CN109872415B - Vehicle speed estimation method and system based on neural network - Google Patents
Vehicle speed estimation method and system based on neural network Download PDFInfo
- Publication number
- CN109872415B CN109872415B CN201811620672.7A CN201811620672A CN109872415B CN 109872415 B CN109872415 B CN 109872415B CN 201811620672 A CN201811620672 A CN 201811620672A CN 109872415 B CN109872415 B CN 109872415B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- training
- speed
- real
- time data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 24
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 33
- 239000011159 matrix material Substances 0.000 claims abstract description 23
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 230000001133 acceleration Effects 0.000 claims description 59
- 238000005070 sampling Methods 0.000 claims description 10
- 238000011176 pooling Methods 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 claims description 3
- 239000013598 vector Substances 0.000 abstract description 3
- 230000001186 cumulative effect Effects 0.000 abstract 1
- 230000010354 integration Effects 0.000 abstract 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Images
Landscapes
- Steering Control In Accordance With Driving Conditions (AREA)
Abstract
本发明公开一种基于神经网络的车速估计方法及系统，该车速估计方法包括获取训练样本和训练输出量，并将训练样本中的训练输入量拓展为8*8的对称车辆实时数据矩阵，其中训练输出量和训练输入量均为向量形式；然后根据对称车辆实时数据矩阵和训练输出量对卷积神经网络进行训练；最后获取当前车辆的实时数据，并将当前车辆的实时数据输入到训练后的卷积神经网络模型中，来估计当前车辆对地的横向速度和纵向速度。本发明利用卷积神经网络，将对称车辆实时数据矩阵作为输入，经过卷积计算收敛，输出车辆对地的实时纵向速度及横向速度，不仅能够适用于车辆打滑的极限工况，计算量小，而且有效的解决后期积分过程中的累计误差问题。
The invention discloses a method and system for estimating vehicle speed based on neural network. The method for estimating vehicle speed includes acquiring training samples and training output, and expanding the training input in the training samples into an 8*8 symmetric vehicle real-time data matrix, wherein The training output and training input are both in the form of vectors; then the convolutional neural network is trained according to the real-time data matrix of the symmetric vehicle and the training output; finally, the real-time data of the current vehicle is obtained, and the real-time data of the current vehicle is input to the post-training The convolutional neural network model of the current vehicle is used to estimate the lateral and longitudinal speed of the current vehicle on the ground. The invention uses the convolutional neural network, takes the real-time data matrix of the symmetric vehicle as the input, and converges through the convolution calculation to output the real-time longitudinal speed and lateral speed of the vehicle on the ground, which is not only applicable to the extreme working condition of the vehicle slipping, but also has a small amount of calculation. And it can effectively solve the cumulative error problem in the later integration process.
Description
Technical Field
The invention relates to the technical field of vehicle speed estimation, in particular to a vehicle speed estimation method and system based on a neural network.
Background
Based on the acquisition value of the acceleration sensor, filtering prediction is carried out through an unscented Kalman method, and the acceleration value is integrated to obtain the corresponding transverse speed and longitudinal speed.
The method is characterized in that the longitudinal speed and the transverse speed are calculated based on the collected four-wheel speed signals and by combining a vehicle four-wheel model, so that the method has better real-time performance, but the defect that the model is built under the condition that the vehicle does not slide and is not suitable for the skid limit working condition of the vehicle.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a vehicle speed estimation method and system based on a neural network.
In order to achieve the purpose, the invention provides the following scheme:
a neural network-based vehicle speed estimation method, the vehicle speed estimation method comprising:
obtaining a training sample; the data in the training sample is vehicle real-time data in a precondition test; the vehicle real-time data comprises four wheel speeds, steering wheel angles, yaw acceleration, longitudinal acceleration and lateral acceleration;
expanding the training input quantity in the training sample into a 8-by-8 symmetric vehicle real-time data matrix;
determining a training output quantity; the training output is the lateral speed and the longitudinal speed of the vehicle to the ground;
training a convolutional neural network according to the symmetric vehicle real-time data matrix and the training output quantity to obtain a trained convolutional neural network model;
acquiring real-time data of a current vehicle; the real-time data of the current vehicle comprise four wheel speeds of the current vehicle, a steering wheel angle of the current vehicle, a yaw acceleration of the current vehicle, a longitudinal acceleration of the current vehicle and a lateral acceleration of the current vehicle;
and inputting the real-time data of the current vehicle into the trained convolutional neural network model, and estimating the transverse speed and the longitudinal speed of the current vehicle to the ground.
