×

Path planning

  • US 8,428,820 B2
  • Filed: 12/07/2006
  • Issued: 04/23/2013
  • Est. Priority Date: 12/09/2005
  • Status: Active Grant
First Claim
Patent Images

1. A method of computing the path of a motor vehicle along a two-dimensional plane required to control the motion of the motor vehicle by means of a driver-assisting system, comprising:

  • providing a path connecting an initial state with a terminal state wherein each of states Zi=[xi, yi, ψ

    i, ci] along the path is characterized by four state coordinates with xi and yi being the Cartesian coordinates of a point Pi, ψ

    i being the direction angle of the tangent, and ci being the curvature of the path at point Pi;

    assembling the path from a number of elementary paths joining at junction points JPi such that the four state coordinates of state Zi have a continuous transition in the junction points JPi;

    assigning the path at least one isosceles planar triangle with vertices Δ

    (P0, P3, P5), where P0 is the starting position and P3 is terminal position of at least a section of the path;

    determining where point P5 is located opposite the base of the planar triangle P0P3| by the intersection of the two legs P0P5 and P5P3 of the planar triangle such that the two legs P0P5, P5P3, are of the same length and, together with side P0P3| of the planar triangle, form the same vertical angle ψ

    i, with angle ψ

    i at vertex P0 of the planar triangle by the angle formed by the tangent of the path and side P0P3| of the triangle;

    determining the parameters of the elementary paths by the parameters of the planar triangle; and

    signaling a control setting of a vehicle steering mechanism according to said determinations.

View all claims
  • 2 Assignments
Timeline View
Assignment View
    ×
    ×