University of Michigan-Dearborn, USA-Yi Lu Murphey
Wuhan University of Technology, China-Honghui Zhu
by SAE International in United States
Lane changes are stressful maneuvers for drivers, particularly during high-speed traffic flows. However, modeling driver’s lane-changing decision and implementation process is challenging due to the complexity and uncertainty of driving behaviors. To address this issue, this article presents a personalized Lane-Changing Model (LCM) for Advanced Driver Assistance System (ADAS) based on deep learning method. The LCM contains three major computational components. Firstly, with abundant inputs of Root Residual Network (Root-ResNet), LCM is able to exploit more local information from the front view video data. Secondly, the LCM has an ability of learning the global spatial-temporal information via Temporal Modeling Blocks (TMBs). Finally, a two-layer Long Short-Term Memory (LSTM) network is used to learn video contextual features combined with lane boundary based distance features in lane change events. The experimental results on a -world driving dataset show that the LCM is capable of learning the latent features of lane-changing behaviors and achieving significantly better performance than other prevalent models.
This content contains downloadable datasets
Annotation ability available
A list of downloadable items.
Please Wait. Loading Available Formats...
This modal will allow the user to download citation information or export specific fields of data into one of three file formats: Excel, comma-separated values, or text.
Sorry, this document is already saved to a project folder.
Share this content via social media.
Share Via Email
Email sent successfully!
Use this modal to save a search.
The current search and criteria have already been saved. Please modify your criteria to save as a new search.
Request Sent successfully!
Search Tips: Search Operators:AND, OR, NOT; Wild Cards; Using Quotes
SAE MOBILUS® Search Tips
Please note: All terms not contained within quotes, will be evaluated using the OR operator. This means that your search results will contain all results that contain ANY of the entered terms.
AND, OR, NOT
SAE MOBILUS supports the use of the AND, OR and NOT BOOLEAN operators ONLY function with the Advanced Search.
Using the asterisk, ( * ) allows you to search for a partial word. For example, entering a keyword search of aero* will search for any word beginning with aero.
Using the question mark, ( ? ) allows you to search for a word with a single character wildcard. For example, searching for t?re would match both tire and tyre
Words in double quotes " " are treated as a phrase. Conducting a search for "heat model" will return all results in which the exact phrase "heat model" is found, while "heat modeling" would NOT be included.
A term combined with a phrase will be joined with an OR operator.
Searching for Standards
When searching for a standard please use the entire Document Number assigned, which includes the abbreviation. A space can be used, or omitted between the abbreviation and number.