CN108828980B - Method and system for predicting over-bending centrifugal force - Google Patents
Method and system for predicting over-bending centrifugal force Download PDFInfo
- Publication number
- CN108828980B CN108828980B CN201810460791.4A CN201810460791A CN108828980B CN 108828980 B CN108828980 B CN 108828980B CN 201810460791 A CN201810460791 A CN 201810460791A CN 108828980 B CN108828980 B CN 108828980B
- Authority
- CN
- China
- Prior art keywords
- centrifugal force
- vehicle
- data
- model
- acceleration
- 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
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B17/00—Systems involving the use of models or simulators of said systems
- G05B17/02—Systems involving the use of models or simulators of said systems electric
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Tires In General (AREA)
- Force Measurement Appropriate To Specific Purposes (AREA)
Abstract
The invention discloses a method and a system for predicting over-bending centrifugal force, and belongs to the field of transportation equipment. A method of predicting over-bent centrifugal force, comprising the steps of: integrating real-time tire pressure data, acceleration data, speed data and torque data of a vehicle; 2, generating a time-sequence feature vector from the integrated data; 3, training a prediction model of the centrifugal force according to the feature vector; and 4, after the prediction model is obtained, inputting the characteristic vector into the prediction model to obtain a predicted value of the centrifugal force of the vehicle for a period of time later. According to the method, a data modeling is carried out by utilizing a deep neural network of machine learning through massive driving sensor data, and the correlation between the tire pressure, the acceleration, the speed and the torque at the current moment and the centrifugal force borne by the vehicle at the later moment is found out, so that the aim of predicting the centrifugal force borne by the vehicle is fulfilled.
Description
Technical Field
The invention belongs to the technical field of vehicle control, and particularly relates to a method and a system for predicting an over-bending centrifugal force.
Background
With the increase of the container port throughput, the container handling efficiency is more and more emphasized by port operation managers. When shortening the transport time, improving container haulage vehicle's average speed, because tractor and planker are the flexible coupling, carry the goods on the planker and when crossing the bend, the gesture of vehicle vertical direction has the slope of certain degree, can not turn on one's side because of the speed is too fast when guaranteeing that the vehicle crosses the bend, need be with speed and acceleration control when crossing the bend in reasonable within range. Therefore, it is predicted that the lateral centrifugal force applied to the transport vehicle during the cornering becomes an important index for determining whether the vehicle can stably and safely negotiate a curve.
Disclosure of Invention
The invention provides a method and a system for predicting over-bending centrifugal force, which are used for predicting the transverse centrifugal force borne by a carrying vehicle in the over-bending process.
The invention can be realized by the following technical scheme:
a prediction method of over-bending centrifugal force is characterized by comprising the following steps: the method comprises the following steps:
step 1: integrating real-time tire pressure data, acceleration data, speed data and torque data of the vehicle;
step 2: generating the integrated data into a time-sequence feature vector;
and step 3: training a prediction model of the centrifugal force according to the feature vector;
and 4, step 4: after the prediction model is obtained, the feature vector is input to the prediction model, and a predicted value of the centrifugal force of the vehicle at a later time is obtained.
Further, the system in the step 1 detects and obtains tire pressure data, acceleration data, speed data and torque data of the vehicle in real time, ensures that time scales of all data are aligned, and outputs the data to the step 2 according to the sequence of the time sequence.
Further, in the step 2, the tire pressure difference of the left tire and the right tire, the lateral centrifugal force applied to the vehicle, the load of the vehicle and the acceleration and deceleration of the vehicle are calculated according to the input data, and are combined into a characteristic vector of t +1 seconds and a characteristic vector of t +2 seconds until the characteristic vector of t + n seconds is output to the step 3.
Further, step 3, the feature vectors are respectively input into the training process of the centrifugal force prediction model after t +1 second, t +2 seconds, … and t + n seconds as training data of the model, whether the prediction error of the model is converged is judged every time the training is completed, if the judgment result is that the model is not converged, the next batch of feature vectors are continuously waited, if the model is converged, the model is output to the vehicle centrifugal force simulator, and the simulator is output to step 4.
