Optimal trajectory planning method and system for intelligent vehicle in continuous curve scene

Abstract

The invention discloses an optimal trajectory planning method and system for an intelligent vehicle in a continuous curve scene, and the method comprises the steps: 1, obtaining the coordinates of a road boundary point of a road where the vehicle is located and a center line point of the road through a sensor, setting a front enveloping circle and a rear enveloping circle with the radius being R in the front and rear directions of a longitudinal axis of the vehicle respectively, representing the vehicle through a front enveloping circle and a rear enveloping circle, and meanwhile, inwards shrinking road boundary points by a distance R in the direction perpendicular to the corresponding center line; step 2, fitting the road boundary points into N road boundaries which have mutually overlapped parts and are described by polynomial sum; and step 3, a bicycle kinematic model is used as a vehicle model of a self-vehicle, and meanwhile, sum is used as a road boundary constraint, a first optimal control problem is established in combination with the vehicle constraint, the optimal control problem is converted into a nonlinear programming problem, and a global optimal trajectory is solved.

Description

technical field [0001] The invention relates to the technical field of intelligent vehicle trajectory planning, in particular to a method and system for optimal trajectory planning of an intelligent vehicle in a continuous curve scene. Background technique [0002] At present, in the trajectory planning of intelligent vehicles on structured roads and continuous curves, most methods still use the road centerline or the basis function of known curves as a reference. In fact, when driving a vehicle artificially, skilled drivers do not will only drive according to the centerline. In the case of a one-way lane or no traffic ahead, the driver tends to take a shorter road, that is, to drive slightly closer to the inside of the curve when turning; in addition, the curve basis function has a fixed The expression is also far from the trajectory of human driving, and it is obviously not a road boundary. When the vehicle encounters an obstacle and needs to avoid it, in most cases the B...

AI Technical Summary

Problems solved by technology

When the vehicle encounters an obstacle and needs to avoid it, in most cases the Bezier curve or nonlinear MPC method is still used. Although the curve fitting method is faster in calculation, it also has the above problems. Non-linear MPC takes a long time to calculate and has poor real-time performance

Images