CN115931095B - Vehicle tail plate weighing method, device, equipment and storage medium - Google Patents
Vehicle tail plate weighing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN115931095B CN115931095B CN202310141151.8A CN202310141151A CN115931095B CN 115931095 B CN115931095 B CN 115931095B CN 202310141151 A CN202310141151 A CN 202310141151A CN 115931095 B CN115931095 B CN 115931095B
- Authority
- CN
- China
- Prior art keywords
- data
- model
- sensor
- descending
- ascending
- 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 126
- 238000005303 weighing Methods 0.000 title claims abstract description 64
- 230000001174 ascending effect Effects 0.000 claims abstract description 97
- 230000000630 rising effect Effects 0.000 claims abstract description 51
- 238000012545 processing Methods 0.000 claims abstract description 35
- 238000007781 pre-processing Methods 0.000 claims abstract description 14
- 230000008569 process Effects 0.000 claims description 81
- 238000012549 training Methods 0.000 claims description 58
- 230000006870 function Effects 0.000 claims description 28
- 230000001133 acceleration Effects 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 13
- 238000011478 gradient descent method Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 230000003068 static effect Effects 0.000 description 8
- 238000009434 installation Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000007306 turnover Effects 0.000 description 2
- 101100001671 Emericella variicolor andF gene Proteins 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005489 elastic deformation Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Vehicle Body Suspensions (AREA)
- Force Measurement Appropriate To Specific Purposes (AREA)
Abstract
The invention relates to the field of artificial intelligence, and discloses a vehicle tail plate weighing method, device and equipment and a storage medium, which are used for improving the weighing precision of a vehicle tail plate. The method comprises the following steps: performing data compensation and data standardization processing on the first sensor data to obtain first standard data; constructing a first rising data set and a first falling data set according to the first standard data; respectively constructing a target ascending model and a target descending model according to the first ascending data set and the first descending data set; collecting second sensor data, and preprocessing the second sensor data to obtain second standard data; and constructing a second ascending data set and a second descending data set according to the second standard data, inputting the second ascending data set and the second descending data set into a target ascending model and a target descending model respectively, and subtracting the output result of the target ascending model from the output result of the target descending model to obtain the weight of the object.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a vehicle tail plate weighing method, device and equipment and a storage medium.
Background
With the continuous improvement of living standard, the urban high-speed promotion is realized, and the demand for cargo transportation in many industries is increasing. In recent years, the requirements for truck mounting tail boards are remarkably increased, the national tail board mounting rate is higher and higher, and the tail board market development is in good opportunity. Correspondingly, the need for tailgate weighing is also increasing in order to better monitor the vehicle's process of transporting cargo.
At present, in the logistics industry, no matter the size of the express delivery volume, the weighing flow of the express delivery is manually operated by manpower, so that the problems of high labor cost, low efficiency and the like are solved; in the sanitation industry, the garbage collection and transportation process cannot monitor the weight of garbage in real time, so that the phenomena of transportation data originality, islanding and the like occur, decision making of the supervision department is influenced, and the garbage is difficult to trace in the transportation process. In the aspect of vehicle-mounted weighing, the weighed object is measured based on sensor data in combination with a related algorithm, and the data is automatically stored for subsequent analysis. The model currently used is mostly a traditional linear regression model that fits too single to the data. The weight of the related sensor and the weight of the object to be weighed are nonlinear, and when the number of vehicles is large to a certain extent, the model needs to be calibrated repeatedly, so that the problems of low installation efficiency and low model generalization capability exist. And for dynamic weighing of tail plate lifting, the model is a static model, cannot adapt to the dynamic process of tail plate lifting and lowering, and can only weigh at a fixed height. Meanwhile, the method cannot be used for multiple vehicles or multiple vehicle types, and the problem of large data error can exist in different tail plate states.
Disclosure of Invention
The invention provides a vehicle tail plate weighing method, device, equipment and storage medium, which are used for improving the weighing precision of a vehicle tail plate.
