Power supply prediction method and system of unmanned vehicle based on parallel neural network

Abstract

The invention discloses a power supply prediction method and system of an unmanned vehicle based on a parallel neural network. The method comprises steps of using a forces sensing sensor to measure rainfall resistance of a vehicle body and using a continuous covariance fusion algorithm to carry out data fusion; using a Kinect camera to carry out image acquisition and using a Sobel operator to carry out processing, wherein the obtained gathered water detection images are simple and clear and the detection precision is quite high; using a three-part neural network to carry out real-time prediction on electric quantity of the unmanned vehicle in an extreme rainstorm environment, wherein data of rainfall resistance and gathered water resistance is processed so that an added value of power consumption rate of the vehicle in the rainstorm environment is obtained and transmission data of road slope loss power and cell temperature is processed so that power consumption rate of the vehicle driving in the normal weather is obtained. According to the invention, by calculating variables under the rainstorm environment and variables under the normal weather in a separated manner, data processing speed is improved and timelines of the system is improved.

Description

technical field [0001] The invention belongs to the field of intelligent monitoring, in particular to a method and system for predicting power supply of unmanned vehicles based on a parallel neural network. Background technique [0002] With the rapid development of science and technology, the research and development of unmanned vehicle technology has attracted more and more attention. However, the current unmanned vehicles cannot completely replace the existing vehicles. The reason is that there are still many restrictions. The most important point is the control of the battery power of unmanned vehicles. How to ensure that unmanned vehicles can correctly control the real-time power is the basis for the realization of all vehicle-mounted unmanned driving technologies. [0003] As far as the existing technical conditions are concerned, the calculation method of battery power is generally the SOC estimation algorithm, which has poor real-time performance and cannot consider...

AI Technical Summary

Problems solved by technology

[0003] As far as the existing technical conditions are concerned, the calculation method of battery power is generally the SOC estimation algorithm, which has poor real-time performance and cannot consider the impact of environmental changes on the power consumption of unmanned vehicles.

Therefore, how to accurately and real-time estimate the battery power of vehicles in extreme environments is a technical problem faced by unmanned vehicles. So far, there is no effective technology that has been disclosed.

Images