Vehicle Behavior Prediction and Generalization Using Imbalanced Learning Techniques
(2021) 24th IEEE International Intelligent Transportation Systems Conference (ITSC2021) p.2003-2010- Abstract
- The use of learning-based methods for vehicle behavior prediction is a promising research topic. However, many publicly available data sets suffer from class distribution skews which limits learning performance if not addressed. This paper proposes an interaction-aware prediction model consisting of an LSTM autoencoder and SVM classifier. Additionally, an imbalanced learning technique, the multiclass balancing ensemble is proposed. Evaluations show that the method enhances model performance, resulting in improved classification accuracy. Good generalization properties of learned models are important and therefore a generalization study is done where models are evaluated on unseen traffic data with dissimilar traffic behavior stemming from... (More)
- The use of learning-based methods for vehicle behavior prediction is a promising research topic. However, many publicly available data sets suffer from class distribution skews which limits learning performance if not addressed. This paper proposes an interaction-aware prediction model consisting of an LSTM autoencoder and SVM classifier. Additionally, an imbalanced learning technique, the multiclass balancing ensemble is proposed. Evaluations show that the method enhances model performance, resulting in improved classification accuracy. Good generalization properties of learned models are important and therefore a generalization study is done where models are evaluated on unseen traffic data with dissimilar traffic behavior stemming from different road configurations. This is realized by using two distinct highway traffic recordings, the publicly available NGSIM US-101 and I80 data sets. Moreover, methods for encoding structural and static features into the learning process for improved generalization are evaluated. The resulting methods show substantial improvements in classification as well as generalization performance. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/f88f18de-1418-425f-9e01-c26ab51a1d2d
- author
- Westny, Theodor ; Frisk, Erik and Olofsson, Björn LU
- publishing date
- 2021
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- 24th IEEE International Intelligent Transportation Systems Conference (ITSC)
- pages
- 2003 - 2010
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- 24th IEEE International Intelligent Transportation Systems Conference (ITSC2021)
- conference location
- Indianapolis, United States
- conference dates
- 2021-09-19 - 2021-09-22
- external identifiers
-
- scopus:85118442722
- DOI
- 10.1109/ITSC48978.2021.9564948
- project
- ELLIIT B14: Autonomous Force-Aware Swift Motion Control
- RobotLab LTH
- language
- English
- LU publication?
- no
- id
- f88f18de-1418-425f-9e01-c26ab51a1d2d
- alternative location
- https://arxiv.org/abs/2109.10656
- date added to LUP
- 2022-07-04 15:42:03
- date last changed
- 2023-04-24 21:09:12
@inproceedings{f88f18de-1418-425f-9e01-c26ab51a1d2d, abstract = {{The use of learning-based methods for vehicle behavior prediction is a promising research topic. However, many publicly available data sets suffer from class distribution skews which limits learning performance if not addressed. This paper proposes an interaction-aware prediction model consisting of an LSTM autoencoder and SVM classifier. Additionally, an imbalanced learning technique, the multiclass balancing ensemble is proposed. Evaluations show that the method enhances model performance, resulting in improved classification accuracy. Good generalization properties of learned models are important and therefore a generalization study is done where models are evaluated on unseen traffic data with dissimilar traffic behavior stemming from different road configurations. This is realized by using two distinct highway traffic recordings, the publicly available NGSIM US-101 and I80 data sets. Moreover, methods for encoding structural and static features into the learning process for improved generalization are evaluated. The resulting methods show substantial improvements in classification as well as generalization performance.}}, author = {{Westny, Theodor and Frisk, Erik and Olofsson, Björn}}, booktitle = {{24th IEEE International Intelligent Transportation Systems Conference (ITSC)}}, language = {{eng}}, pages = {{2003--2010}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, title = {{Vehicle Behavior Prediction and Generalization Using Imbalanced Learning Techniques}}, url = {{http://dx.doi.org/10.1109/ITSC48978.2021.9564948}}, doi = {{10.1109/ITSC48978.2021.9564948}}, year = {{2021}}, }