Further, in the training process of the centrifugal force prediction model, the model uses the structure of a deep neural network, the value of each parameter in the model is calculated through a training method of error reverse transfer, the model consists of a characteristic input layer, a hidden layer and an output layer, a characteristic vector enters from the characteristic input layer, and is output to the output layer after the kernel function operation of the multiple hidden layers, wherein the kernel function of the hidden layer is a nonlinear continuous conductible function.
Further, step 4, inputting the feature vector into a vehicle centrifugal force simulator, predicting a centrifugal force value for a period of time later through the simulator, and storing the centrifugal force prediction value with a time scale to represent a predicted value of the centrifugal force at the moment t +1, a predicted value of the centrifugal force at the moment t +2, a predicted value of the centrifugal force at the moment t +3, … and a predicted value of the centrifugal force at the moment t + n, which are borne by the vehicle when the vehicle passes a bend.
A system applying the method for predicting the over-bending centrifugal force is characterized in that: the device comprises a main controller, a tire pressure sensor, an acceleration sensor, a speed sensor and a torque sensor, wherein the tire pressure sensor, the speed sensor, the acceleration sensor and the torque sensor are respectively connected with the main controller, the main controller receives signals of the sensors, trains a centrifugal force prediction model according to real-time tire pressure, acceleration, speed and torque data, and predicts the over-bending centrifugal force of a vehicle in a later period by using the obtained centrifugal force prediction model.
Further, the tire pressure sensors are arranged on the tires at the front left, front right, rear left and rear right positions.
Further, the acceleration sensor comprises a running acceleration sensor parallel to the running direction of the vehicle and a transverse acceleration sensor perpendicular to the running direction of the vehicle and parallel to the ground, and the running acceleration sensor and the transverse acceleration sensor are respectively connected with the main controller.
Advantageous effects
According to the technical scheme, a data modeling is carried out through massive driving sensor data and a deep neural network learned by a machine, and the correlation between the tire pressure, the acceleration, the speed and the torque at the current moment and the centrifugal force borne by the vehicle at the later moment is found out, so that the aim of predicting the centrifugal force borne by the vehicle is fulfilled; the method adopts a deep learning modeling mode to predict the centrifugal force borne by the vehicle, changes the control lag of the prior art which is corrected based on measurement data, can predict the centrifugal force of the vehicle, not only can improve the safety of the vehicle in the process of driving and turning, but also can provide assistance for the control of the driving process.
Drawings
FIG. 1 is a flow chart of the centrifugal force prediction of the present invention;
FIG. 2 is a flow chart of predictive model training of the present invention;
FIG. 3 is a schematic diagram of a deep neural network of the present invention;
FIG. 4 is a flow chart of the centrifugal force prediction calculation of the present invention;
FIG. 5 is a schematic diagram of a predicted value of centrifugal force corresponding to a vehicle cornering process according to the present invention;
FIG. 6 is a schematic diagram of the system of the present invention.
Detailed Description
The present invention will be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings.
The system is based on data collected by a vehicle control platform, and obtains a predicted value of the transverse centrifugal force of the vehicle for a period of time after feature vector generation, a centrifugal force prediction model and centrifugal force prediction.
The system comprises a main controller 5, a tire pressure sensor 6, an acceleration sensor 7, a speed sensor 8 and a torque sensor 9, wherein the tire pressure sensor 6, the speed sensor 8, the acceleration sensor 7 and the torque sensor 9 are respectively connected with the main controller 5, and the main controller 5 receives sensor signals to obtain real-time tire pressure, acceleration, speed and torque data. The acceleration sensor 7 includes a running acceleration sensor parallel to the running direction of the vehicle, and a lateral acceleration sensor perpendicular to the running direction of the vehicle and parallel to the ground, and the running acceleration sensor and the lateral acceleration sensor are respectively connected to the main controller 5.
