A planning method for vehicle collision avoidance trajectories under pedestrian crossing conditions

Abstract

The invention discloses a method for planning vehicle collision avoidance trajectories under pedestrian crossing conditions. When a pedestrian crossing the street suddenly appears in front of the vehicle, a series of collision avoidance candidate trajectories are planned, specifically including spatial path planning and temporal movement. planning; and then transmit the candidate trajectory to the optimization module to perform multi-objective optimization with safety, stability and traffic efficiency indicators to obtain an optimal trajectory. The invention plans a safe, stable and efficient trajectory for the vehicle in real time under the pedestrian crossing condition to realize active collision avoidance, and can ensure that the vehicle avoids obstacles safely, stably and efficiently.

Description

technical field [0001] The invention relates to an advanced auxiliary driving technology, in particular to a planning method for vehicle collision avoidance trajectories under pedestrian crossing conditions. Background technique [0002] In recent decades, more and more practitioners have invested in the research of intelligent driving, and high-tech companies such as Google have joined forces with Tesla, Volvo and other global car companies to develop advanced assisted driving systems such as active collision avoidance systems. The field has achieved great results. However, as one of the important components of ADAS, the active collision avoidance system needs to deal with all kinds of uncertain factors brought by all complex traffic scenes in the real traffic environment so as to determine a trajectory that can ensure the safety of various traffic bodies. there's still a long way to go. Traditional active collision avoidance systems are mainly aimed at obstacles that are...

AI Technical Summary

Problems solved by technology

[0003] As two agents in the traffic environment, on the one hand, pedestrians have strong mobility, and their decision-making is affected by the motion state of surrounding vehicles, making it difficult to predict their future state accurately; Many uncertain factors, which make the driver's collision avoidance decision confusing

This makes it difficult for the vehicle-to-vehicle collision avoidance system to be directly used for pedestrian collision avoidance

[0004] In addition, trajectory planning technology is one of the key technologies of the advanced driver assistance system. At the same time, the high safety of pedestrian collision avoidance has high requirements for the motion state of the vehicle. However, traditional trajectory planning research focuses on the spatial path planning, while ignoring motion planning in time

Images