CN112417598A - Multi-source fusion vehicle state parallel estimation method - Google Patents
Multi-source fusion vehicle state parallel estimation method Download PDFInfo
- Publication number
- CN112417598A CN112417598A CN202011309519.XA CN202011309519A CN112417598A CN 112417598 A CN112417598 A CN 112417598A CN 202011309519 A CN202011309519 A CN 202011309519A CN 112417598 A CN112417598 A CN 112417598A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- neural network
- steering wheel
- longitudinal acceleration
- estimation
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000004927 fusion Effects 0.000 title claims abstract description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 27
- 230000001133 acceleration Effects 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000001914 filtration Methods 0.000 claims abstract description 14
- 238000005457 optimization Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000012360 testing method Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Biology (AREA)
- Automation & Control Theory (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Computational Mathematics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Medical Informatics (AREA)
- Steering Control In Accordance With Driving Conditions (AREA)
Abstract
The invention relates to a multi-source fusion vehicle state parallel estimation method, which comprises the following steps: s1, acquiring real values corresponding to the steering wheel corner and the longitudinal acceleration of the vehicle at each moment in the running process, and dividing the real values into a training set and a test set according to the proportion after preprocessing; step S2, constructing a neural network and training a training set; s3, constructing an extended Kalman filtering estimation model; step S4: in the running process of a vehicle, a steering wheel corner sensor and a longitudinal acceleration sensor are used for acquiring a steering wheel corner and a longitudinal acceleration, and the steering wheel corner and the longitudinal acceleration are respectively transmitted to a trained neural network estimation model and an extended Kalman filtering estimation model for processing; step S5: and respectively obtaining estimation results, and performing signal fusion and optimization processing to obtain a final estimation result. The method can quickly and effectively obtain the vehicle state estimation, and plays a decisive role in the accurate control of the subsequent vehicle active safety.
Description
Technical Field
The invention relates to the field of vehicle state parameter estimation design, in particular to a multi-source fusion vehicle state parallel estimation method.
Background
With the development of the automobile industry, automobiles are not only simple mechanical bodies, but also gradually intelligentized and automated. Various sensors are additionally arranged on the vehicle to collect external information required by the vehicle. The vehicle can judge the current vehicle state through the collected information, and then various vehicle dynamic control systems such as ESP, ABS and the like are called. On the mass production vehicle, because the manufacturing cost needs to be controlled, the collection of the external information parameters of the vehicle cannot be performed by a sensor with high price, therefore, some other parameters to be estimated to the vehicle active control or passive control need to be performed through the collected signals by the sensor carried by the current mass production vehicle, such as: centroid slip angle, yaw rate, and longitudinal vehicle speed.
Disclosure of Invention
In view of the above, the present invention provides a multi-source fused parallel estimation method for vehicle states,
in order to achieve the purpose, the invention adopts the following technical scheme:
a multi-source fusion vehicle state parallel estimation method comprises the following steps:
s1, acquiring real values corresponding to the steering wheel corner and the longitudinal acceleration of the vehicle at each moment in the running process, and dividing the real values into a training set and a test set according to the proportion after preprocessing;
step S2, constructing a neural network, training a training set and obtaining a trained neural network estimation model;
s3, constructing an extended Kalman filtering estimation model;
step S4: in the running process of a vehicle, a steering wheel corner sensor and a longitudinal acceleration sensor are used for acquiring a steering wheel corner and a longitudinal acceleration, and the steering wheel corner and the longitudinal acceleration are respectively transmitted to a trained neural network estimation model and an extended Kalman filtering estimation model for processing;
step S5: and performing signal fusion and optimization processing according to estimation results respectively obtained by the trained neural network estimation model and the extended Kalman filtering estimation model to obtain a final estimation result.
Further, the real values of the steering wheel angle and the longitudinal acceleration comprise a centroid slip angle, a yaw rate and a longitudinal vehicle speed.
Further, the neural network adopts a radial basis function neural network to predict the relevant vehicle parameters, wherein the neural network implies that the layer node q is 9, and the nodes of the input and output layers are 1.
Further, the neural network training process is divided into two stages, namely a non-instructor learning stage and an instructor learning stage.
