Maneuver Optimization for Autonomous Vehicles using Trajectory Templates

Date

2024-8-26

Author

Arslan, Ömer Faruk

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

466
views

0
downloads

Cite This

Precise and efficient maneuvering of autonomous driving systems is paramount to ensure safety, comfort, and compliance with traffic regulations. In particular, there are various important regular navigation tasks, including lane keeping, lane changes, turns at intersections, or merging on highways. In addition, there are several safety-critical maneuvers, such as avoiding obstacles or overtaking a slow vehicle. Each of these maneuvers requires the real-time computation of safe and efficient vehicle trajectories. This thesis addresses the challenge of optimizing maneuvers of autonomous vehicles through the innovative use of trajectory templates. Specifically, we employ bi-elementary paths constructed using clothoids to represent avoidance and overtake maneuvers. Clothoids have the advantage of a linearly changing curvature, which makes them easily trackable by autonomous vehicles. Moreover, bi-elementary paths are parametrized by a few parameters and have an accurate analytical approximation in the form of arc-splines. This allows the analytical evaluation of geometric and kinematic constraints posed on the desired trajectories and, hence, effectively reduces the computational effort of the trajectory optimization. The thesis illustrates the benefits of the proposed method by comprehensive simulation experiments with static and dynamic obstacles.

Subject Keywords

Maneuver Optimization, Avoidance Maneuver, Overtake Maneuver, Recovery Maneuvers, Bi-Elementary Paths, Clothoids, Optimization, Bi-section Search

URI

https://hdl.handle.net/11511/111409

Collections

Graduate School of Natural and Applied Sciences, Thesis