Cluster Naturalistic Driving Encounters Using Deep Unsupervised Learning

Li, Sisi; Wang, Wenshuo; Mo, Zhaobin; Zhao, Ding

doi:10.1109/ivs.2018.8500529

Cited by 21 publications

(9 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on a set of behaviours, manoeuvre-based motion prediction approaches employ estimation techniques, for instance Gaussian Processes (Christopher, 2009), (Joseph et al, 2011) to then estimate most probable future manoeuvres. Deep-learning techniques have also been applied to cluster vehicle encounters (Li et al, 2018).…”

Section: Perceptionmentioning

confidence: 99%

AI perspectives in Smart Cities and Communities to enable road vehicle automation and smart traffic control

Englund¹,

Aksoy²,

Alonso‐Fernandez³

et al. 2021

Preprint

View full text Add to dashboard Cite

Smart Cities and Communities (SCC) constitute a new paradigm in urban development. SCC ideates on a data-centered society aiming at improving efficiency by automating and optimizing activities and utilities. Information and communication technology along with internet of things enables data collection and with the help of artificial intelligence (AI) situation awareness can be obtained to feed the SCC actors with enriched knowledge. This paper describes AI perspectives in SCC and gives an overview of AI-based technologies used in traffic to enable road vehicle automation and smart traffic control. Perception, Smart Traffic Control and Driver Modelling are described along with open research challenges and standardization to help introduce advanced driver assistance systems in traffic. AI technologies provide accurate prediction and classification; however, the models do not provide any evidence on their output making them hard to trust for a human operator. In addition, there are currently no methods that can be used to describe requirements of how the data should be annotated in order to train an accurate model.

show abstract

Section: Perceptionmentioning

confidence: 99%

AI perspectives in Smart Cities and Communities to enable road vehicle automation and smart traffic control

Englund¹,

Aksoy²,

Alonso‐Fernandez³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…• Decoder: The hidden representation h from the encoder is then mapped back tox through a symmetric multilayer network. Subsequently, an autoencoder with a simple multilayer network can be easily derived using (2) without considering the past observations, which has been thoroughly investigated in [22]. Driving behavior, however, is a dynamic process in nature which depends on past information [25], [26].…”

Section: A Deep Autoencodersmentioning

confidence: 99%

“…For example, Pokorny et al [21] proposed a topological trajectory clustering by considering the relative persistent homology between motion trajectories. Our previous work [22] directly applied a common autoencoder to extract features of characterizing driving encounters for clustering, but the learned representations were not interpretable. To the best of our knowledge, great efforts on tackling a bunch of single trajectories have been made by utilizing off-theshelf algorithms, but no one efficient approach is suitable for clustering a set of multi-vehicle trajectories.…”

mentioning

confidence: 99%

Clustering of Driving Encounter Scenarios Using Connected Vehicle Trajectories

Wang

Ramesh

Zhu

et al. 2020

IEEE Trans. Intell. Veh.

Self Cite

View full text Add to dashboard Cite

Multi-vehicle interaction behavior classification and analysis offer in-depth knowledge to make an efficient decision for autonomous vehicles. This paper aims to cluster a wide range of driving encounter scenarios based only on multi-vehicle GPS trajectories. Towards this end, we propose a generic unsupervised learning framework comprising two layers: feature representation layer and clustering layer. In the layer of feature representation, we combine the deep autoencoders with a distancebased measure to map the sequential observations of driving encounters into a computationally tractable space that allows quantifying the spatiotemporal interaction characteristics of two vehicles. The clustering algorithm is then applied to the extracted representations to gather homogeneous driving encounters into groups. Our proposed generic framework is then evaluated using 2,568 naturalistic driving encounters. Experimental results demonstrate that our proposed generic framework incorporated with unsupervised learning can cluster multi-trajectory data into distinct groups. These clustering results could benefit decisionmaking policy analysis and design for autonomous vehicles.Index Terms-Multi-vehicle behavior clustering, autoencoder, vehicle trajectory, unsupervised learning.

show abstract

“…Based on a set of behaviours, manoeuvre-based motion prediction approaches employ estimation techniques, for instance Gaussian Processes [20,21] to then estimate most probable future manoeuvres. Recently, techniques such as deep-learning have also been applied to cluster vehicle encounters [22]. To compare a query and reference manoeuvres, manoeuvre-based techniques employ different distance metrics such as Hausdorff, Longest Common Subsequence, Euclidian distance, etc.…”

Section: Related Workmentioning

confidence: 99%

Predicting Agent Behaviour and State for Applications in a Roundabout-Scenario Autonomous Driving

Muhammad

Åstrand

2019

Sensors

View full text Add to dashboard Cite

As human drivers, we instinctively employ our understanding of other road users’ behaviour for enhanced efficiency of our drive and safety of the traffic. In recent years, different aspects of assisted and autonomous driving have gotten a lot of attention from the research and industrial community, including the aspects of behaviour modelling and prediction of future state. In this paper, we address the problem of modelling and predicting agent behaviour and state in a roundabout traffic scenario. We present three ways of modelling traffic in a roundabout based on: (i) the roundabout geometry; (ii) mean path taken by vehicles inside the roundabout; and (iii) a set of reference trajectories traversed by vehicles inside the roundabout. The roundabout models are compared in terms of exit-direction classification and state (i.e., position inside the roundabout) prediction of query vehicles inside the roundabout. The exit-direction classification and state prediction are based on a particle-filter classifier algorithm. The results show that the roundabout model based on set of reference trajectories is better suited for both the exit-direction and state prediction.

show abstract

Cluster Naturalistic Driving Encounters Using Deep Unsupervised Learning

Cited by 21 publications

References 22 publications

AI perspectives in Smart Cities and Communities to enable road vehicle automation and smart traffic control

AI perspectives in Smart Cities and Communities to enable road vehicle automation and smart traffic control

Clustering of Driving Encounter Scenarios Using Connected Vehicle Trajectories

Predicting Agent Behaviour and State for Applications in a Roundabout-Scenario Autonomous Driving

Contact Info

Product

Resources

About