Further, the tutor-free learning stage specifically includes:
(1) given the initial center vector C of each hidden nodei(0) (i ═ 1,2,. q), learning rate β (0)<β(0)<1) And a threshold value epsilon for judging to stop the calculation
(2) Node for calculating Euclidean distance and calculating minimum distance
In the formula, k is a sample serial number; r is a central vector ci(k-1) hidden node sequence number closest to input sample x (k);
(3) center of adjustment
In the formula, β (k) is a learning rate. β (k) ═ β (k-1)/(1+ int (k/q))1/2(ii) a int (·) denotes rounding (·);
(4) determining cluster quality
The above (2) and (3) were repeated for all samples k (k ═ 1, 2.., N) until the following was satisfied
If the formula is satisfied, finishing clustering;
further, the teacher learning stage specifically includes:
when c is going toiAfter the determination, training the weight between the hidden layer and the output layer, and if the weight is a linear equation set, the problem of linear optimization becomes full-time; connection weight w between hidden layer and output layer of radial basic bible networkkiThe (k 1, 2.. times.l; i 1, 2.. times.q) learning algorithm is
wki(k+1)＝wki(k)+η(tk-yk)ui(x)/uTu
Wherein u is [ u ]1(x),u2(x),..,uq(x)]T；ui(x) Is a Gaussian function; η is the learning rate.
Further, the extended kalman filter estimation model specifically includes: the vehicle three-degree-of-freedom dynamic model is used as a state equation and an observation equation, and the centroid slip angle, the yaw angular velocity and the longitudinal vehicle speed are estimated based on the extended Kalman filtering algorithm
In the formula, omega is the yaw angular velocity, beta is the centroid slip angle,vxIs the longitudinal speed, k1And k2Yaw stiffness, a and b are the lengths from the center of mass of the vehicle to the front and rear axes, Iz is the moment of inertia, m is the total vehicle mass, delta is the steering wheel angle, axAnd ayLongitudinal acceleration and lateral acceleration.
Further, the signal fusion and optimization module compares the signals estimated by different algorithms with the reference signal to obtain different confidence degrees of the two estimated signals, wherein the signal with the high confidence degree has a large weight value, and otherwise, the signal with the low confidence degree has a small weight value.
Compared with the prior art, the invention has the following beneficial effects:
the method can quickly and effectively obtain the vehicle state estimation, and plays a decisive role in the accurate control of the subsequent vehicle active safety.
Drawings
Fig. 1 is a schematic diagram of the principle of the present invention.
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
Referring to fig. 1, the invention provides a multi-source fusion vehicle state parallel estimation method, which includes the following steps:
s1, acquiring real values corresponding to the steering wheel corner and the longitudinal acceleration of the vehicle at each moment in the running process, and dividing the real values into a training set and a test set according to the proportion after preprocessing;
step S2, constructing a neural network, training a training set and obtaining a trained neural network estimation model;
s3, constructing an extended Kalman filtering estimation model;
step S4: in the running process of a vehicle, a steering wheel corner sensor and a longitudinal acceleration sensor are used for acquiring a steering wheel corner and a longitudinal acceleration, and the steering wheel corner and the longitudinal acceleration are respectively transmitted to a trained neural network estimation model and an extended Kalman filtering estimation model for processing;
step S5: and performing signal fusion and optimization processing according to estimation results respectively obtained by the trained neural network estimation model and the extended Kalman filtering estimation model to obtain a final estimation result.
In the present embodiment, the actual values of the steering wheel angle and the longitudinal acceleration include the centroid slip angle, the yaw rate, and the longitudinal vehicle speed.
In the embodiment, the neural network adopts a radial basis function neural network to predict the relevant vehicle parameters, wherein the neural network implies that the layer node q is 9, and the nodes of the input and output layers are 1.
The neural network training process is divided into two stages, namely a non-instructor learning stage and an instructor learning stage.
