Automotive Tracking Technique Using a New IMM Based PDA-SVSF Conferences

abstract

• Car tracking algorithms are important for a number of applications, including self-driving cars and vehicle safety systems. The probabilistic data association (PDA) algorithm, in conjunction with Kalman Filter (KF), and interacting multiple model (IMM) are well studied, specifically in the aero-tracking applications. This paper studies single targets while performing maneuvers in the presence of clutter, which is a common scenario for road vehicle tracking applications. The relatively new smooth variable structure filter (SVSF) is demonstrated to be robust and stable filtering strategy under the presence of modeling uncertainties. In this paper, SVSF based PDA technique is combined with IMM method. The new method, referred to as IMM-PDA-SVSF is simulated under several possible car motion scenarios. Also, the algorithm is tested on a real experimental data acquired by GPS device.

authors

• Attari, Mina
• Habibi, Saeid

status

• published 

publication date

• November 14, 2014

published in

• Volume 12: Transportation Systems  Journal

presented at event

• ASME 2014 International Mechanical Engineering Congress and Exposition  Conference

keywords

• ALGORITHM
• Engineering
• Engineering, Mechanical
• PROBABILISTIC DATA ASSOCIATION
• Science & Technology
• Technology
• Transportation
• Transportation Science & Technology
• automotive tracking
• interacting multiple model
• probabilistic data association
• smooth variable structure filter

Digital Object Identifier (DOI)

• 10.1115/imece2014-36412

International Standard Book Number (ISBN) 13

• 978-0-7918-4961-3

volume

• 12