The main controller in the scheme adopts a platform architecture processor of X86, runs a linux operating system, and can load and run small database software and other required middleware software.
The scheme is connected with an external sensor and a controller mainly in a CANBUS2.0 protocol mode.
The tire pressure sensors 6 are respectively installed in the left front, left rear, right front and right rear tires of the tractor or the left front, left rear, right front and right rear tires of the tractor and the planker, and real-time tire pressure data are collected and transmitted back to the controller.
The tire pressure data in the scheme is realized by directly measuring the tire pressure by using the pressure sensor installed in each tire, transmitting pressure information from the inside of the tire to a tire pressure detection unit (TPMS) of a vehicle-mounted controller through a wireless transmitter, and then transmitting the pressure information to a main controller through a CANUSB2.0 bus by a tire pressure detection module.
The tire pressure data in this scheme needs to include the pressure data of the tire on the front left, front right, back left, back right side position.
The lateral acceleration sensor is arranged on a chassis of the tractor or the carriage, collects real-time lateral acceleration data of the vehicle and transmits the data back to the controller, and the controller calculates the centrifugal force borne by the trailer according to the lateral acceleration data.
The running acceleration sensor is arranged on the chassis of the tractor, the speed sensor is arranged on the wheels of the tractor, the real-time acceleration and speed data of the running direction of the vehicle are collected and transmitted back to the controller, and the controller calculates the load of the vehicle according to the acceleration and speed data.
In order to acquire accurate acceleration signals and angular velocity data in real time, the present embodiment introduces an Inertial Measurement Unit (IMU) unit (acceleration sensor 7). The main controller is directly connected with the inertia measuring device to collect driving acceleration data, transverse acceleration data and relevant angular velocity data in all directions fed back by the inertia measuring device in real time.
The CANBUS2.0 protocol is used for communication between the main controller 5 and the inertial measurement unit (acceleration sensor 7).
The speed sensor 8 is a device for detecting the vehicle speed of the vehicle, and the output signal of the vehicle speed sensor may be a magnetoelectric ac signal, a hall digital signal, or a photoelectric digital signal.
The vehicle speed signal in the scheme is directly from any one of the signals, and the main controller obtains a square wave signal through rectification and then calculates the number of square waves in a period to obtain the instantaneous vehicle speed.
The torque sensor 9 is arranged on the output side of the electric engine or the driving motor, collects real-time torque data and transmits the real-time torque data back to the controller, and the controller calculates the acceleration and deceleration of the vehicle.
The engine torque (torque sensor 9) in the scheme is obtained by the vehicle control platform after relevant data is obtained, and actual torque data is obtained through calculation.
The torque data is transmitted by the vehicle control platform to the master controller 5 in real time via the CANBUS2.0 bus. The associated braking subsystem of the vehicle control platform or the vehicle control platform itself is represented by a final drive 10.
After the system acquires the tire pressure data, the acceleration data, the speed data and the torque data by adopting the device, the tire pressure difference of the left and right tires, the transverse centrifugal force borne by the vehicle, the load of the vehicle and the acceleration and deceleration of the vehicle at the current moment are calculated. And obtaining past tire pressure data, acceleration data, speed data, torque data, tire pressure difference of left and right tires, transverse centrifugal force applied to the vehicle, vehicle load, acceleration and deceleration of the vehicle and the like from a historical library to combine into a characteristic vector of t +1 second and a characteristic vector of t +2 seconds till the characteristic vector of t + n seconds.
And respectively inputting the feature vector time marks into a feature input layer of the centrifugal force prediction model after t +1 second, t +2 seconds, … and t + n seconds, and outputting the feature input layer to an output layer through nonlinear conversion of hidden layers 1 to n, wherein the nonlinear conversion of the hidden layers is completed by a kernel function, the kernel function uses a nonlinear and continuously-derivable function, and parameters of the kernel function use parameters updated in the previous round in the nonlinear conversion process.
