A Time-Consistent Local Path Planning Method for Unmanned Vehicles

Abstract

The invention discloses a local path planning method for an unmanned vehicle with time consistency. Firstly, vehicle motion differential constraints are added to the planned path curve, so that the vehicle can accurately track the generated trajectory; secondly, the path planning trajectory curve model The model parameters have been improved, which adds more degrees of freedom to the planning path, and facilitates the transformation of the curvature constraint problem into a mathematical solution optimization problem; finally, there is a comparison between the cycle optimization results before and after the partial path method obtained by using the cycle optimization solution. large difference, which leads to the problem of jitter when the vehicle performs real-time path tracking. The present invention divides the local path planning section into two parts: weak planning and strong planning. The weak planning section is obtained from the strong planning result of the previous planning cycle. segment, which not only ensures the time consistency of the planning results, but also enables the local path planning to have a higher obstacle avoidance ability.

Description

technical field [0001] The invention belongs to the related research field of local path planning of intelligent vehicles, and relates to a method for planning local paths of unmanned vehicles with time consistency. [0002] technical background [0003] According to statistics, in 2016, my country's domestic car sales reached 28.028 million, an increase of 13.7% compared to 2015, ranking first in the world for eight consecutive years. By the end of 2015, the total mileage across the country had reached 4,577,300 kilometers, a year-on-year increase of 145,000 kilometers. The increase in car ownership and the total mileage of roads has led to frequent traffic accidents and serious traffic congestion. According to statistics from the Traffic Management Bureau of the Ministry of Public Security, there were about 165,000 car accidents nationwide in 2016, resulting in about 52,000 deaths and 168,000 injuries. Analysis of urban traffic data by experts shows that human factors are...

AI Technical Summary

Problems solved by technology

[0007] For the current local path planning methods based on path generation and path selection for unmanned vehicles, the planned curves lack vehicle kinematics constraints, cannot guarantee the accuracy and safety of vehicle path tracking, and rely on optimization solutions to obtain The poor cycle consistency of the local planning path method can easily lead to a step in the planning result that affects the safe driving of the vehicle. A time-consistent local path planning method for unmanned vehicles is proposed.

Images