Driving behavior prediction method and system based on bidirectional long and short term memory network

Abstract

The invention discloses a driving behavior prediction method and system based on a bidirectional long and short term memory network, relates to the technical field of intelligent driving, and solves the technical problem that the driving behavior prediction of a driver is not accurate enough. According to the method, the function of quickly predicting the future longitudinal and transverse driving behaviors of the driver under the short time window is achieved, the prediction precision is high, the future longitudinal and transverse driving behaviors of the driver are predicted under the lightweight vehicle-mounted processor, and the economic cost is low. The method can widely serve the vehicle auxiliary driving system, improves the safety of the auxiliary driving system, enhances the understanding of the auxiliary system for a driver, facilitates the realization of safe driving, and has practicability and commercial value. Meanwhile, the deep learning method based on big data has very high prediction precision, robustness and generalization ability, is helpful for deep development of an advanced driving assistance system, and is widely suitable for driver behavior prediction under different vehicles.

Description

technical field [0001] The present application relates to the technical field of intelligent driving, and in particular, to a driving behavior prediction method and system based on a bidirectional long-short-term memory network. Background technique [0002] With the development of society, vehicles have become a wide range of means of transportation. According to a survey by the World Health Organization, an average of 1.3 million people die every year due to traffic accidents, and the economic loss caused by them is as high as 3% of the GDP of most countries. Among them, the main cause of traffic accidents comes from the driver's wrong manipulation behavior, which is often caused by fatigue, distraction, and tension. Therefore, predicting the driving behavior of the driver can provide the advanced driver assistance system with the possible manipulation behavior data of the future driver, which in turn plays a significant role in improving the driving safety of the vehicle ...

AI Technical Summary

Problems solved by technology

Therefore, a complete in-vehicle assistance system needs to be able to understand the driver's manipulation behavior. If it cannot predict the driver's driving behavior in the short term in the future and take reasonable assistance measures, the system itself will conflict with the driver and have a negative impact The driving state of the driver, which leads to a decrease in the driving safety of the vehicle

[0004] In the prior art, there is no means of predicting the short-term behavior of the driver with a lightweight vehicle-mounted system. Therefore, there is a wide demand and hardware foundation for a lightweight vehicle-mounted system that can predict the short-term longitudinal and lateral maneuvering behavior of the driver.

Images