Optionally, the inputting the real-time data of the current vehicle into the trained convolutional neural network model to estimate the lateral speed and the longitudinal speed of the current vehicle to the ground specifically includes:
converting the real-time data of the current vehicle according to the form of the training input quantity;
and inputting the converted real-time data of the current vehicle into the trained convolutional neural network model, and estimating the transverse speed and the longitudinal speed of the current vehicle to the ground.
Optionally, the values of the four wheel speeds, the steering wheel angles, the yaw acceleration, the longitudinal acceleration and the lateral acceleration are used as the training input quantities of the training samples in a vector form.
Optionally, the training input amount is [ wh1 wh2 wh3 wh4 str yaw log lat ]; wherein wh1 wh2 wh3 wh4 represents a first wheel speed, a second wheel speed, a third wheel speed, and a fourth wheel speed, respectively; str denotes a steering wheel angle; yaw represents Yaw acceleration; log represents a longitudinal acceleration; lat represents lateral acceleration.
Optionally, the symmetric vehicle real-time data matrix is:
【wh1 wh1 wh1 wh1 wh1 wh1 wh1 wh1】
【wh1 wh2 wh2 wh2 wh2 wh2 wh2 wh2】
【wh1 wh2 wh3 wh3 wh3 wh3 wh3 wh3】
【wh1 wh2 wh3 wh4 wh4 wh4 wh4 wh4】
【wh1 wh2 wh3 wh4 str str str str】
【wh1 wh2 wh3 wh4 str yaw yaw yaw】
【wh1 wh2 wh3 wh4 str yaw log log】
【wh1 wh2 wh3 wh4 str yaw log lat】。
optionally, the training output is the lateral speed and the longitudinal speed of the vehicle to the ground measured by the two-axis optical sensor at the same time.
Optionally, the training output is [ vlat vllog ]; where vlat represents the lateral velocity and vlog represents the longitudinal velocity.
A neural network-based vehicle speed estimation system, the vehicle speed estimation system method comprising:
the training sample acquisition module is used for acquiring a training sample; the data in the training sample is vehicle real-time data in a precondition test; the vehicle real-time data comprises four wheel speeds, steering wheel angles, yaw acceleration, longitudinal acceleration and lateral acceleration;
the symmetric vehicle real-time data matrix determining module is used for expanding the training input quantity in the training sample into an 8 × 8 symmetric vehicle real-time data matrix;
a training output quantity determining module for determining a training output quantity; the training output is the lateral speed and the longitudinal speed of the vehicle to the ground;
the convolutional neural network training module is used for training a convolutional neural network according to the symmetric vehicle real-time data matrix and the training output quantity to obtain a trained convolutional neural network model;
the current vehicle real-time data acquisition module is used for acquiring the real-time data of the current vehicle; the real-time data of the current vehicle comprise four wheel speeds of the current vehicle, a steering wheel angle of the current vehicle, a yaw acceleration of the current vehicle, a longitudinal acceleration of the current vehicle and a lateral acceleration of the current vehicle;
and the current vehicle-to-ground transverse speed and longitudinal speed estimation module is used for inputting the real-time data of the current vehicle into the trained convolutional neural network model and estimating the transverse speed and the longitudinal speed of the current vehicle to the ground.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention provides a vehicle speed estimation method and system based on a neural network, which utilize a convolution neural network to output real-time longitudinal speed and transverse speed of a vehicle to the ground by taking a symmetric vehicle real-time data matrix as input through convolution calculation convergence.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a neural network-based vehicle speed estimation method according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a vehicle speed estimation system based on a neural network according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a vehicle speed estimation method and a vehicle speed estimation system based on a neural network, which are not only suitable for the limited working condition of vehicle slip, but also have small calculated amount.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
The invention utilizes the convolutional neural network to accurately obtain the transverse speed and the longitudinal speed of the vehicle to the ground in real time.
