Pedestrian crossing behavior prediction method for plane intersection

Abstract

The invention relates to a pedestrian crossing behavior prediction method for a plane intersection. The pedestrian crossing behavior prediction method comprises the following steps of 1, designing aninstant reward function; 2, establishing a full convolutional neural network long and short term memory (FCNLSTM) model to predict an action reward function; 3, training a full convolutional neural network long and short term memory (FCNLSTM) model based on reinforcement learning; 4, predicting pedestrian crossing behaviors and carrying out danger early warning. According to the technical scheme,a complex pedestrian motion model does not need to be established, a large number of labeled data sets do not need to be prepared, pedestrian crossing behavior characteristics at the plane intersection are autonomously learned, behaviors such as walking, stopping and fast running are predicted, and particularly, pedestrian crossing behaviors when dangers such as human-vehicle collision and rubbingare induced are predicted in real time; therefore, danger early warning is carried out on pedestrians and passing vehicles crossing the street, the traffic accident rate of key road sections such asplane intersections can be reduced, and the safety of the pedestrians in the traffic environment is guaranteed.

Description

technical field [0001] The invention relates to a prediction method, in particular to a pedestrian crossing behavior prediction method facing level intersections, and belongs to the technical field of behavior modeling and prediction of traffic participants. Background technique [0002] Pedestrians are the main participants in road traffic, and their behavior is an important factor affecting traffic safety, especially at level intersections where there are a large number of crossing behaviors such as school entrances and exits and unsignaled entrances. Nearly 70% of the total number of traffic accidents. Therefore, aiming at the recognition and prediction of pedestrian crossing behavior at level intersections, especially the real-time prediction of pedestrian crossing behavior when dangers such as pedestrian-vehicle collision and rubbing are induced, and the danger warning of pedestrians crossing the street and passing vehicles is the key to building an intelligent road. I...

AI Technical Summary

Problems solved by technology

For model-based methods, it is impossible to construct an explicit mathematical model to describe pedestrian crossing behavior at level intersections

For data-driven deep learning methods, it is difficult to obtain massive labeled datasets to extract interdependence and random features of pedestrian crossing behavior

Images