In the training process, the output value of the output layer is the predicted value of the centrifugal force borne by the vehicle, and the parameters of the kernel function are updated layer by layer from the hidden layer n to the hidden layer 1 by the error back propagation mode of the predicted value and the measured value.
And after one round of training is finished, judging whether the parameter change of the kernel functions of the hidden layer before and after updating is converged, if not, continuing waiting for the next batch of feature vectors, and if so, outputting the parameters of the kernel functions of the models to a vehicle centrifugal force simulator.
After a prediction model with converged parameters is obtained, input parameters are input into an input layer of a vehicle centrifugal force simulator, an output value of an output layer is obtained after nonlinear conversion of a hidden layer, the converged model parameters are used as parameters of a kernel function used in the conversion process, the output value is a centrifugal force predicted value from the t +1 moment to the t + n moment of a vehicle, and the difference between the calculation process of the simulator and the training process of the prediction model is that the parameters of the kernel function are updated without using a method of error back propagation in the calculation process of the simulator, so that the transverse centrifugal force borne by the vehicle at multiple moments in the vehicle bending process is predicted.
Example 1
The main controller 5 calculates tire pressure difference of left and right tires, lateral centrifugal force borne by the vehicle, vehicle load and acceleration and deceleration of the vehicle at the current time t after acquiring tire pressure data, acceleration data, torque data and vehicle speed data in real time, normalizes the numerical value to 0.2-0.6 and writes the numerical value into a Redis database.
Tire pressure, acceleration, torque, vehicle speed, tire pressure difference of left and right tires, transverse centrifugal force borne by the vehicle, vehicle load, acceleration and deceleration of the vehicle, difference values of the quantities, a 2-power value and a 3-power value from t moment to t-10 moment are obtained from a Redis database and input into a feature input layer of the prediction model as input feature vectors, a sigmoid function is used as a kernel function by a hidden layer, and the result is converted by the hidden layer to output a predicted value of the centrifugal force to an output layer of the prediction model as the predicted value of the transverse centrifugal force borne by the vehicle for t +1 second.
And when the transverse centrifugal force of the vehicle in t +1 seconds is obtained through calculation, the transverse centrifugal force is used as an actual measured value of the transverse centrifugal force of the vehicle in t +1 seconds, parameters of all sigmoid functions in the prediction model are updated by using an error back propagation algorithm proposed by Rumelhart in 1986, and when the absolute value of the variation of the parameters is less than 1E-10, the parameters are judged to be converged, and the model is considered to be trained and output to a vehicle centrifugal force simulator.
When the trained model is adopted for actual prediction, real-time tire pressure data, acceleration data, torque data and vehicle speed data are also processed, calculating the tire pressure difference of left and right tires, the transverse centrifugal force borne by the vehicle, the vehicle load and the acceleration and deceleration of the vehicle at the current time t, normalizing the numerical values to 0.2-0.6, writing the numerical values into a Redis database, obtaining the tire pressure, the acceleration, the torque, the vehicle speed, the tire pressure difference of the left and right tires, the transverse centrifugal force borne by the vehicle, the vehicle load, the acceleration and deceleration of the vehicle, the difference value of the quantities, the 2-power value and the 3-power value from the Redis database, inputting the tire pressure, the acceleration, the torque, the vehicle speed, the tire pressure difference of the left and right tires, the transverse centrifugal force borne by the vehicle, the vehicle load, the acceleration and deceleration of the vehicle, the 2-power value and the 3-power value into a characteristic input layer of a prediction model, using a sigmoid function as a kernel function by a hidden layer, outputting the predicted value of the centrifugal.
According to the technical scheme, a data modeling is carried out through massive driving sensor data and a deep neural network learned by a machine, and the correlation between the tire pressure, the acceleration, the speed and the torque at the current moment and the centrifugal force borne by the vehicle at the later moment is found out, so that the aim of predicting the centrifugal force borne by the vehicle is fulfilled; the method adopts a deep learning modeling mode to predict the centrifugal force borne by the vehicle, changes the control lag of the prior art which is corrected based on measurement data, can predict the centrifugal force of the vehicle, not only can improve the safety of the vehicle in the process of driving and turning, but also can provide assistance for the control of the driving process.
