Environmental perception system used for autonomous vehicle and based on sensor network

Abstract

The embodiment of the invention discloses an environmental perception system which is used for an autonomous vehicle and based on a sensor network. The environmental perception system comprises multiple sensors, a cloud platform, communication equipment, a memory and a controller; multiple sensors are configured on a target vehicle and other vehicles; the cloud platform is used for acquiring othervehicles in an environment where the target vehicle is located and forming global information of the environment, where the target vehicle is located, according to local environmental information; the memory memorizes historical statistical information of the local environmental information; and the controller compares the local environmental information with the historical statistical information, then generates decision credibility of the local environmental information and forms a driving decision according to the global information and the decision credibility of the local environmental information. The technical scheme of the invention comprehensively considers the decision credibility of the local environmental information for making the driving decision as well as the global information, a more effective driving decision can be made, and safety of the autonomous vehicle is improved.

Description

technical field [0001] The present disclosure relates to the technical field of automatic driving, in particular to a sensor network-based environment perception system for automatic driving vehicles. Background technique [0002] With the development of automobile intelligent technology, smart cars can realize automatic driving. Vehicle lidar and millimeter wave radar are important sensors for autonomous vehicles. LiDAR can scan road conditions, draw three-dimensional maps, and accurately identify traffic signs. Millimeter wave radar is also one of the common environmental information perception sensors, which can perceive road terrain, lane information and signal lights. Different sensors have their own characteristics in different environments. For example, in a city with a complex environment, a lidar with a larger detection angle should be selected, and a millimeter-wave radar can be selected in a harsh outdoor environment with poor lighting conditions. [0003] Duri...

AI Technical Summary

Problems solved by technology

However, the analysis of the reliability of the detection results of each sensor is ignored. Once some sensors fail or misjudgment occurs, it will affect the judgment of the entire perception system, thereby reducing the correctness of vehicle decision-making.

Moreover, even if it is possible to avoid making vehicle driving decisions based on misjudged sensor data, the driving decisions made in this situation are only correct in the local overview of the vehicle's surrounding environment. Compared with the global overview of the vehicle's surrounding environment, Such driving decisions are often not optimal

Images