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.


Path Planning and Control for Extreme Maneuvers in Autonomous Vehicles


E. Jelavic

Master Thesis, HS15 (10459)

Most of the people will associate the term "extreme driving" with driving at the friction and handling limits. Colloquially, these maneuvers are often referred to as "drifting". Professional drivers, such as rally or stunt drivers, deliberately bring a vehicle into this mode of operation where tires are said to be saturated (operating at friction limit). Unconventional control inputs such as weight transfer or handbrake are often used to control the vehicle in this mode of operation. By saturating the tires (especially rear tires), professional drivers are exploiting control possibilities that are beyond the skill of an average driver and have not been harnessed in automotive control design yet. Characteristic for extreme maneuvers are high lateral forces and high yaw rates. Through understanding of the principles that professional drivers use to generate the extreme trajectories and understanding how to track these trajectories, someday it might be possible to utilize extreme driving techniques in collision avoidance systems of the autonomous vehicles.

In this thesis, a systematic approach to extreme driving maneuvers is presented. Considerations and results presented in this thesis refer to a four-wheeled vehicle (e.g. a car). The task of performing an extreme maneuver is broken down into two steps: model-based trajectory planning and trajectory following. Firstly, an investigation on required model complexity is conducted. An approach for trajectory planning with high order models is given. This path planning approach is based on a set of rules describing an extreme maneuver. A control approach for following trajectories containing extreme maneuvers has been developed and its effectiveness has been assessed in a simulation. Control approach works on switched control principle; it switches between actively controlling the system using state feedback and controlling the system in the open-loop fashion by merely executing pre-recorded commands without any feedback. Additionally, an offthe- shelf Radio-Controlled car (RC car) has been modified and transformed into an experimental platform. Control algorithm was tested on the RC car performing a sliding parallel park into a tight parking spot. Guidelines for possible future work are given.

Supervisors: Francesco Borrelli (UCB), Manfred Morari


Type of Publication:

(12)Diploma/Master Thesis

M. Morari

File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2016:IFA_5437
Permanent link