Contextual scene segmentation of driving behavior based on double articulation analyzer

Takenaka, Kazuhito; Bando, Takashi; Nagasaka, Shogo; Taniguchi, Tadahiro; Hitomi, Kentarou

doi:10.1109/iros.2012.6385614

Cited by 27 publications

(14 citation statements)

References 10 publications

(12 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is also assumed that a boundary between two adjacent driving words represents a contextual changing point. These assumptions are supported by several previous studies [1], [3]- [5]. Fig.…”

Section: B Daa and Predictionsupporting

confidence: 87%

“…Takenaka et al reported that the DAA was able to infer contextual changing points of driving-behavior data as boundaries of driving words [1]. 1 They used a sticky HDP-HMM for segmenting driving-behavior data and obtaining driving letter sequences.…”

Section: A Daa Using Hdp-hsmmmentioning

confidence: 99%

“…From this viewpoint, driving behavior is considered as a sequence of meaningful driving-behavior chunks. To analyze observed driving-behavior time series data, Takenaka et al proposed using the double articulation analyzer (DAA) model on the basis of this idea [1].…”

Section: Introductionmentioning

confidence: 99%

“…Several types of time series data, including human speech signal, human motion data, and human driving behavior, are assumed to have a double articulation structure. Regarding driving-behavior data, Takenaka et al insisted that a latent word corresponds to meaningful driving-behavior units such as "starting with turning right" [1]. Latent words are assumed to be the latent information behind observed driving behavior, which we listed above as examples.…”

Section: Introductionmentioning

confidence: 99%

“…Takenaka et al applied the DAA model to driving-behavior data [1]. They found that the boundaries of driving words represent contextual changing points more properly than the boundaries of driving letters.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Unsupervised Hierarchical Modeling of Driving Behavior and Prediction of Contextual Changing Points

Taniguchi

Nagasaka

Hitomi

et al. 2015

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

An unsupervised learning method, called double articulation analyzer with temporal prediction (DAA-TP), is proposed on the basis of the original DAA model. The method will enable future advanced driving assistance systems to determine driving context and predict possible scenarios of driving behavior by segmenting and modeling incoming driving-behavior time series data. In previous studies, we applied the DAA model to driving-behavior data and argued that contextual changing points can be estimated as changing points of chunks. A sequence prediction method, which predicts the next hidden state sequence, was also proposed in a previous study. However, the original DAA model does not model the duration of chunks of driving behavior and is not able to do a temporal prediction of the scenarios. Our DAA-TP method explicitly models the duration of chunks of driving behavior on the assumption that driving-behavior data have a two-layered hierarchical structure, i.e., double articulation structure. For this purpose, the hierarchical Dirichlet process hidden semi-Markov model is used for explicitly modeling the duration of segments of driving-behavior data. A Poisson distribution is also used to model the duration distribution of driving-behavior segments. The duration distribution of chunks of driving-behavior data is also theoretically calculated using the reproductive property of the Poisson distribution. We also propose a calculation method for obtaining the probability distribution of the remaining duration of current driving words as a mixture of Poisson distribution with a theoretical approximation for unobserved driving words. This method can calculate the posterior probability distribution of the next termination time of chunks by explicitly modeling all probable chunking results for observed data. The DAA-TP was applied to a synthetic data set having a double articulation structure to evaluate its model consistency. To evaluate the effectiveness of DAA-TP, we applied it to a drivingbehavior data set recorded at actual factory circuits. The DAA-TP could predict the next termination time of chunks more accurately than the compared methods. We also report the qualitative results for understanding the potential capability of DAA-TP.

show abstract

Section: B Daa and Predictionsupporting

confidence: 87%

Section: A Daa Using Hdp-hsmmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Unsupervised Hierarchical Modeling of Driving Behavior and Prediction of Contextual Changing Points

Taniguchi

Nagasaka

Hitomi

et al. 2015

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

Model-based Behavioral Causality Analysis of Handball with Delayed Transfer Entropy

Itoda¹,

Watanabe²,

Takefuji³

2015

Procedia Computer Science

View full text Add to dashboard Cite

Variable Selection and Modeling of Drivers’ Decision in Overtaking Behavior Based on Logistic Regression Model With Gazing Information

Nwadiuto

Yoshino²,

Okuda

et al. 2021

IEEE Access

View full text Add to dashboard Cite

This paper investigates the decision-making characteristics of the driver in the overtaking on the highway road. For the research purpose, a novel method was proposed by introducing a logistic regression model accompanied by the statistical test technique, which does not require prior knowledge about the explanatory variables. This study hypothesizes that the driver's gazing behavior is crucial for the decision-making process in driving and hence, the line-of-sight information was introduced to estimate driver's gazing behavior in the model of driver's decision specifically for reproducing the overtaking driving behavior accurately. Consequently, the proposed model realized a high describability on the decision of the driver when performing the overtaking driving task, which is one of the significant advancements of the present study with respect to the past similar studies. This study integrates the perspectives of intelligent vehicle design and cognitive science by revealing which factor the driver pays attention to in a changeable driving environment due to various observable factors. In experiments based on the driving simulator with six human subjects, the overtaking behavior was successfully estimated by specifying a set of variables to reconstruct the driver's behavior and then the proposed model provided a minimum set of necessary variables accompanied with key coefficients. In conclusion, the proposed approach based on a simple logistic regression model demonstrated driving behaviors with an accurate estimation of the driver's intention without the need for prior knowledge, and it may contribute to higher describability for various driving actions in a dynamic environment.

show abstract

Contextual scene segmentation of driving behavior based on double articulation analyzer

Cited by 27 publications

References 10 publications

Unsupervised Hierarchical Modeling of Driving Behavior and Prediction of Contextual Changing Points

Unsupervised Hierarchical Modeling of Driving Behavior and Prediction of Contextual Changing Points

Model-based Behavioral Causality Analysis of Handball with Delayed Transfer Entropy

Variable Selection and Modeling of Drivers’ Decision in Overtaking Behavior Based on Logistic Regression Model With Gazing Information

Contact Info

Product

Resources

About