Double-flow vehicle-mounted pedestrian and vehicle prediction method based on boundary frame and distance prediction

Abstract

The invention provides a double-flow vehicle-mounted pedestrian and vehicle prediction method based on boundary frame and distance prediction. The method comprises the main contents: pedestrian trajectory prediction, Bayesian modeling, recurrent neural network (RNN) encoder-decoder, distance prediction, and training and reasoning. The process comprises that a distance prediction flow is used for predicting a most possible vehicle distance sequence, and the boundary frame flow is composed of a Bayesian RNN encoder-decoder structure and is used for predicting the attitude distribution on a pedestrian trajectory and capturing cognition and arbitrary uncertainty; since the prediction flow of the distance prediction method is used for estimating a prediction point, the prediction flow is trained by minimizing the mean square error of a training set; and the Bayesian boundary frame prediction flow is trained by estimating and minimizing the KL divergence approximate to weight distribution. The double-flow system structure including the pedestrian boundary frame prediction and the vehicle distance prediction is adopted, the time required for prediction is greatly shortened, and the prediction accuracy of the model is also significantly improved by the uncertainty estimation.

Description

technical field [0001] The present invention relates to the field of pedestrian and vehicle prediction, in particular to a dual-stream vehicle pedestrian and vehicle prediction method based on bounding box and distance prediction. Background technique [0002] With the rapid development of the domestic automobile market, automobiles have become the main way for people to travel, bringing great convenience to people's lives, but at the same time, the incidence of traffic accidents is also increasing year by year. Therefore, how to improve the performance of automobiles and reduce the accident rate is a problem that many automobile manufacturers and researchers have been working hard to solve. In recent years, computer vision technology has developed rapidly, and advanced driver assistance systems are also becoming a development trend. Predict road conditions through the on-board system, such as predicting the distance to obstacles, detecting pedestrians in front of the vehic...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that the prediction time is too long, the purpose of the present invention is to provide a dual-stream vehicle pedestrian vehicle prediction method based on bounding box and distance prediction. The ranging flow will predict the most likely vehicle distance sequence, and the bounding box flow is determined by Bayesian The RNN encoder-decoder architecture is used to predict the distribution of attitudes on pedestrian trajectories, capturing cognition and arbitrary uncertainty. Since the odometry prediction flow is used to predict point estimates, it is used to minimize the mean square on the training set. error to train it, Bayesian bounding box prediction flow is trained by estimating and minimizing the KL divergence of its approximate weight distribution

Images