Distributed Nonlinear Model Predictive Control for Connected Vehicles Trajectory Tracking and Collision Avoidance with Ellipse Geometry

authored by: Mohamed Elsayed Hasan Abdelaal, Steffen Schön
Abstract: In the last decade, Model Predictive Control has drawn much attention from both academia and industry to be a successful guidance and control algorithm in autonomous driving because of its ability to handle complex nonlinear constrained systems, especially with the rapidly expanding technologies that enable fast implementation of its optimization problem. This paper considers Non-Cooperative Distributed Nonlinear Model Predictive Control (NMPC) for simultaneous trajectory tracking and collision avoidance of connected autonomous/semi-autonomous vehicles. The connected vehicles are considered as a Network Control System (NCS) of dynamically decoupled agents with only coupling constraints, and it is formulated as a distributed Optimal Control Problem (OCP). It is assumed that the connected vehicles have a communication link to exchange their intentions. The coordination among the agents is achieved by Priority-Based techniques where a priority is assigned for each one to satisfy the prediction consistency. For each agent, a nonlinear bicycle model is used to predict a sequence of the states and then optimize it with respect to a sequence of control inputs. The objective function of the OCP is to track the planned trajectory. In order to achieve a normal driving behavior, comfort driving, and provide consistency of the simplified kinematic model with the actual complex vehicle model, constraints are added to the control inputs and their rate of change. In order to achieve collision avoidance among the networked vehicles, a geometric shape must be selected to represent the vehicle geometry. In this paper, an elliptic disk is selected for that as it represents the geometry of the vehicle better that the traditional circular one. A separation condition between each pair of elliptic disks is formulated as time-varying state constraints for the OCP, and is proved by developing sufficient conditions for the separation. The algorithm is validated using MATLAB simulation with the aid of ACADO toolkit.
Organisation(s): Faculty of Civil Engineering and Geodetic Science
Institute of Geodesy
Type: Conference contribution
Pages: 2100-2111
No. of pages: 12
Publication date: 2019
Publication status: Published
ASJC Scopus subject areas: Software, Communication, Information Systems, Electrical and Electronic Engineering, Computer Networks and Communications, Computer Science Applications
Electronic version(s): https://doi.org/10.33012/2019.16911 (Access: Closed)
: Details in the research portal "Research@Leibniz University"

BibTeX

@inproceedings{1abcfd3685f84969a2036fd1ca6c0090,
title = "Distributed Nonlinear Model Predictive Control for Connected Vehicles Trajectory Tracking and Collision Avoidance with Ellipse Geometry",
abstract = "In the last decade, Model Predictive Control has drawn much attention from both academia and industry to be a successful guidance and control algorithm in autonomous driving because of its ability to handle complex nonlinear constrained systems, especially with the rapidly expanding technologies that enable fast implementation of its optimization problem. This paper considers Non-Cooperative Distributed Nonlinear Model Predictive Control (NMPC) for simultaneous trajectory tracking and collision avoidance of connected autonomous/semi-autonomous vehicles. The connected vehicles are considered as a Network Control System (NCS) of dynamically decoupled agents with only coupling constraints, and it is formulated as a distributed Optimal Control Problem (OCP). It is assumed that the connected vehicles have a communication link to exchange their intentions. The coordination among the agents is achieved by Priority-Based techniques where a priority is assigned for each one to satisfy the prediction consistency. For each agent, a nonlinear bicycle model is used to predict a sequence of the states and then optimize it with respect to a sequence of control inputs. The objective function of the OCP is to track the planned trajectory. In order to achieve a normal driving behavior, comfort driving, and provide consistency of the simplified kinematic model with the actual complex vehicle model, constraints are added to the control inputs and their rate of change. In order to achieve collision avoidance among the networked vehicles, a geometric shape must be selected to represent the vehicle geometry. In this paper, an elliptic disk is selected for that as it represents the geometry of the vehicle better that the traditional circular one. A separation condition between each pair of elliptic disks is formulated as time-varying state constraints for the OCP, and is proved by developing sufficient conditions for the separation. The algorithm is validated using MATLAB simulation with the aid of ACADO toolkit.",
author = "Abdelaal, {Mohamed Elsayed Hasan} and Steffen Sch{\"o}n",
note = "Funding Information: This work was supported by the German Research Foundation (DFG) as a part of the Research Collaboration in dynamic SENSor networks (i.c.sens) [GRK2159].; 32nd International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS+ 2019 ; Conference date: 16-09-2019 Through 20-09-2019",
year = "2019",
doi = "10.33012/2019.16911",
language = "English",
series = "Proceedings of the Satellite Division's International Technical Meeting",
publisher = "Institute of Navigation",
pages = "2100--2111",
booktitle = "Proceedings of the 32nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2019)",
}