Functional discretization of space using Gaussian processes for road intersection crossing
Authors:
- 1 November 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 156-162
Abstract
A framework for the discretization of navigable space within and around a cross intersection is proposed in this paper. The purpose of our approach is to capture the manner in which drivers manoeuvre in an intersection in order to facilitate and understand the decision-making tasks. Gaussian processes are used to learn and predict the most likely trajectories taken by multiple drivers in different situations. The merging and crossing areas are found by searching for the overlap between two predicted trajectories, whereas the area approaching the intersection is discretized by using the most probable occupancy. The generated areas are stored in the vehicle navigation map. The correlation between the proposed discretization and the driver's behaviour is demonstrated. The proposed framework enables also the discretization of the vehicle velocity profile, information that can be used to govern the decision-making function.Keywords
This publication has 8 references indexed in Scilit:
- Towards autonomous navigation of unsignalized intersections under uncertainty of human driver intent2015
- Probabilistic road context inference for autonomous vehicles2015
- Hybrid sampling Bayesian Occupancy Filter2014
- Lanelets: Efficient map representation for autonomous driving2014
- Probabilistically safe motion planning to avoid dynamic obstacles with uncertain motion patternsAutonomous Robots, 2013
- Driver Behavior Classification at Intersections and Validation on Large Naturalistic Data SetIEEE Transactions on Intelligent Transportation Systems, 2012
- A probabilistic model for estimating driver behaviors and vehicle trajectories in traffic environments2010
- Human Driver Model and Driver Decision Making for Intersection Driving2007 IEEE Intelligent Vehicles Symposium, 2007