The first aspect of the invention provides a vehicle tail plate weighing method, which comprises the following steps: training data acquisition is carried out on a vehicle tail board based on preset sensor point positions, so that first sensor data are obtained; performing data compensation and data standardization processing on the first sensor data to obtain first standard data; constructing a first ascending data set and a first descending data set corresponding to the vehicle tail board according to the first standard data; respectively constructing a rising model and a falling model according to the first rising data set and the first falling data set, and carrying out model training on the rising model and the falling model to obtain a target rising model and a target falling model; collecting second sensor data to be processed, and preprocessing the second sensor data to obtain second standard data; and constructing a second ascending data set and a second descending data set according to the second standard data, inputting the second ascending data set and the second descending data set into the target ascending model and the target descending model respectively, and subtracting the output result of the target ascending model from the output result of the target descending model to obtain the weight of the object.
With reference to the first aspect, in a first implementation manner of the first aspect of the present invention, the acquiring training data of the vehicle tail board based on the preset sensor point location to obtain first sensor data includes: installing a plurality of sensors corresponding to a vehicle tail board based on preset sensor points, wherein the plurality of sensors comprise a hydraulic sensor for measuring lifting pressure, a hydraulic sensor for measuring overturning pressure, an inertial sensor for measuring lifting angle and acceleration and an inertial sensor for measuring overturning angle; and acquiring sensor data of the vehicle tail board in the multiple tail board lifting processes based on the plurality of sensors to obtain first sensor data.
With reference to the first aspect, in a second implementation manner of the first aspect of the present invention, the performing data compensation and data normalization processing on the first sensor data to obtain first standard data includes: performing data compensation on the first sensor data to obtain compensated data; and carrying out data standardization processing on the compensated data to obtain first standard data.
With reference to the first aspect, in a third implementation manner of the first aspect of the present invention, the performing data compensation on the first sensor data to obtain compensated data includes: constructing a compensation dataset based on the first sensor data; fitting the objective function according to the compensation data set to obtain an error function; and correcting the reading of the hydraulic sensor according to the error function to obtain compensated data.
With reference to the first aspect, in a fourth implementation manner of the first aspect of the present invention, the performing data normalization processing on the compensated data to obtain first standard data includes: respectively extracting lifting hydraulic pressure sensor data, overturning hydraulic pressure sensor data, lifting angle sensor data and overturning angle sensor data corresponding to the compensated data; respectively calculating the lifting hydraulic pressure sensor data, the overturning hydraulic pressure sensor data, the lifting angle sensor data and the average value corresponding to the overturning angle sensor data to obtain a lifting hydraulic pressure sensor average value, an overturning hydraulic pressure sensor average value, a lifting angle sensor average value and an overturning angle sensor average value; and carrying out centering and standardization operation on the lifting hydraulic sensor average value, the overturning hydraulic sensor average value, the lifting angle sensor average value and the overturning angle sensor average value to obtain first standard data.
With reference to the first aspect, in a fifth implementation manner of the first aspect of the present invention, the constructing, according to the first standard data, a first rising data set and a first falling data set corresponding to the vehicle tail board includes: extracting time sequence data information corresponding to the first standard data; acquiring time sequence data of a rising process and time sequence data of a falling process of the first standard data based on the time sequence data information; and adding Gaussian noise into the time sequence data of the ascending process and the time sequence data of the descending process respectively to obtain a first ascending data set and a first descending data set corresponding to the vehicle tail board.
With reference to the first aspect, in a sixth implementation manner of the first aspect of the present invention, the building a lifting model and a falling model according to the first lifting data set and the first falling data set, and performing model training on the lifting model and the falling model to obtain a target lifting model and a target falling model, includes: respectively constructing an ascending model and a descending model according to the first ascending data set and the first descending data set; respectively setting loss functions of the rising model and the falling model; and updating network parameters through a gradient descent method based on the loss function, and performing model training on the ascending model and the descending model to obtain a target ascending model and a target descending model.
