A motion planning method for unmanned vehicles based on multi-objective optimization

Abstract

The invention discloses a motion planning method for unmanned vehicles based on multi-objective optimization, which maps the vehicle and the environment from the Cartesian coordinate system to the Frenet coordinate system; establishes a mathematical model for the multi-objective path planning problem of the unmanned vehicle; uses linear The dynamic programming method is used for path planning; the trajectory is described by a piecewise quintic polynomial, with the minimum trajectory slope, minimum curvature, comfortable riding experience, and the closest distance to the path obtained by linear dynamic programming as the optimization goals, and the piecewise quintic polynomial is used to connect the points. The position, first-order derivative, second-order derivative, and third-order derivative are used as equality constraints, and the road natural boundary constraints and obstacle boundary constraints are used as inequality constraints to establish a mathematical model for the multi-target trajectory generation problem of unmanned vehicles; Optimal solution of vehicle multi-target trajectory generation problem. The invention solves the problem that the path obtained by the path planning method based on random scattered points is difficult to conform to the constraints of vehicle kinematics.

Description

technical field [0001] The invention belongs to the technical field of intelligent driving and its planning, and in particular relates to a motion planning method for unmanned vehicles based on multi-objective optimization. Background technique [0002] Since its introduction in 1886, the automobile field has developed rapidly, and its technological development and popularization have become an important symbol to measure the degree of modernization and civilization of a country or region. However, with the popularity of cars and the exponential growth of ownership, people are also suffering from serious annoyances caused by worsening traffic jams and accidents. Therefore, intelligent driving technology has received people's attention. In the urban road environment, the driving conditions are relatively simple, and there are maps with higher precision, so it is easier to realize intelligent driving. The 2017 national key research and development plan has clearly listed the...

AI Technical Summary

Problems solved by technology

However, the traditional path planning method based on randomly scattered points, such as RRT, does not satisfy the vehicle kinematics and dynamics constraints; the traditional trajectory generation method based on curve fitting transforms the multi-objective weighting in the multi-objective optimization problem Target optimization problem, the weight is difficult to determine

Images