Journal
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
Volume 509, Issue -, Pages 989-997Publisher
ELSEVIER SCIENCE BV
DOI: 10.1016/j.physa.2018.06.082
Keywords
Speed; Estimation; Traffic flow; Multiple kernel learning; Support vector regression
Categories
Funding
- China NSFC Program [61603257]
- National 863 Key Program [2012AA112307]
Ask authors/readers for more resources
Industrial loop detectors (ILDs) are the most common traffic detectors. In Shanghai, most of the ILDs are installed in a single loop way, which can detect various parameters, such as flow, saturation, and so on. However, they cannot detect the speed directly, which is one of the key inputs of intelligent transportation systems (ITS) for identifying the traffic state. Thus, this paper is dedicated to estimate speed accurately. It proposes a new algorithm that multiple kernel support vector regression (MKL-SVR) to complete this goal, which improves the accuracy and robustness of the speed estimation. Extensive experiments have been performed to evaluate the performances of MKL-SVR, compared with polynomial fitting, BP neural networks and SVR. All results indicate that the performances of MKL-SVR are the best and most robust. (C) 2018 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available