The tutor-free learning stage specifically comprises the following steps:
(1) given the initial center vector C of each hidden nodei(0) (i ═ 1,2,. q), learning rate β (0)<β(0)<1) And a threshold value epsilon for judging to stop the calculation
(2) Node for calculating Euclidean distance and calculating minimum distance
In the formula, k is a sample serial number; r is a central vector ci(k-1) hidden node sequence number closest to input sample x (k);
(3) center of adjustment
In the formula, β (k) is a learning rate. β (k) ═ β (k-1)/(1+ int (k/q))1/2(ii) a int (·) denotes rounding (·);
(4) determining cluster quality
The above (2) and (3) were repeated for all samples k (k ═ 1, 2.., N) until the following was satisfied
If the formula is satisfied, finishing clustering;
the learning stage of the instructor is as follows:
when c is going toiAfter the determination, training the weight between the hidden layer and the output layer, and if the weight is a linear equation set, the problem of linear optimization becomes full-time; connection weight w between hidden layer and output layer of radial basic bible networkkiThe (k 1, 2.. times.l; i 1, 2.. times.q) learning algorithm is
wki(k+1)＝wki(k)+η(tk-yk)ui(x)/uTu
Wherein u is [ u ]1(x),u2(x),..,uq(x)]T；ui(x) Is a Gaussian function; η is the learning rate.
In this embodiment, the extended kalman filter estimation model specifically includes: the vehicle three-degree-of-freedom dynamic model is used as a state equation and an observation equation, and the centroid slip angle, the yaw angular velocity and the longitudinal vehicle speed are estimated based on the extended Kalman filtering algorithm
In the formula, omega is yaw angular velocity, beta is barycenter slip angle and vxIs the longitudinal speed, k1And k2Yaw stiffness, a and b are the lengths from the center of mass of the vehicle to the front and rear axes, Iz is the moment of inertia, m is the total vehicle mass, delta is the steering wheel angle, axAnd ayLongitudinal acceleration and lateral acceleration.
In this embodiment, the signal fusion and optimization module compares the signals estimated by different algorithms with the reference signal to obtain different confidence degrees of the two estimated signals, and the signal with the higher confidence degree has a higher weight value, whereas the signal with the lower weight value. Specifically, the difference is made between the estimated signal and the reference signal according to the two algorithms, the magnitude of the two errors is compared within the estimated time of each step, the signal with small error is judged as the signal with high confidence, the assigned weight is larger, and otherwise, the assigned weight is small. Wherein, the sum of the weights of the two is 1.
In this embodiment, the signal fusion and optimization process specifically includes:
the calculated values of the centroid slip angle, the horizontal plaque angular velocity and the longitudinal vehicle speed can be obtained through a theoretical calculation method, but the calculated values are caused by error accumulation of the sensors and error accumulation of integral. Over time, the confidence of the calculated values is reduced and therefore can only be used as a reference for one signal.
And the signal fusion part takes the theoretical value as a calibration object, respectively uses two different algorithms to be the worst with the reference signal, and uses a dichotomy to search the weight suitable for each signal, wherein the signal with small error has large full weight and the signal with large error has small weight. Finally, the update value is the addition of each signal multiplied by its own weight.
Centroid slip angle:
yaw rate:
longitudinal vehicle speed:
wherein: omega and vxIs the yaw rate and the longitudinal speed, a, estimated by the neural networkyAs a lateral acceleration, ntFor the wheel speed,/wThe tire circumference.
The above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.
Claims (8)
1. A multi-source fusion vehicle state parallel estimation method is characterized by comprising the following steps:
s1, acquiring real values corresponding to the steering wheel corner and the longitudinal acceleration of the vehicle at each moment in the running process, and dividing the real values into a training set and a test set according to the proportion after preprocessing;
step S2, constructing a neural network, training a training set and obtaining a trained neural network estimation model;
s3, constructing an extended Kalman filtering estimation model;
step S4: in the running process of a vehicle, a steering wheel corner sensor and a longitudinal acceleration sensor are used for acquiring a steering wheel corner and a longitudinal acceleration, and the steering wheel corner and the longitudinal acceleration are respectively transmitted to a trained neural network estimation model and an extended Kalman filtering estimation model for processing;
step S5: and performing signal fusion and optimization processing according to estimation results respectively obtained by the trained neural network estimation model and the extended Kalman filtering estimation model to obtain a final estimation result.
2. The multi-source fused vehicle state parallel estimation method according to claim 1, wherein the real values of the steering wheel angle and the longitudinal acceleration comprise a centroid slip angle, a yaw rate and a longitudinal vehicle speed.