Fig. 1 is a schematic flow chart of a vehicle speed estimation method based on a neural network according to an embodiment of the present invention, and as shown in fig. 1, the vehicle speed estimation method based on the neural network according to the embodiment of the present invention includes the following steps:
step 101: training samples are obtained.
Training data in the sample as vehicle real-time data in the precondition test; the vehicle real-time data comprises four wheel speeds, steering wheel corners, yaw acceleration, longitudinal acceleration and lateral acceleration, and the values of the four wheel speeds, the steering wheel corners, the yaw acceleration, the longitudinal acceleration and the lateral acceleration are used as training input quantities of training samples in a vector form.
The training input was [ wh1 wh2 wh3 wh4 str yaw log lat ]. Wherein wh1 wh2 wh3 wh4 represents a first wheel speed, a second wheel speed, a third wheel speed, and a fourth wheel speed, respectively; str denotes a steering wheel angle; yaw represents Yaw acceleration; log represents a longitudinal acceleration; lat represents lateral acceleration.
Step 102: and expanding the training input quantity in the training sample into an 8-by-8 symmetric vehicle real-time data matrix. The symmetric vehicle real-time data matrix is as follows:
【wh1 wh1 wh1 wh1 wh1 wh1 wh1 wh1】
【wh1 wh2 wh2 wh2 wh2 wh2 wh2 wh2】
【wh1 wh2 wh3 wh3 wh3 wh3 wh3 wh3】
【wh1 wh2 wh3 wh4 wh4 wh4 wh4 wh4】
【wh1 wh2 wh3 wh4 str str str str】
【wh1 wh2 wh3 wh4 str yaw yaw yaw】
【wh1 wh2 wh3 wh4 str yaw log log】
【wh1 wh2 wh3 wh4 str yaw log lat】。
step 103: a training output is determined.
And the transverse speed and the longitudinal speed of the vehicle to the ground, which are measured by the double-shaft optical sensor at the same time, are used as training output.
The training output is [ vlat vlog ]; where vlat represents the lateral velocity and vlog represents the longitudinal velocity.
Step 104: and training the convolutional neural network according to the symmetric vehicle real-time data matrix and the training output quantity to obtain a trained convolutional neural network model.
There are two main types of network layers in convolutional neural networks, namely convolutional layers and sampling layers. The convolutional layer is used for extracting various characteristics of the training sample; the sampling layer is used for abstracting the original signal, so that training parameters are greatly reduced, and the degree of overfitting of the model can be reduced.
When the input training input quantity passes through the convolutional layer, the convolutional core of the convolutional layer is calculated in the input training input quantity by a sliding window. Each parameter of the convolution kernel is a weight parameter in the neural network. And multiplying each parameter of the convolution kernel by the data in the corresponding training sample to obtain the result of the convolution layer.
After the convolutional layer outputs the features of the training input quantity, the features are input into the sampling layer for processing in order to further reduce the overfitting degree of the network training parameters and the model. The sampling layer is used for acting on each output characteristic and reducing the size of the output characteristic, and particularly, each output characteristic is processed by adopting a maximum pooling method so as to reduce the size of the output characteristic.
And comparing the data subjected to the final layer of pooling with the output data of the training sample to obtain the approximation degree of each transverse speed and each longitudinal speed, and selecting the most similar result for outputting.
Step 105: acquiring real-time data of a current vehicle; the real-time data of the current vehicle comprises four wheel speeds of the current vehicle, a steering wheel angle of the current vehicle, a yaw acceleration of the current vehicle, a longitudinal acceleration of the current vehicle and a lateral acceleration of the current vehicle.
