent years. If the failure time of an automobile can be predicted, it can bring tangible benefits to automobile fleet management. The Multi-Grained Cascade Forest (gcForest) is a tree-based deep learning algorithm, which was originally developed for image classification, but is potentially a helpful tool in automobile maintenance. This study aims to introduce the gcForest into automobile maintenance based on automobile historical maintenance data and geographical information system (GIS) data. The experimental results reveal that the gcForest shows merits in automobile time-between-failure (TBF) modelling, while it requires less computational cost.

show abstract

“…In earlier work [5, 6] we give a structural analysis of the instantaneous vehicle behaviour state including behaviour trend and degree simultaneously based on the vehicle images, where it only needs to detect and locate the vehicle without tracking. Inspired by structured learning, we argue that the degree of the turning angle represents the extent to which the turning behaviour occurs, so we design the structured label where the curves with different bending angles represent turning behaviours and straight lines represent straight‐through behaviour.…”

Section: Introductionmentioning

confidence: 99%