A driving evaluation method and system for a car driver
An evaluation system and driver technology, applied in the field of driving information, can solve problems such as failure to truly reflect driving skills, failure to consider environmental factors, and too one-sided evaluation results.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0036] A driving evaluation method for a car driver, comprising the steps of:
[0037] Step 1, collect the driving data, environmental data and driver information data of a certain journey of the vehicle;
[0038] Step 2, sending the collected driving data, environment data and driver information data to the server;
[0039] Step 3, the server analyzes the received information data and gives the driver a driving score.
[0040] We know that driving data is the data that can most directly reflect the driving state of the vehicle, and the driving state of the vehicle can also reflect the driving behavior of the driver to a high degree. However, it is not comprehensive enough to evaluate the driver's driving behavior only based on the vehicle's driving state or driving data, because environmental factors will also affect the driving state of the vehicle to a certain extent, such as the slippery degree of the ground, the slope or flatness of the ground etc., even weather, temper...
Embodiment 2
[0044] The step 3 specifically includes:
[0045] Step 3.1, forming a sample matrix of the driving data, environment data and driver information data;
[0046] Step 3.2, define the parameter matrix and bias vector of the neural network, wherein the parameter matrix and bias vector are obtained by training sample data, and the initial data are all 0; the sample data is obtained through a large number of tests to ensure Representative of the parameter matrix and bias vector.
[0047] Step 3.3, multiplying the sample matrix and the parameter matrix, and adding the bias vector to perform an activation function to obtain the first neural network output matrix;
[0048] Step 3.4, using the first neural network output matrix as an input matrix, multiplying it with the parameter matrix, and adding the bias vector to obtain a second neural network output matrix;
[0049] Step 3.5, back propagating the second neural network output matrix and the real result to obtain the algorithm mod...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More