Automatic driving motion planning method based on parallel solution

Abstract

The invention relates to an automatic driving motion planning method based on parallel solution, and belongs to the technical field of automatic driving. The method comprises the following steps: S1, constructing a motion planning problem, and discretizing the motion planning problem; S2, converting the discretized motion planning problem into a parallel motion planning problem by introducing a global variable and a slack variable; and S3, for the constructed parallel motion planning problem, carrying out numerical solution by adopting a layered alternating direction multiplier method. According to the method, the optimization dimension and the constraint scale of the automatic driving motion planning problem are greatly reduced, the problem difficulty is simplified, and meanwhile, the carrying load of a single vehicle-mounted processor is shared to a plurality of parallel processors, so that the solving real-time performance of the automatic driving motion planning problem is remarkably improved.

Description

technical field [0001] The invention belongs to the technical field of automatic driving, and relates to a motion planning method for automatic driving based on parallel solution. Background technique [0002] The motion planning system is one of the key technologies in the automatic driving system. It can plan a collision-free and passable optimal driving path from the starting point to the ending point according to the road condition information, the driver's travel purpose and the global map database. [0003] According to whether there is a model structure, motion planning methods can be divided into model-free methods and model-based optimization methods. Model-free methods mainly include rule-based methods, artificial potential field methods, and sampling methods. Among them, the rule-based method cannot deal with new scenarios beyond the rules, the artificial potential field method lacks reasoning and prediction of vehicles and the environment, the sampling method fa...

AI Technical Summary

Problems solved by technology

[0004] However, as the traffic scene becomes more complex, the number of optimization indicators and nonlinear constraints in NMPC motion planning also increases, especially the increase of non-convex obstacle avoidance constraints, which greatly increases the complexity of solving motion planning problems

Moreover, the current solution method of NMPC motion planning is usually solved by a vehicle-mounted single-core processor. Due to the limited computing power of a single-core processor, this makes the real-time solution of NMPC motion planning more difficult.

Images