Step 106: and inputting the real-time data of the current vehicle into the trained convolutional neural network model, and estimating the transverse speed and the longitudinal speed of the current vehicle to the ground.
The method specifically comprises the following steps: converting the real-time data of the current vehicle according to the form of the training input quantity; and inputting the converted real-time data of the current vehicle into the trained convolutional neural network model, and estimating the transverse speed and the longitudinal speed of the current vehicle to the ground.
In order to achieve the purpose, the invention also provides a vehicle speed estimation system based on the neural network.
As shown in fig. 2, a vehicle speed estimation system method provided by the embodiment of the invention includes:
a training sample obtaining module 100, configured to obtain a training sample; the data in the training sample is vehicle real-time data in a precondition test; the vehicle real-time data comprises four wheel speeds, steering wheel angles, yaw acceleration, longitudinal acceleration and lateral acceleration.
And a symmetric vehicle real-time data matrix determining module 200, configured to expand the training input amount in the training sample into an 8 × 8 symmetric vehicle real-time data matrix.
A training output determination module 300 for determining a training output; the training output is the lateral and longitudinal speed of the vehicle over the ground.
And the convolutional neural network training module 400 is configured to train a convolutional neural network according to the symmetric vehicle real-time data matrix and the training output quantity, so as to obtain a trained convolutional neural network model.
A current vehicle real-time data acquisition module 500, configured to acquire real-time data of a current vehicle; the real-time data of the current vehicle comprises four wheel speeds of the current vehicle, a steering wheel angle of the current vehicle, a yaw acceleration of the current vehicle, a longitudinal acceleration of the current vehicle and a lateral acceleration of the current vehicle.
And a current vehicle-to-ground transverse speed and longitudinal speed estimation module 600, configured to input real-time data of the current vehicle into the trained convolutional neural network model, and estimate a transverse speed and a longitudinal speed of the current vehicle to ground.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (6)
1. A vehicle speed estimation method based on a neural network is characterized by comprising the following steps:
obtaining a training sample; the data in the training sample is vehicle real-time data in a precondition test; the vehicle real-time data comprises four wheel speeds, steering wheel angles, yaw acceleration, longitudinal acceleration and lateral acceleration;
expanding the training input quantity in the training sample into a 8-by-8 symmetric vehicle real-time data matrix; the symmetric vehicle real-time data matrix is as follows:
【wh1 wh1 wh1 wh1 wh1 wh1 wh1 wh1】
【wh1 wh2 wh2 wh2 wh2 wh2 wh2 wh2】
【wh1 wh2 wh3 wh3 wh3 wh3 wh3 wh3】
【wh1 wh2 wh3 wh4 wh4 wh4 wh4 wh4】
【wh1 wh2 wh3 wh4 str str str str】
【wh1 wh2 wh3 wh4 stryaw yaw yaw】
【wh1 wh2 wh3 wh4 stryaw log log】
【wh1 wh2 wh3 wh4 stryaw log lat】；
wherein wh1 wh2 wh3 wh4 represents a first wheel speed, a second wheel speed, a third wheel speed, and a fourth wheel speed, respectively; str denotes a steering wheel angle; yaw represents Yaw acceleration; log represents a longitudinal acceleration; lat represents lateral acceleration;
determining a training output quantity; the training output is the lateral speed and the longitudinal speed of the vehicle to the ground;
training a convolutional neural network according to the symmetric vehicle real-time data matrix and the training output quantity to obtain a trained convolutional neural network model;
the convolutional neural network comprises two types of network layers, namely a convolutional layer and a sampling layer;
when the input training input quantity passes through the convolutional layer, the convolutional core of the convolutional layer is obtained by calculation through a sliding window in the input training input quantity; each parameter of the convolution kernel is a weight parameter in the neural network; multiplying each parameter of the convolution kernel by data in the corresponding training sample to obtain a result of the convolution layer;
after the convolutional layer outputs the characteristics of the training input quantity, inputting the characteristics into a sampling layer for processing in order to further reduce the overfitting degree of the network training parameters and the model; the sampling layer is used for acting on each output characteristic and reducing the size of the output characteristic, and specifically, each output characteristic is processed by adopting a maximum pooling method so as to reduce the size of the output characteristic;
comparing the data subjected to the final layer of pooling with the output data of the training sample to obtain the approximation degree of each transverse speed and each longitudinal speed, and selecting the most similar result for output;
acquiring real-time data of a current vehicle; the real-time data of the current vehicle comprise four wheel speeds of the current vehicle, a steering wheel angle of the current vehicle, a yaw acceleration of the current vehicle, a longitudinal acceleration of the current vehicle and a lateral acceleration of the current vehicle;
and inputting the real-time data of the current vehicle into the trained convolutional neural network model, and estimating the transverse speed and the longitudinal speed of the current vehicle to the ground.
