Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Multi-object adaptive cruise control

Author(s):

R. Möbus
Conference/Journal:

no. 17742
Abstract:

In this thesis the development and implementation of a multi-object adaptive cruise control (ACC) system is presented. A sensor fusion configuration as well as object tracking and sensor fusion algorithms are presented to obtain a thorough representation of the traffic scene ahead of an ACC-controlled vehicle. The sensor fusion configuration includes a 77GHz radar sensor and an IR laser sensor for object detection. A monocular CCD camera system is employed for lane recognition and the lane assignment of the detected objects. Experimental results of all presented algorithms are given.

The control model and the control objectives of a multi-object ACC system are presented. The multi-object ACC problem is looked at as a constrained optimal control problem incorporating the dynamics of the traffic scene, the driver’s desire to cruise at a certain velocity, the lane assignment of the other road users, the objective of respecting certain minimum distances to other road users and to adapt the velocity to the flow of the other road users. Additionally, overtaking a preceding vehicle on the right can be avoided. The choice of the relevant object is implicitly determined by the cost function and the optimization criteria. Constraints imposed by physical limitations as well as by comfort and safety considerations can be included and a receding horizon control strategy is applied.

In a first step the constrained finite time optimal control (CFTOC) problem with a piecewise affine (PWA) system description is solved via dynamic programming. The resulting explicit solution is tested in a research vehicle and it is shown that real-time operation and most control objectives of the multi-object ACC problem can be achieved.

In a second step, the multi-object ACC problem is looked at as a CFTOC problem with a mixed logical dynamical (MLD) system description. With an efficient way to represent and evaluate the explicit solution to the corresponding multi-parametric mixed integer quadratic program, it is possible to include all desired control objectives in the problem formulation and still obtain an explicit solution suitable for real-time operation. Simulation results of this multi-object ACC control approach are presented and the controller is compared to a reference ACC controller. With the efficient controller representation the multi-object ACC controller is implemented on the ECU of a standard production platform vehicle to confirm the simulation results in real traffic.

Year:

2008
Type of Publication:

(03)Ph.D. Thesis
Supervisor:



File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhDThesis { Xxx:2008:IFA_3063,
    author={R. M{\"o}bus},
    title={{Multi-object adaptive cruise control}},
    school={},
    year={2008},
    address={},
    month=apr,
    url={http://control.ee.ethz.ch/index.cgi?page=publications;action=details;id=3063}
}
Permanent link