Learning To Predict Driver Behavior From Observation
Abstract
This paper focuses on modeling and predicting human driving behavior, with the long term goal of anticipating the behavior of the driver before dangerous situations occur. We formulate this problem as a Learning from Demonstration problem, and show how standard supervised learning methods do not perform well in this task. The main contribution of this paper is a new approach we call indirect prediction. The key idea of indirect prediction is not to predict the behavior directly, but rather to build a model that predicts how certain features of the world state will change over time, and then use those to predict the necessary behavior in order to achieve those changes. We show how this apparently counterintuitive idea directly addresses one of the key reasons for which supervised learning does not perform well for LfD. In addition, we show how using ideas from context-based reasoning can also improve the accuracy of behavior modeling.
Publication Date
1-1-2017
Publication Title
AAAI Spring Symposium - Technical Report
Volume
SS-17-01 - SS-17-08
Number of Pages
506-512
Document Type
Article; Proceedings Paper
Personal Identifier
scopus
Copyright Status
Unknown
Socpus ID
85028734754 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/85028734754
STARS Citation
Ontañón, Santiago; Lee, Yi Ching; Snodgrass, Sam; Winston, Flaura K.; and Gonzalez, Avelino J., "Learning To Predict Driver Behavior From Observation" (2017). Scopus Export 2015-2019. 7426.
https://stars.library.ucf.edu/scopus2015/7426