Variations and modifications to the above-described embodiments may become apparent to those skilled in the art from the disclosure and teachings of the above specification. Therefore, the present invention is not limited to the above-mentioned embodiments, and any obvious improvement, replacement or modification made by those skilled in the art based on the present invention is within the protection scope of the present invention.
In addition, although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.
Claims (7)
1. A prediction method of over-bending centrifugal force is characterized by comprising the following steps: the method comprises the following steps:
step 1: integrating real-time tire pressure data, acceleration data, speed data and torque data of the vehicle;
step 2: generating the integrated data into a time-sequence feature vector;
and step 3: training a prediction model of the centrifugal force according to the feature vector;
and 4, step 4: after the prediction model is obtained, inputting the characteristic vector into the prediction model to obtain a predicted value of the centrifugal force of the vehicle for a period of time;
calculating the tire pressure difference of the left tire and the right tire, the transverse centrifugal force borne by the vehicle, the vehicle load and the acceleration and deceleration of the vehicle according to the input data in the step 2, combining the tire pressure difference, the transverse centrifugal force borne by the vehicle, the vehicle load and the acceleration and deceleration of the vehicle into a characteristic vector of t +1 seconds and a characteristic vector of t +2 seconds, and outputting the characteristic vectors of t + n seconds to the step 3;
and 3, respectively inputting the characteristic vectors serving as training data of the model into the training process of the centrifugal force prediction model after t +1 second, t +2 seconds, … seconds and t + n seconds, judging whether the model is converged or not after finishing training once, namely judging whether the prediction error of the model is converged or not, if the judgment result is that the model is not converged, continuing waiting for the next batch of characteristic vectors, and if the model is converged, outputting the model to a vehicle centrifugal force simulator, and outputting the simulator to the step 4.
2. The method of predicting an over-bending centrifugal force of claim 1, wherein: and (3) detecting in real time by the system in the step 1 to obtain tire pressure data, acceleration data, speed data and torque data of the vehicle, ensuring the time scales of all the data to be aligned, and outputting to the step 2 according to the sequence of the time sequence.
3. The method of predicting an over-bending centrifugal force of claim 1, wherein: in the training process of the centrifugal force prediction model, the model uses the structure of a deep neural network, the value of each parameter in the model is calculated by a training method of error reverse transfer, the model consists of a characteristic input layer, a hidden layer and an output layer, a characteristic vector enters from the characteristic input layer, is output to the output layer after the kernel function operation of a plurality of layers of hidden layers, and the kernel function of the hidden layer is a nonlinear continuous derivative function.
4. The method of predicting an over-bending centrifugal force of claim 1, wherein: and 4, firstly inputting the characteristic vector into a vehicle centrifugal force simulator, predicting a centrifugal force value for a period of time later through the simulator, and storing the centrifugal force prediction value with a time scale so as to represent the predicted value of the centrifugal force at the moment t +1, the predicted value of the centrifugal force at the moment t +2, the predicted value of the centrifugal force at the moment t +3 and the predicted values of the centrifugal force at the moments … and t + n when the vehicle passes a bend.
