Automatic driving trajectory generation system in high-speed scene

Abstract

The invention relates to an automatic driving trajectory generation system in a high-speed scene, including a sensor group, an execution system, a planning subsystem, a trajectory screening module, and a trajectory optimization module; the sensor group collects information and transmits it to the sensor fusion module; the sensor fusion module builds two dimensional coordinate system, get the original road data and original vehicle state data, and send them to the behavior planning subsystem; get the collection of sampling points, and send them to the trajectory screening module; the trajectory screening module generates candidate trajectories; the candidate trajectories are composed of sampling points; the trajectory screening module one by one , calculate the cost value between the sampling points, filter out the best path data, and send it to the trajectory optimization module; the trajectory optimization module obtains the optimal smooth curve, and sends it to the execution system; the execution system controls the vehicle to enter the end point along the target trajectory. The invention reduces the amount of calculation, improves the path planning speed, and ensures the real-time output; the curvature of the planned path is continuous, and the safety and operability are high; the optimal solution can be obtained.

Description

technical field [0001] The invention relates to the technical field of unmanned vehicles, in particular to an automatic driving trajectory generation system in high-speed scenarios. Background technique [0002] The path planning of unmanned vehicles depends on the decision-making and planning system in the ECU. In the complex and changeable road environment, the corresponding driving behavior instructions are made in real time according to the received environmental information, which includes the trajectory planning work. The trajectory is a series of trajectory curves of the unmanned vehicle from the current state to the next target state. Trajectories need to satisfy the vehicle's motion and dynamic constraints to avoid collisions with obstacles in a realistic urban environment. [0003] Most of the current mainstream trajectory generation systems use algorithms based on curve interpolation and search algorithms based on sampling. [0004] The former has the advantages...

AI Technical Summary

Problems solved by technology

get the optimal solution

Images