A second aspect of the present invention provides a vehicle tailgate weighing apparatus comprising:
the acquisition module is used for acquiring training data of the vehicle tail board based on preset sensor point positions to obtain first sensor data;
the processing module is used for carrying out data compensation and data standardization processing on the first sensor data to obtain first standard data;
the building module is used for building a first ascending data set and a first descending data set corresponding to the vehicle tail board according to the first standard data;
The training module is used for respectively constructing a rising model and a falling model according to the first rising data set and the first falling data set, and carrying out model training on the rising model and the falling model to obtain a target rising model and a target falling model;
the preprocessing module is used for acquiring second sensor data to be processed and preprocessing the second sensor data to obtain second standard data;
and the calculation module is used for constructing a second ascending data set and a second descending data set according to the second standard data, inputting the second ascending data set and the second descending data set into the target ascending model and the target descending model respectively, and subtracting the output result of the target ascending model from the output result of the target descending model to obtain the weight of the object.
With reference to the second aspect, in a first implementation manner of the second aspect of the present invention, the acquisition module is specifically configured to: installing a plurality of sensors corresponding to a vehicle tail board based on preset sensor points, wherein the plurality of sensors comprise a hydraulic sensor for measuring lifting pressure, a hydraulic sensor for measuring overturning pressure, an inertial sensor for measuring lifting angle and acceleration and an inertial sensor for measuring overturning angle; and acquiring sensor data of the vehicle tail board in the multiple tail board lifting processes based on the plurality of sensors to obtain first sensor data.
With reference to the second aspect, in a second implementation manner of the second aspect of the present invention, the processing module further includes: the compensation unit is used for carrying out data compensation on the first sensor data to obtain compensated data; and the normalization unit is used for performing data normalization processing on the compensated data to obtain first standard data.
With reference to the second aspect, in a third implementation manner of the second aspect of the present invention, the compensation unit is specifically configured to: constructing a compensation dataset based on the first sensor data; fitting the objective function according to the compensation data set to obtain an error function; and correcting the reading of the hydraulic sensor according to the error function to obtain compensated data.
With reference to the second aspect, in a fourth implementation manner of the second aspect of the present invention, the normalization unit is specifically configured to: respectively extracting lifting hydraulic pressure sensor data, overturning hydraulic pressure sensor data, lifting angle sensor data and overturning angle sensor data corresponding to the compensated data; respectively calculating the lifting hydraulic pressure sensor data, the overturning hydraulic pressure sensor data, the lifting angle sensor data and the average value corresponding to the overturning angle sensor data to obtain a lifting hydraulic pressure sensor average value, an overturning hydraulic pressure sensor average value, a lifting angle sensor average value and an overturning angle sensor average value; and carrying out centering and standardization operation on the lifting hydraulic sensor average value, the overturning hydraulic sensor average value, the lifting angle sensor average value and the overturning angle sensor average value to obtain first standard data.
With reference to the second aspect, in a fifth implementation manner of the second aspect of the present invention, the building block is specifically configured to: extracting time sequence data information corresponding to the first standard data; acquiring time sequence data of a rising process and time sequence data of a falling process of the first standard data based on the time sequence data information; and adding Gaussian noise into the time sequence data of the ascending process and the time sequence data of the descending process respectively to obtain a first ascending data set and a first descending data set corresponding to the vehicle tail board.
With reference to the second aspect, in a sixth implementation manner of the second aspect of the present invention, the training module is specifically configured to: respectively constructing an ascending model and a descending model according to the first ascending data set and the first descending data set; respectively setting loss functions of the rising model and the falling model; and updating network parameters through a gradient descent method based on the loss function, and performing model training on the ascending model and the descending model to obtain a target ascending model and a target descending model.
A third aspect of the present invention provides a vehicle tailgate weighing apparatus comprising: a memory and at least one processor, the memory having instructions stored therein; the at least one processor invokes the instructions in the memory to cause the vehicle tailgate weighing device to perform the vehicle tailgate weighing method described above.
A fourth aspect of the invention provides a computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the vehicle tailgate weighing method described above.