2. The vehicle speed estimation method based on the neural network as claimed in claim 1, wherein the inputting the real-time data of the current vehicle into the trained convolutional neural network model to estimate the lateral speed and the longitudinal speed of the current vehicle to the ground specifically comprises:
converting the real-time data of the current vehicle according to the form of the training input quantity;
and inputting the converted real-time data of the current vehicle into the trained convolutional neural network model, and estimating the transverse speed and the longitudinal speed of the current vehicle to the ground.
3. The neural-network-based vehicle speed estimation method according to claim 1, characterized in that numerical values of the four wheel speeds, the steering wheel angle, the yaw acceleration, the longitudinal acceleration, the lateral acceleration are vector-form as the training input amounts of the training samples.
4. The neural-network-based vehicle speed estimation method of claim 1, wherein the training output is a lateral speed and a longitudinal speed of the vehicle to the ground measured at the same time via a two-axis optical sensor.
5. The neural-network based vehicle speed estimation method of claim 4, wherein the training output is [ vlatvlog ]; where vlat represents the lateral velocity and vlog represents the longitudinal velocity.
6. A neural network-based vehicle speed estimation system, the vehicle speed estimation system method comprising:
the training sample acquisition module is used for acquiring a training sample; the data in the training sample is vehicle real-time data in a precondition test; the vehicle real-time data comprises four wheel speeds, steering wheel angles, yaw acceleration, longitudinal acceleration and lateral acceleration;
the symmetric vehicle real-time data matrix determining module is used for expanding the training input quantity in the training sample into an 8 × 8 symmetric vehicle real-time data matrix; the symmetric vehicle real-time data matrix is as follows:
【wh1 wh1 wh1 wh1 wh1 wh1 wh1 wh1】
【wh1 wh2 wh2 wh2 wh2 wh2 wh2 wh2】
【wh1 wh2 wh3 wh3 wh3 wh3 wh3 wh3】
【wh1 wh2 wh3 wh4 wh4 wh4 wh4 wh4】
【wh1 wh2 wh3 wh4 str str str str】
【wh1 wh2 wh3 wh4 stryaw yaw yaw】
【wh1 wh2 wh3 wh4 stryaw log log】
【wh1 wh2 wh3 wh4 stryaw log lat】；
wherein wh1 wh2 wh3 wh4 represents a first wheel speed, a second wheel speed, a third wheel speed, and a fourth wheel speed, respectively; str denotes a steering wheel angle; yaw represents Yaw acceleration; log represents a longitudinal acceleration; lat represents lateral acceleration;
a training output quantity determining module for determining a training output quantity; the training output is the lateral speed and the longitudinal speed of the vehicle to the ground;
the convolutional neural network training module is used for training a convolutional neural network according to the symmetric vehicle real-time data matrix and the training output quantity to obtain a trained convolutional neural network model;
the convolutional neural network comprises two types of network layers, namely a convolutional layer and a sampling layer;
when the input training input quantity passes through the convolutional layer, the convolutional core of the convolutional layer is obtained by calculation through a sliding window in the input training input quantity; each parameter of the convolution kernel is a weight parameter in the neural network; multiplying each parameter of the convolution kernel by data in the corresponding training sample to obtain a result of the convolution layer;
after the convolutional layer outputs the characteristics of the training input quantity, inputting the characteristics into a sampling layer for processing in order to further reduce the overfitting degree of the network training parameters and the model; the sampling layer is used for acting on each output characteristic and reducing the size of the output characteristic, and specifically, each output characteristic is processed by adopting a maximum pooling method so as to reduce the size of the output characteristic;
