Skip to content

Quintic G2-Splines for Trajectory Planning of Autonomous Vehicles

Authors: Aurelio Piazzi, Corrado Guarino Lo Bianco

Published: 2000 (Conference Paper)

Source: IEEE Intelligent Vehicles Symposium (IV)

Algorithm: Quintic G2-Spline

DOI: 10.1109/IVS.2000.898341

Summary

Proposes a steering method for connecting pairs of states in [x, y, yaw, curvature] state space using paths that are quintic polynomials in x, y position coordinates. This leaves four free tuning parameters eta1, eta2, eta3, eta4, which influence the shape of the paths and can be set according to heuristics or can be explicitly optimized, which is investigated in the companion work by the same authors "Optimal trajectory planning with quintic G2-splines".

Abstract

Presents a motion planning primitive for car-like vehicles. It is a completely parametrized quintic spline, denoted as eta-spline, that allows interpolation of an arbitrary sequence of points with overall second order geometric G2-continuity. Issues such as minimality, regularity, symmetry, and flexibility of these G2-splines are addressed in the exposition. The development of the new primitive is tightly connected to the flatness based control of nonholonomic car-like vehicles.

Primary

Paper researchgate.net

Standard

DOI 10.1109/IVS.2000.898341

Alternate

ieeexplore.ieee.org ieeexplore.ieee.org

Tags

  • Path planning

  • Motion planning

  • Splines

  • G2 continuity

  • Autonomous vehicles

  • Nonholonomic systems

  • Flatness-based control

  • Differential flatness

  • Curvature continuity