According to the technical scheme provided by the invention, training data are acquired for the tail board of the vehicle based on the preset sensor point positions to obtain the first sensor data, the method can be suitable for a plurality of different application scenes, the errors of the data are more comprehensively analyzed, processed and calculated, the dynamic loading and unloading process of the tail board is analyzed, so that the characteristics of input data are enriched, then an ascending model and a descending model are respectively constructed according to the first ascending data set and the first descending data set, and model training is carried out on the ascending model and the descending model to obtain a target ascending model and a target descending model; by training by using more different vehicle data, the model has generalization, can be deployed to different vehicles of the same vehicle type without recalibration and training, improves the efficiency of manual installation, and obtains low-noise and high-authenticity data based on an error compensation algorithm for data preprocessing, thereby reducing the influence of symmetrical weight precision; the invention considers the dynamic process of lifting the tail plate, fully utilizes the time sequence characteristics of the data, and better accords with the characteristics of the data, thereby improving the weighing precision of the tail plate of the vehicle.
Drawings
FIG. 1 is a schematic view of an embodiment of a vehicle tailgate weighing method according to an embodiment of the invention;
FIG. 2 is a flow chart of the data normalization process in an embodiment of the present invention;
FIG. 3 is a flow chart of data set construction in an embodiment of the invention;
FIG. 4 is a flow chart of object model construction in an embodiment of the invention;
FIG. 5 is a schematic view of an embodiment of a vehicle tailgate weighing apparatus according to an embodiment of the invention;
FIG. 6 is a schematic view of another embodiment of a vehicle tailgate weighing apparatus according to an embodiment of the invention;
FIG. 7 is a schematic view of an embodiment of a vehicle tailgate weighing apparatus according to an embodiment of the invention;
FIG. 8 is a schematic view of a tailgate and sensor mounting locations in accordance with an embodiment of the invention;
FIG. 9 is a diagram showing a mechanical analysis of the tail plate structure in an embodiment of the invention;
FIG. 10 is a schematic diagram of hydraulic sensor data in an embodiment of the invention;
FIG. 11 is a graph showing the rising value of the hydraulic sensor according to the embodiment of the present invention;
FIG. 12 is a graph showing the decrease value of the hydraulic sensor according to the embodiment of the present invention;
FIG. 13 is a schematic diagram of a timing model architecture according to an embodiment of the present invention;
FIG. 14 is a graph showing relative error in weighing a vehicle in an embodiment of the invention;
fig. 15 is an absolute error of vehicle weighing in an embodiment of the invention.
Detailed Description
The embodiment of the invention provides a vehicle tail plate weighing method, device and equipment and a storage medium, which are used for improving the weighing precision of the vehicle tail plate. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present invention is described below with reference to fig. 1, and an embodiment of a vehicle tail board weighing method according to an embodiment of the present invention includes:
S101, training data acquisition is carried out on a vehicle tail board based on preset sensor point positions, and first sensor data are obtained;
it will be appreciated that the execution body of the present invention may be a vehicle tail board weighing device, and may also be a terminal or a server, and is not limited herein. The embodiment of the invention is described by taking a server as an execution main body as an example.
Specifically, the sensor is first mounted to the appropriate location of the tailgate. The vehicle tail plate basically needs to be provided with four sensors for measurement, namely a hydraulic sensor for measuring lifting pressure, a hydraulic sensor for measuring overturning pressure, an inertial sensor for measuring lifting angle and acceleration and an inertial sensor for measuring overturning angle. And (5) collecting sensor data. Based on the installed position, on the same vehicle model and different vehicle models, weights with different weights are lifted to acquire data of the sensors so as to obtain original sensor data for constructing a training set. Because of the differences in the mechanical structures of the sensor mounting portions of different vehicles and the small changes in the mechanical structures of the vehicles themselves, there are differences in the sensor data collected by different vehicles, and even in the data between the multiple ascent and descent processes of the same vehicle. It is therefore necessary to collect sensor data from multiple tailgate lifting processes for different vehicles. In the lifting process, weights with standard weights are placed on the tail plate, the set weights are recorded to serve as tag values, raw data measured by the sensor are processed to serve as input data, and first sensor data are obtained.