comparing the data subjected to the final layer of pooling with the output data of the training sample to obtain the approximation degree of each transverse speed and each longitudinal speed, and selecting the most similar result for output;
the current vehicle real-time data acquisition module is used for acquiring the real-time data of the current vehicle; the real-time data of the current vehicle comprise four wheel speeds of the current vehicle, a steering wheel angle of the current vehicle, a yaw acceleration of the current vehicle, a longitudinal acceleration of the current vehicle and a lateral acceleration of the current vehicle;
and the current vehicle-to-ground transverse speed and longitudinal speed estimation module is used for inputting the real-time data of the current vehicle into the trained convolutional neural network model and estimating the transverse speed and the longitudinal speed of the current vehicle to the ground.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811620672.7A CN109872415B (en) | 2018-12-28 | 2018-12-28 | Vehicle speed estimation method and system based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811620672.7A CN109872415B (en) | 2018-12-28 | 2018-12-28 | Vehicle speed estimation method and system based on neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109872415A CN109872415A (en) | 2019-06-11 |
| CN109872415B true CN109872415B (en) | 2021-02-02 |
Family
ID=66917234
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811620672.7A Active CN109872415B (en) | 2018-12-28 | 2018-12-28 | Vehicle speed estimation method and system based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109872415B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102023111544A1 (en) * | 2023-05-04 | 2024-11-07 | Ford Global Technologies Llc | Method for estimating the reference speed of a vehicle for boundary conditions |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110288154A (en) * | 2019-06-25 | 2019-09-27 | 北京百度网讯科技有限公司 | Velocity prediction method, device, equipment and medium |
| JP6772351B1 (en) * | 2019-09-18 | 2020-10-21 | Ｔｏｙｏ Ｔｉｒｅ株式会社 | Tire physical information estimation system |
| KR102729857B1 (en) * | 2019-12-10 | 2024-11-12 | 현대자동차주식회사 | System of modeling abs controller of vehicle |
| CN111077335B (en) * | 2020-01-22 | 2021-03-02 | 滴图(北京)科技有限公司 | Vehicle speed detection method, vehicle speed detection device and readable storage medium |
| CN111428905B (en) * | 2020-02-11 | 2020-12-08 | 北京理工大学 | A method and system for predicting longitudinal vehicle speed under all operating conditions |
| CN111693044A (en) * | 2020-06-19 | 2020-09-22 | 南京晓庄学院 | Fusion positioning method |
| CN114114183B (en) * | 2021-09-09 | 2025-04-04 | 深圳大学 | Vehicle-mounted SAR speed estimation method, system and terminal based on deep learning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011120201A1 (en) * | 2010-04-01 | 2011-10-06 | 江苏六维物流设备实业有限公司 | Piler neural network control technology |
| CN105270383A (en) * | 2014-05-30 | 2016-01-27 | 福特全球技术公司 | Vehicle speed profile prediction using neural networks |
| CN105946861A (en) * | 2016-06-02 | 2016-09-21 | 大连理工大学 | NAR neural network vehicle speed prediction method based on driving intention recognition |
| CN107368914A (en) * | 2017-06-15 | 2017-11-21 | 淮阴工学院 | A kind of hypervelocity based on BP neural network distinguishes model optimization method |
| CN107527113A (en) * | 2017-08-01 | 2017-12-29 | 北京理工大学 | A kind of operating mode Forecasting Methodology of hybrid car travel operating mode |
| CN107909678A (en) * | 2017-11-29 | 2018-04-13 | 思建科技有限公司 | One kind driving risk evaluating method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107729986B (en) * | 2017-09-19 | 2020-11-03 | 平安科技（深圳）有限公司 | Driving model training method, driver identification method, device, equipment and medium |
| CN108622105B (en) * | 2018-04-16 | 2019-12-31 | 吉林大学 | Prediction and early warning system of safe vehicle speed in curves based on multiple regression analysis |
-
2018