3. The multi-source fused vehicle state parallel estimation method according to claim 1, wherein the neural network predicts relevant vehicle parameters by using a radial basis function neural network, wherein a hidden layer node q of the neural network is 9, and nodes of input and output layers are 1.
4. The multi-source fused vehicle state parallel estimation method according to claim 3, wherein the neural network training process is divided into two stages, namely a non-instructor learning stage and an instructor learning stage.
5. The multi-source fused vehicle state parallel estimation method according to claim 4,
the tutor-free learning stage specifically comprises the following steps:
(1) given each hidden nodeInitial center vector C of pointsi(0) (i ═ 1,2,. q), learning rate β (0)<β(0)<1) And a threshold value epsilon for judging to stop the calculation
(2) Node for calculating Euclidean distance and calculating minimum distance
In the formula, k is a sample serial number; r is a central vector ci(k-1) hidden node sequence number closest to input sample x (k);
(3) center of adjustment
In the formula, β (k) is a learning rate. β (k) ═ β (k-1)/(1+ int (k/q))1/2(ii) a int (·) denotes rounding (·);
(4) determining cluster quality
Repeating the above (2) and (3) for all samples k (k ═ 1, 2., N) until the following formula condition is satisfied, and then finishing clustering;
6. the multi-source fused vehicle state parallel estimation method according to claim 4,
the instructor learning stage specifically comprises:
when c is going toiAfter the determination, training the weight between the hidden layer and the output layer, and if the weight is a linear equation set, the problem of linear optimization becomes full-time; connection weight w between hidden layer and output layer of radial basic bible networkkiThe (k 1, 2.. times.l; i 1, 2.. times.q) learning algorithm is
wki(k+1)＝wki(k)+η(tk-yk)ui(x)/uTu
Wherein u is [ u ]1(x),u2(x),..,uq(x)]T；ui(x) Is a Gaussian function; η is the learning rate.
7. The multi-source fusion vehicle state parallel estimation method according to claim 1, wherein the extended kalman filter estimation model is specifically: the vehicle three-degree-of-freedom dynamic model is used as a state equation and an observation equation, and the centroid slip angle, the yaw angular velocity and the longitudinal vehicle speed are estimated based on the extended Kalman filtering algorithm
In the formula, omega is yaw angular velocity, beta is barycenter slip angle and vxIs the longitudinal speed, k1And k2Yaw stiffness, a and b are the lengths from the center of mass of the vehicle to the front and rear axes, Iz is the moment of inertia, m is the total vehicle mass, delta is the steering wheel angle, axAnd ayLongitudinal acceleration and lateral acceleration.
8. The multi-source-fused vehicle state parallel estimation method according to claim 1, wherein the signal fusion and optimization module compares signals estimated by different algorithms with a reference signal to obtain different confidence degrees of the two estimated signals, the signal with the high confidence degree is weighted more heavily, and the weighted value is weighted less heavily.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011309519.XA CN112417598A (en) | 2020-11-20 | 2020-11-20 | Multi-source fusion vehicle state parallel estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011309519.XA CN112417598A (en) | 2020-11-20 | 2020-11-20 | Multi-source fusion vehicle state parallel estimation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112417598A true CN112417598A (en) | 2021-02-26 |
Family
ID=74774875
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011309519.XA Pending CN112417598A (en) | 2020-11-20 | 2020-11-20 | Multi-source fusion vehicle state parallel estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112417598A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113916565A (en) * | 2021-12-14 | 2022-01-11 | 禾多科技（北京）有限公司 | Steering wheel zero deflection angle estimation method and device, vehicle and storage medium |
| CN113937893A (en) * | 2021-10-22 | 2022-01-14 | 大连交通大学 | Rail vehicle running state sensing system and method based on machine learning |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108545081A (en) * | 2018-03-20 | 2018-09-18 | 北京理工大学 | Slip angle estimation method and system based on robust Unscented kalman filtering |