S102, performing data compensation and data standardization processing on first sensor data to obtain first standard data;
specifically, data compensation and data standardization processing are performed on the first sensor data, so as to obtain first standard data. Because the tail board can have overweight, weightlessness and tail board inclination in the ascending and descending processes, the data measured by the sensor has a certain error, and the obtained data needs to be compensated. The compensated data is normalized. In the process of training the time sequence model, in order to ensure the reliability of the result, the original index data needs to be subjected to centering and standardization processing. Furthermore, since the distribution of the data measured by the sensors of different vehicles is not uniform due to the difference in mechanical structure, it is necessary to perform centering and normalization processing on the data.
S103, constructing a first ascending data set and a first descending data set corresponding to the tail board of the vehicle according to the first standard data;
the rising data set and the falling data set are constructed based on the normalized data. Since in the rising process of an actual scene a person may need to rise together with the object under consideration and only in the falling process, it is necessary to construct a rising dataset and a falling dataset, i.e. a rising dataset and a falling dataset. The normalized data requires further extraction of the data sets of the rising and falling portions.
S104, respectively constructing a rising model and a falling model according to the rising data set and the falling data set, and carrying out model training on the rising model and the falling model to obtain a target rising model and a target falling model;
specifically, based on the constructed ascending data set and descending data set, a corresponding initial model is constructed. The time sequence characteristics are provided because the data sets are up and down and updated with the passage of time. As shown in fig. 11 and 12, when the data of fig. 10 are amplified, if there is a nonlinear characteristic and a time series characteristic, a model for processing the corresponding characteristics, that is, a model for measuring an ascent and a model for measuring a descent, needs to be constructed.
S105, acquiring second sensor data to be processed, and preprocessing the second sensor data to obtain second standard data;
s106, a second ascending data set and a second descending data set are built according to the second standard data, the second ascending data set and the second descending data set are respectively input into a target ascending model and a target descending model, and the output result of the target ascending model is subtracted from the output result of the target descending model to obtain the weight of the object.
Specifically, the trained model is deployed. After the sensors are installed on the tail plate positions of different vehicles, the data of the sensors are collected through lifting objects with different weights, then the collected sensor data are subjected to numerical compensation and standardization, second sensor data to be processed are collected, the second sensor data are preprocessed to obtain second standard data, then an ascending data set and a descending data set are constructed, and then the second standard data are input into a target model to be subjected to weight calculation, so that the weight of the object is calculated.
According to the embodiment of the invention, training data acquisition is carried out on the tail board of the vehicle based on the preset sensor point positions to obtain first sensor data, the method can be suitable for a plurality of different application scenes, the errors of the data are more comprehensively analyzed, processed and calculated, the dynamic loading and unloading process of the tail board is analyzed, so that the characteristics of input data are enriched, then an ascending model and a descending model are respectively constructed according to an ascending data set and a descending data set, and model training is carried out on the ascending model and the descending model to obtain a target ascending model and a target descending model; by training by using more different vehicle data, the model has generalization, can be deployed to different vehicles of the same vehicle type without recalibration and training, improves the efficiency of manual installation, and obtains low-noise and high-authenticity data based on an error compensation algorithm for data preprocessing, thereby reducing the influence of symmetrical weight precision; the invention considers the dynamic process of lifting the tail plate, fully utilizes the time sequence characteristics of the data, and better accords with the characteristics of the data, thereby improving the weighing precision of the tail plate of the vehicle.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
1. installing a plurality of sensors corresponding to a vehicle tail board based on preset sensor points, wherein the plurality of sensors comprise a hydraulic sensor for measuring lifting pressure, a hydraulic sensor for measuring overturning pressure, an inertial sensor for measuring lifting angle and acceleration and an inertial sensor for measuring overturning angle;
2. and acquiring sensor data of the vehicle tail board in the multiple tail board lifting processes based on the multiple sensors, and obtaining first sensor data.