- 2018-12-28 CN CN201811620672.7A patent/CN109872415B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011120201A1 (en) * | 2010-04-01 | 2011-10-06 | 江苏六维物流设备实业有限公司 | Piler neural network control technology |
| CN105270383A (en) * | 2014-05-30 | 2016-01-27 | 福特全球技术公司 | Vehicle speed profile prediction using neural networks |
| CN105946861A (en) * | 2016-06-02 | 2016-09-21 | 大连理工大学 | NAR neural network vehicle speed prediction method based on driving intention recognition |
| CN107368914A (en) * | 2017-06-15 | 2017-11-21 | 淮阴工学院 | A kind of hypervelocity based on BP neural network distinguishes model optimization method |
| CN107527113A (en) * | 2017-08-01 | 2017-12-29 | 北京理工大学 | A kind of operating mode Forecasting Methodology of hybrid car travel operating mode |
| CN107909678A (en) * | 2017-11-29 | 2018-04-13 | 思建科技有限公司 | One kind driving risk evaluating method and system |
Non-Patent Citations (1)
| Title |
|---|
| 基于BP神经网络及其优化算法的汽车车速预测;谢洁;《中国优秀硕士学位论文全文数据库信息科技辑》;20150115;正文5-40页 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102023111544A1 (en) * | 2023-05-04 | 2024-11-07 | Ford Global Technologies Llc | Method for estimating the reference speed of a vehicle for boundary conditions |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109872415A (en) | 2019-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109872415B (en) | Vehicle speed estimation method and system based on neural network | |
| CN114021441A (en) | A CNN-BiLSTM-based ship motion attitude prediction method | |
| CN106295637B (en) | A vehicle recognition method based on deep learning and reinforcement learning | |
| CN110595775A (en) | Rolling bearing fault diagnosis method based on multi-branch and multi-scale convolutional neural network | |
| CN112733386B (en) | A deep neural network-based inversion method for multicomponent reactive solute transport parameters | |
| CN104251712B (en) | MEMS gyro random error compensation method based on wavelet multi-scale analysis | |
| CN109618288B (en) | Wireless sensor network distance measurement system and method based on deep convolutional neural network | |
| CN106092600A (en) | A kind of pavement identification method for proving ground strengthening road | |
| CN111161224A (en) | Classification and evaluation system and method of casting internal defects based on deep learning | |
| CN112578419B (en) | A GPS data reconstruction method based on GRU network and Kalman filtering | |
| CN106202756A (en) | Based on monolayer perceptron owing determines blind source separating source signal restoration methods | |
| CN108846200A (en) | A kind of quasi-static Bridge Influence Line recognition methods based on iterative method | |
| CN113203464A (en) | Sensor fault detection method of dynamic vehicle-mounted weighing system | |
| CN119313717B (en) | Vehicle-mounted camera visibility inversion method, device, medium and electronic equipment | |
| CN109976311A (en) | Single order fixed set point control system sensor fault diagnosis method and diagnostic system | |
| CN120198495A (en) | A Kalman filter attitude estimation method based on SigKAN network assistance | |
| CN115631428A (en) | Unsupervised image fusion method and system based on structural texture decomposition | |
| Zhao et al. | Tire-Road friction coefficients adaptive estimation through image and vehicle dynamics integration | |
| CN119622456A (en) | A method for training end-to-end autonomous driving policies | |
| CN114759904A (en) | Data processing method, device, equipment, readable storage medium and program product | |
| CN111627056A (en) | Depth estimation-based driving visibility determination method and device | |
| CN116148193B (en) | Water quality monitoring method, device, equipment and storage medium | |
| CN114154231B (en) | A vehicle driving state estimation system and method based on deep reinforcement learning | |
| CN116129237A (en) | Training method of image detection model, image detection method and device | |
| CN115046550A (en) | Inertial attitude determination positioning method applied to automobile |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |