Unsupervised drive topic finding from driving behavioral data

Bando, Takashi; Takenaka, Kazuhito; Nagasaka, Shogo; Taniguchi, Tadairo

doi:10.1109/ivs.2013.6629467

Cited by 37 publications

(29 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

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%

“…For topic modeling, latent Dirichlet allocation (LDA) is widely used. Bando et al used the DAA model with LDA for modeling driving behavior [5]. Integrating the two models and improving the prediction performance are for future work.…”

Section: Discussionmentioning

confidence: 99%

“…Taniguchi et al conducted a qualitative evaluation on the extracted driving words and argued that driving words differ even at the same place on the same course if the external situation changes, e.g., pedestrian passes in front of the car or a proceeding vehicle appears in front the car [4]. Bando et al argued that the DAA can segment driving-behavior data into favorably organized chunks from the viewpoint of topic modeling [5]. They also proposed a drive annotation method using the DAA and a topic modeling method [6].…”

Section: Introductionmentioning

confidence: 99%

See 2 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: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

“…Following to Bando et al, we used the latent Dirichlet allocation (LDA) for abstraction of driving situation symbols [9]. The LDA [11] proposed by Blei et al, one of the most basic Fig.…”

Section: B Abstraction Via Latent Topic Modelmentioning

confidence: 99%

“…Bando et al proposed a framework for automatically clustering driving situation symbols into "driving topics" [9]. They used latent Dirichlet allocation (LDA) for clustering extracted driving situations into a small number of driving topics in accordance with the emergence frequencies of the physical behavioral features observed in each driving situation.…”

Section: Introductionmentioning

confidence: 99%

Generating contextual description from driving behavioral data

Bando

Takenaka

Nagasaka

et al. 2014

2014 IEEE Intelligent Vehicles Symposium Proceedings

Self Cite

View full text Add to dashboard Cite

This paper presents an automatic translation method from time-series driving behavior into natural language with contextual information. Nowadays, various advanced driver-assistance systems (ADASs) have been developed to reduce the number of traffic accidents and multiple ADASs are required to reduce further accidents. For such multiple ADASs, considering the context of driving and selecting appropriate assistance is key because the systems have to handle extremely complicated driving situations consisting of drivers (and their intents and maneuvers), environments (including other traffic participants such as vehicles and pedestrians), and vehicles dynamics. In this paper, time-series driving behavior is segmented into typical driving situation symbols, and the natural language expression of each situation is generated via the behavioral feature distribution observed in each situation. Owing to the symbolization of the driving behavior, the generated behavioral descriptions can be associated with their causes and results not on an actual time axis but on a situation-symbol axis as contextual descriptions, e.g., "letting up on gas pedal to pass tollgate." The effectiveness of the proposed method was evaluated by using an actual data set of more than eight hours over a distance of 300 km in total. Although contextual expressions are very diverse even among human drivers, the proposed method obtained an agreement of more than 70%.

show abstract