| CN108715166A (en) * | 2018-04-28 | 2018-10-30 | 南京航空航天大学 | Intact stability index method of estimation based on deep learning |
| CN109466558A (en) * | 2018-10-26 | 2019-03-15 | 重庆邮电大学 | A kind of coefficient of road adhesion estimation method based on EKF and BP neural network |
-
2020
- 2020-11-20 CN CN202011309519.XA patent/CN112417598A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108545081A (en) * | 2018-03-20 | 2018-09-18 | 北京理工大学 | Slip angle estimation method and system based on robust Unscented kalman filtering |
| CN108715166A (en) * | 2018-04-28 | 2018-10-30 | 南京航空航天大学 | Intact stability index method of estimation based on deep learning |
| CN109466558A (en) * | 2018-10-26 | 2019-03-15 | 重庆邮电大学 | A kind of coefficient of road adhesion estimation method based on EKF and BP neural network |
Non-Patent Citations (5)
| Title |
|---|
| 宗长富 等: "基于扩展卡尔曼滤波的信息融合技术在车辆状态估计中的应用", 《机械工程学报》 * |
| 张凤娇 等: "基于深度学习的极限工况下车辆的状态估计", 《重庆理工大学学报(自然科学)》 * |
| 王震坡 等: "基于自适应无迹卡尔曼滤波的分布式驱动电动汽车车辆状态参数估计", 《北京理工大学学报》 * |
| 胡盛斌: "RBF神经网络", 《非线性多关节机器人系统滑模控制》 * |
| 赵万忠等: "基于无迹卡尔曼滤波的汽车状态参数估计", 《华南理工大学学报(自然科学版)》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113937893A (en) * | 2021-10-22 | 2022-01-14 | 大连交通大学 | Rail vehicle running state sensing system and method based on machine learning |
| CN113916565A (en) * | 2021-12-14 | 2022-01-11 | 禾多科技（北京）有限公司 | Steering wheel zero deflection angle estimation method and device, vehicle and storage medium |
| CN113916565B (en) * | 2021-12-14 | 2022-03-11 | 禾多科技（北京）有限公司 | Steering wheel zero deflection angle estimation method and device, vehicle and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112417598A (en) | Multi-source fusion vehicle state parallel estimation method | |
| CN108715166B (en) | Vehicle stability index estimation method based on deep learning | |
| CN110688729B (en) | LSTM-IDM (least squares-inverse discrete cosine transform) following characteristic fusion method based on adaptive Kalman filtering, storage medium and equipment | |
| CN110588657B (en) | Joint estimation method for vehicle motion state and road gradient | |
| CN113447021B (en) | MEMS inertial navigation system positioning enhancement method based on LSTM neural network model | |
| CN113408047B (en) | Vehicle dynamics prediction model based on time-lag feedback neural network, training data acquisition method and training method | |
| CN108931233B (en) | Road side slope value detection method and device | |
| CN115406446A (en) | Multi-axis special vehicle state estimation method based on neural network and unscented Kalman filtering | |
| CN112373484B (en) | Method for acquiring vehicle mass dynamics based on feedforward neural network | |
| CN107741269B (en) | Weighing sensor test compensation method based on fuzzy recognition | |
| CN112270039A (en) | Distributed asynchronous fusion-based nonlinear state estimation method for drive-by-wire chassis vehicle | |
| CN117719519A (en) | Vehicle running state estimation method | |
| CN112550294B (en) | Path tracking control method based on vehicle fault signal isolation | |
| CN113030940B (en) | Multi-star convex type extended target tracking method under turning maneuver | |
| CN109916484B (en) | Combined weighing method and device for weighing equipment | |
| CN113705865B (en) | Automobile stability factor prediction method based on deep neural network | |
| CN109916483B (en) | Weighing equipment combined monitoring method and device | |
| CN114001759B (en) | Array MEMS sensor control method and system | |
| CN114061592B (en) | Adaptive robust AUV navigation method based on multiple models | |
| CN111461288B (en) | Full-speed segment detection method and system for geometric parameters of track | |
| CN112965965A (en) | Outlier elimination method and system based on fuzzy prediction system and computer related product | |
| CN113386781B (en) | Intelligent vehicle track tracking control method based on data-driven vehicle dynamics model | |
| Ghosn et al. | A Robust Hybrid Observer for Side-slip Angle Estimation | |
| CN113704684B (en) | Centralized fusion robust filtering method | |
| CN117574016A (en) | Rear wheel steering angle calculation method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210226 |
|
| RJ01 | Rejection of invention patent application after publication |