Model Predictive Path Integral Control for Car Driving with Dynamic Cost Map

Alexander Buyval, Aidar Gabdullin, Alexander Klimchik

2018

Abstract

Path planning in a complex dynamic environment is one of the key subsystems in an autonomous vehicle. This paper presents an extension of Model Predictive Path Integral (MPPI) control method which is able to take moving objects into account while path planning and driving. To obtain real-time performance, cost map update with respect to dynamic objects both as basic MPPI is implemented as a set of concurrent processes using CUDA technology. The algorithm’s performance is demonstrated on a model of a stock car in a simulation environment.

Paper Citation

in Harvard Style

Klimchik A. (2018). Model Predictive Path Integral Control for Car Driving with Dynamic Cost Map.In Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO, ISBN 978-989-758-321-6, pages 248-254. DOI: 10.5220/0006901702480254

in Bibtex Style

@conference{icinco18,
author={Alexander Klimchik},
title={Model Predictive Path Integral Control for Car Driving with Dynamic Cost Map},
booktitle={Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,},
year={2018},
pages={248-254},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006901702480254},
isbn={978-989-758-321-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,
TI - Model Predictive Path Integral Control for Car Driving with Dynamic Cost Map
SN - 978-989-758-321-6
AU - Klimchik A.
PY - 2018
SP - 248
EP - 254
DO - 10.5220/0006901702480254