Modeling dangerous driving events based on in-vehicle data using Random Forest and Recurrent Neural Network

Alvarez-Coello, Daniel; Klotz, Benjamin; Wilms, Daniel; Fejji, Sofien; Gómez, Jorge Marx; Troncy, Raphaël

doi:10.1109/ivs.2019.8814069

Cited by 21 publications

(14 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance of the model for classifying each class was not mentioned in the work. Alvarez-Coello et al [22] proposed and split the supervised multi-class driving maneuver classification problem into two parts. The authors developed a binary classifier by applying RF in order to classify aggressive and non-aggressive driving events and the result was transferred to the RNN model to recognize the type of maneuver.…”

Section: Deep Learning Approachesmentioning

confidence: 99%

“…However, in a real-life situation, this setup may be interrupted because the drivers may use smartphone for communication and navigation [14]. To overcome this problem, some researchers utilized previously collected sensor fusion dataset [20]- [22]. Researchers proposed various fuzzy inference, machine learning, deep learning techniques to classify driving behavior from inertial sensors data [20].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Driving Maneuver Classification Using Domain Specific Knowledge and Transfer Learning

2021

View full text Add to dashboard Cite

With the increasing number of vehicles, the usage of technology has also been increased in the transportation system. Although automobile companies are using advanced technologies to develop high performing transports, traffic safety still remains to be a concerning issue. Drivers' driving behavior is considered as one of the key factors of the traffic safety, which could be monitored from their individual driving maneuvers. In this paper, we present a supervised learning model and a semi-supervised transfer learning model for the classification of driving maneuvers from the sensor fusion time series data. The semi-supervised model consists of an unsupervised long-short term memory (LSTM) autoencoder and a supervised LSTM classifier. The supervised model consists of a supervised LSTM model. Because of using LSTM, both of the models can analyze time-series data. In the semi-supervised model, the LSTM encoder learns from unlabeled data as a compressed low dimensional feature vector, which then transfers the learning to the supervised LSTM classifier to classify the driving maneuvers. With the proposed models, we use domain specific knowledge data of the driving environment, such as data changing rules of various driving maneuvers as well as the temporal features over time. We use class functions for seven driving maneuver types and convert those into binary feature vector to use with the LSTM models. We present a comparative analysis of the per class accuracy of the proposed semi-supervised and supervised models with and without using domain-specific knowledge, where the models with the domain specific knowledge outperform. Our proposed semi-supervised and supervised models are compared with the other existing approaches, where our models trained with the domain specific knowledge provide better performance. We also compare the per class accuracy for both the supervised and semi-supervised models, where all the maneuver class accuracy for supervised model is above 98% and semi-supervised model is above 95%. Although the supervised model outperforms the semi-supervised model, semi-supervised model would be more beneficial in applications where the labeled driving maneuvers data is hard to capture or insufficient.INDEX TERMS Driving maneuver classification, domain specific knowledge, LSTM autoencoder, semisupervised learning, transfer learning.

show abstract

Section: Deep Learning Approachesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Driving Maneuver Classification Using Domain Specific Knowledge and Transfer Learning

2021

View full text Add to dashboard Cite

show abstract

“…Moreover, drivers' behavior is closely related to driving maneuvers. Another researcher in [16] focused on an approach to implement supervised time series classification with Random Forest (RF) and Recurrent Neural Network separately. They introduced RF to classify aggressive events and the result of RF transferred to recognize the type of maneuver.…”

Section: Existing Approachmentioning

confidence: 99%

An approach towards data change rule based classification of driving maneuver with LSTM network

Sarker¹,

Haque²

2020

Preprint

View full text Add to dashboard Cite

The proposed work develops a Long Short Term Memory (LSTM) model for multi class classification of driving maneuver from sensor fusion time series dataset. The work also analyzes the significance of sensor fusion data change rule and utilized the idea with deep learning time series multi class classification of driving maneuver. We also proposed some hypotheses which are proven by the experimental results. The proposed model provides Train Accuracy: 99.98, Test Accuracy: 97.2021, Precision: 0.974848, Recall: 0.960154 and F1 score: 0.967028. The Mean Per Class Error (MPCE) is 0.01386. The significant rules can accelerate the feature extraction process of driving data. Moreover, it helps in automatic labeling of unlabeled dataset. Our future approach is to develop a tool for generating categorical label for unlabeled dataset. Besides, we have plan to optimize the proposed classifier using grid search. <br>

show abstract

“…RF is categorized in supervised learning algorithms. Several studies [49], [50] discuss using RF implemented in CV environments to classify supervised timeseries based on driver behaviour and to classify vehicle recognition to differentiate among road users such as pedestrians, bicycles, cars and others. As a result, these two studies show that RF is effective to prevent overfitting for driver behaviour and successfully integrate the data and processing to identify and differentiate road user.…”

Section: ) Decision Tree (Dt)mentioning

confidence: 99%

Conceptual Model for Connected Vehicles Safety and Security using Big Data Analytics

Razali¹,

Shamsaimon²,

Wook³

et al. 2020

IJACSA

View full text Add to dashboard Cite

The capability of Connected Vehicles (CVs) connecting to a nearby vehicle, surrounding infrastructure and cyberspace presents a high risk in the aspect of safety and security of the CV and others. Data volume generated from the sensors and infrastructure in CVs environment are enormous. Thus, CVs implementations require a real-time big data processing and analytics to detect any anomaly in the CVs's environment which are physical layer, network layer and application layer. CVs are exposed to various vulnerabilities associated with exploitations or malfunctions of the components in each layer that could result in various safety and security event such as congestion and collision. The safety and security risks added an extra layer of required protection for the CVs implementation that need to be studied and refined. To address this gap, this research aims to determine the basic components of safety and security for CVs implementation and propose a conceptual model for safety and security in CVs by applying the machine learning and deep learning techniques. The proposed model is highly correlated to safety and security and could be applied in congestion and collision prediction.

show abstract

Modeling dangerous driving events based on in-vehicle data using Random Forest and Recurrent Neural Network

Cited by 21 publications

References 22 publications

Driving Maneuver Classification Using Domain Specific Knowledge and Transfer Learning

Driving Maneuver Classification Using Domain Specific Knowledge and Transfer Learning

An approach towards data change rule based classification of driving maneuver with LSTM network

Conceptual Model for Connected Vehicles Safety and Security using Big Data Analytics

Contact Info

Product

Resources

About