5. A system applying the method of predicting over-bent centrifugal force of claim 1, wherein: the device comprises a main controller, a tire pressure sensor, an acceleration sensor, a speed sensor and a torque sensor, wherein the tire pressure sensor, the speed sensor, the acceleration sensor and the torque sensor are respectively connected with the main controller, the main controller receives signals of the sensors, and a centrifugal force prediction model is trained according to real-time tire pressure, acceleration, speed and torque data;
calculating the tire pressure difference of left and right tires, the transverse centrifugal force borne by the vehicle, the load of the vehicle and the acceleration and deceleration of the vehicle according to the input data, combining the tire pressure difference, the transverse centrifugal force borne by the vehicle and the acceleration and deceleration of the vehicle into a characteristic vector of t +1 seconds and a characteristic vector of t +2 seconds, outputting the characteristic vectors of t + n seconds, and training a centrifugal force prediction model according to the characteristic vectors;
and respectively inputting the characteristic vectors as training data of the model into the training process of the centrifugal force prediction model after t +1 second, t +2 seconds, … and t + n seconds, judging whether the model converges or not every time the training is finished, namely judging whether the prediction error of the model converges or not, if the judgment result is that the model does not converge, continuously waiting for the next batch of characteristic vectors, if the model converges, outputting the model into a vehicle centrifugal force simulator, outputting the simulator, and predicting the over-bending centrifugal force of the vehicle at a later period by using the obtained centrifugal force prediction model.
6. The system of claim 5, wherein: the tire pressure sensors are arranged on the tires on the front left, front right, back left and back right positions.
7. The system of claim 5, wherein: the acceleration sensor comprises a running acceleration sensor parallel to the running direction of the vehicle and a transverse acceleration sensor perpendicular to the running direction of the vehicle and parallel to the ground, and the running acceleration sensor and the transverse acceleration sensor are respectively connected with the main controller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810460791.4A CN108828980B (en) | 2018-05-15 | 2018-05-15 | Method and system for predicting over-bending centrifugal force |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810460791.4A CN108828980B (en) | 2018-05-15 | 2018-05-15 | Method and system for predicting over-bending centrifugal force |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108828980A CN108828980A (en) | 2018-11-16 |
| CN108828980B true CN108828980B (en) | 2021-06-11 |
Family
ID=64148731
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810460791.4A Active CN108828980B (en) | 2018-05-15 | 2018-05-15 | Method and system for predicting over-bending centrifugal force |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108828980B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111547044B (en) * | 2020-03-31 | 2022-01-28 | 桂林电子科技大学 | Truck curve automatic driving training method and system |
| CN114234972B (en) * | 2022-03-01 | 2022-05-24 | 长沙金维信息技术有限公司 | Inertial navigation method for inertial navigation system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1749722A1 (en) * | 2005-08-01 | 2007-02-07 | Delphi Technologies, Inc. | Rollover warning and detection method for transport vehicles |
| CN101027961A (en) * | 2006-03-01 | 2007-09-05 | 株式会社久保田 | Detecting device and controlling device for working vehicle postures |
| JP4244453B2 (en) * | 1999-07-21 | 2009-03-25 | 株式会社デンソー | Body slip angle estimation method and apparatus |
| CN102556160A (en) * | 2010-12-29 | 2012-07-11 | 天津市友达机电液成套设备有限公司 | Turning radius display and vehicle speed alarm method for wheeled vehicle and device thereof |
| CN102923024A (en) * | 2011-08-08 | 2013-02-13 | 林其禹 | Fully autonomous seat system for automobile and execution method for fully autonomous seat system for automobile |
| CN105117524A (en) * | 2015-07-30 | 2015-12-02 | 西安航空制动科技有限公司 | Dynamic simulation method of using differential braking to control aircraft turning process |
| CN106494406A (en) * | 2015-09-08 | 2017-03-15 | 星克跃尔株式会社 | Bend guidance method, bend guider, electronic installation and program |
