Fuzzy Logic-Based Decision Models for Mandatory Lane Changes on Freeways

This research performs a comparative analysis of two fuzzy logic-based methodologies for making binary decisions for mandatory lane changes on freeways. A survey was conducted to determine the most frequently used parameters when drivers decide if they could begin a mandatory lane change. Four of the most frequently used parameters were selected. Next, mandatory lane changes in the NGSIM’s U.S. Highway 101 vehicle trajectory data were identified, and the four parameters were extracted for each mandatory lane changing occurrence. The extracted data were partitioned into training and test data sets. In this research, three lane changing decision models, based on two methodologies, were compared. They were: (1) an existing Fuzzy Inference System (FIS) originally developed for discretionary lane changes; (2) the same FIS but with an internal threshold, adapted to obtain the best performance for mandatory lane changes; and (3) a newly developed Adaptive Neuro Fuzzy Inference System (ANFIS). Based on the test results, the ANFIS is recommended, as it gave a higher overall correct decision rate compared to the FIS.

  • Supplemental Notes:
    • This paper was sponsored by TRB committee ABJ70 Standing Committee on Artificial Intelligence and Advanced Computing Applications.
  • Corporate Authors:

    Transportation Research Board

  • Authors:
    • Vechione, Matthew
    • Cheu, Ruey Long
  • Conference:
  • Date: 2019


  • English

Media Info

  • Media Type: Digital/other
  • Features: References;
  • Pagination: 4p

Subject/Index Terms

Filing Info

  • Accession Number: 01697439
  • Record Type: Publication
  • Report/Paper Numbers: 19-00318
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 7 2018 9:27AM