A Vehicle Obstacle Avoidance Trajectory Planning Method Based on Gaussian Pseudospectral Method

Abstract

The invention belongs to the technical field of vehicle obstacle avoidance trajectory planning, in particular to a vehicle obstacle avoidance trajectory planning method based on the Gaussian pseudospectral method; It is an optimal control problem, which includes vehicle kinematics and dynamics constraints, initial and final state constraints and path constraints, with the optimal obstacle avoidance time as the performance index, and then, based on the Gaussian pseudospectral method, the optimal control problem Carry out discretization processing to transform it into a nonlinear programming problem satisfying constraints, and obtain the optimal obstacle avoidance trajectory by solving the nonlinear programming problem.

Description

technical field [0001] The invention belongs to the technical field of vehicle obstacle avoidance trajectory planning, in particular to a vehicle obstacle avoidance trajectory planning method based on a Gaussian pseudospectral method. Background technique [0002] At present, with the continuous increase of car ownership, there are a series of urgent problems such as road congestion and frequent traffic accidents. Among them, vehicle collisions cause a considerable proportion of major traffic accidents. Therefore, obstacle avoidance trajectory planning As a key part of self-driving vehicles, it has received a lot of attention and in-depth research from the academic community. [0003] At present, the artificial potential field method and the virtual force field method are two traditional methods for obstacle avoidance trajectory planning. Among them, the trajectory planned by the artificial potential field method is smooth and safe. However, this algorithm is easy to fall in...

AI Technical Summary

Problems solved by technology

[0003] At present, the artificial potential field method and the virtual force field method are two traditional methods for obstacle avoidance trajectory planning. Among them, the trajectory planned by the artificial potential field method is smooth and safe. However, this algorithm is easy to fall into a local minimum when solving. However, The virtual force field method is an obstacle avoidance trajectory planning algorithm that combines the grid method and the artificial potential field method. It has great advantages in uncertain dynamic environments, but it is difficult to apply in complex traffic scenarios.

[0004] In addition, intelligent optimization algorithms have also been developed in obstacle avoidance trajectory planning, but each algorithm has its own advantages and disadvantages. For example, although the rapid expansion random number algorithm has a faster search speed, the randomness of the algorithm makes the planned path The curvature changes too much, even a small range of right angle changes, resulting in an unsmooth path; and the genetic algorithm needs to be based on a large amount of driver experience, so it has the disadvantages of being unable to learn and having poor flexibility

Images