| CN107972672A (en) * | 2016-10-21 | 2018-05-01 | 奥迪股份公司 | Drive assist system and driving assistance method |
-
2018
- 2018-05-15 CN CN201810460791.4A patent/CN108828980B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4244453B2 (en) * | 1999-07-21 | 2009-03-25 | 株式会社デンソー | Body slip angle estimation method and apparatus |
| EP1749722A1 (en) * | 2005-08-01 | 2007-02-07 | Delphi Technologies, Inc. | Rollover warning and detection method for transport vehicles |
| CN101027961A (en) * | 2006-03-01 | 2007-09-05 | 株式会社久保田 | Detecting device and controlling device for working vehicle postures |
| CN102556160A (en) * | 2010-12-29 | 2012-07-11 | 天津市友达机电液成套设备有限公司 | Turning radius display and vehicle speed alarm method for wheeled vehicle and device thereof |
| CN102923024A (en) * | 2011-08-08 | 2013-02-13 | 林其禹 | Fully autonomous seat system for automobile and execution method for fully autonomous seat system for automobile |
| CN105117524A (en) * | 2015-07-30 | 2015-12-02 | 西安航空制动科技有限公司 | Dynamic simulation method of using differential braking to control aircraft turning process |
| CN106494406A (en) * | 2015-09-08 | 2017-03-15 | 星克跃尔株式会社 | Bend guidance method, bend guider, electronic installation and program |
| CN107972672A (en) * | 2016-10-21 | 2018-05-01 | 奥迪股份公司 | Drive assist system and driving assistance method |
Non-Patent Citations (1)
| Title |
|---|
| 数学建模神经网络预测模型及程序;未显示;《百度文库》;20160828;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108828980A (en) | 2018-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP2882232B2 (en) | Vehicle weight center slip angle measuring device | |
| CN103407451B (en) | A kind of road longitudinal and additional forces method of estimation | |
| US10279815B2 (en) | Vehicle controls based on the measured weight of freight | |
| CN106004994A (en) | Continuous estimation of surface friction coefficient based on eps and vehicle models | |
| CN103180179B (en) | Sensor device for automobile | |
| CN104973052A (en) | Speed bump detection apparatus and navigation data updating apparatus and method using the same | |
| CN111201173B (en) | Method for determining a friction value of a contact between a vehicle tire and a lane and method for controlling a vehicle function of a vehicle | |
| CN109606378A (en) | Vehicle running state estimation method towards non-Gaussian noise environment | |
| CN103029705A (en) | Technique to improve mileage and driving stability for vehicle | |
| CN103279675B (en) | Tire-road attachment coefficient and the method for estimation of slip angle of tire | |
| CN113498529B (en) | Target tracking method and device | |
| CN108828980B (en) | Method and system for predicting over-bending centrifugal force | |
| CN110687907B (en) | Intelligent automobile decoupling prediction controller based on model dynamic reconstruction and control method thereof | |
| CN101650248A (en) | Dynamic detection method of barycenter height of truck | |
| CN110386189A (en) | Interference signal is accessed into datum quantity in Cascade control | |
| CN107585207B (en) | A kind of vehicle line traffic control four-wheel steering system and its control method | |
| CN115432009B (en) | Automatic driving vehicle track tracking control system | |
| CN110910531A (en) | Rapid pavement friction coefficient detection method based on vehicle-mounted OBD information | |
| JP5337090B2 (en) | Vehicle characteristic information estimation device and warning device using the same | |
| CN110435564A (en) | Determine vehicle drive behavior | |
| CN113341994B (en) | Intelligent automobile path tracking control method based on optimal control of piecewise affine system | |
| KR20180116663A (en) | System and method for converting autonomous driving control | |
| CN104670222B (en) | The driving path temporal voting strategy module and Ride Control System and method of sliding-modes | |
| CN113791598B (en) | Four-wheel moment distribution ring testing device under extreme working condition and torque optimizing method | |
| CN113479177B (en) | Vehicle brake control method, system, computer device, and storage medium |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220907 Address after: Room 643, 999 Changning Road, Changning District, Shanghai 200050 Patentee after: SHANGHAI WISDOM INFORMATION TECHNOLOGY CO.,LTD. Address before: No.163 Haibin Middle Road, Haiyang City, Yantai City, Shandong Province Patentee before: MACHINERY TECHNOLOGY (SHAN DONG) HEAVY INDUSTRY Co.,Ltd. Patentee before: SHANGHAI WISDOM INFORMATION TECHNOLOGY CO.,LTD. |
|
| TR01 | Transfer of patent right |