United States Department of Transportation
i
Prediction and Behaviors for Driver Assistance and Socially Cooperative Autonomous Driving
-
2018-10-31
-
[PDF-1.59 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Resource Type:
-
Geographical Coverage:
-
Corporate Publisher:
-
Abstract:The goal of this project was to develop better prediction and behaviors generation methods with application to driver assistance and autonomous driving. Prediction of the behavior of other cars in traffic allows both driver assistance and autonomous driving systems to better plan their immediate future behavior. The resultant behaviors can then be more “socially cooperative”, i.e., they can increase the predictability, fluidity, and safety of inter-car interactions. This report presents work on two different scenarios: 1) roundabouts or traffic circles, in which drivers need to gauge the behaviors of cars in the traffic circle in order to enter and exit safely; 2) lane change situations, in which prediction of whether surrounding cars will change lanes and if so, what their likely trajectory will be, can improve the smoothness and safety of egovehicle motion. The report first presents the roundabout/traffic circle work, then appends a conference paper published at IROS (IEEE International Conference on Robots and Systems) 2017 summarizing the lane change prediction work.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