Specifically, as shown in the vehicle tail board of fig. 8, 1 is a hydraulic sensor for measuring lifting pressure, 2 is a hydraulic sensor for measuring overturning pressure, 3 is an inertial sensor for measuring lifting angle, and 4 is an inertial sensor for measuring overturning.θIs the included angle between the tail plate and the ground. Sensor data of multiple tail board lifting processes of different vehicles are collected. In the lifting process, the weight with standard weight is placed on the tail plate, the set weight is recorded as a label value, and raw data measured by the sensor are processed and then used as input data.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
(1) Performing data compensation on the first sensor data to obtain compensated data;
(2) And carrying out data standardization processing on the compensated data to obtain first standard data.
Specifically, data compensation is performed on the data of the sensor. Because the tail board can have overweight, weightlessness and tail board inclination in the ascending and descending processes, the data measured by the sensor has a certain error, and the obtained data needs to be compensated. The compensated data is normalized. In the process of training the time sequence model, in order to ensure the reliability of the result, the original index data needs to be subjected to centering and standardization processing.
In a specific embodiment, the process of performing step (1) may specifically include the following steps:
1. constructing a compensation dataset based on the first sensor data;
2. fitting the objective function according to the compensation data set to obtain an error function;
3. and correcting the reading of the hydraulic sensor according to the error function to obtain compensated data.
Specifically, the tail plate structure is required to be analyzed correspondingly, the tail plate of FIG. 8 is abstracted and further analyzed, as shown in the analysis chart of FIG. 9, F 1 In order to lift the cylinder against the force exerted by the object,afor the upward lifting acceleration value,gthe acceleration of the gravity is that,θthe angle value between the tail plate and the horizontal direction after the tail plate is inclined downwards,αthe angle value of the telescopic rod and the tail plate where the angle sensor is arranged in the tail plate lifting process,hthe height value for lifting the tail plate. The acceleration of the strain sensor along the direction opposite to the earth center is calculated by resolving the gyroscope and accelerometer data of the inertial sensoraIncluded angle between tail board and groundθIncluded angle between telescopic rod and tail plateα. When the tail plate is in a static state, the object is put on the tail plate, and the included angle is formedθWhen=0, the lifting cylinder receives the force exerted by the objectF 1 The method comprises the following steps:
F 1 =f 1 (m,α) =L 1 (F ax coxα-F ay sinα)/L 2 sinβ
wherein the method comprises the steps off 1 (m,α) Is thatF 2 Mass with objectmAngle ofαThe functional relationship that exists.F ax AndF ay is thatF a Decomposition in the horizontal direction and in the vertical direction,F ax =PL 3 /h 1 =mgL 3 /h 1 ，F ay =P(1+L 3 ctgα/h 1 ) =mg(1+L 3 ctgα/h 1 )，α=kθ+b. Constant, L as shown in FIG. 9 1 、L 2 、L 3 At the mostThe respective lengths of the upper structure are such that,βis the included angle between the two rods of the uppermost structure,h 1 for the distance from E point to F point, L 1 、L 2 、L 3 、β、h 1 Are all constant.
At this time, the hydraulic sensor readss 1 And the force applied by the sameF 1 Has the following functional relationship:
s 1= g 1 (F 1 )=g 1 (f 1 (m))
wherein the method comprises the steps ofg 1 Representing hydraulic sensor readingss 1 And the force applied by the sameF 1 A functional relationship between them.
When the tail plate ascends or descends, the included angle θWhen the object is overweight or weightless, the lifting cylinder receives the force exerted by the objectF 2 The process is as follows:
F 2 =f 2 (m,α,α) =L 1 (F ax coxα-F ay sinα)/L 2 sinβ
wherein the method comprises the steps off 1 (m,α,a) Is thatF 2 Mass with objectmAngle ofαAcceleration ofaThe functional relationship that exists is that,αthe angle value of the tail plate when lifting is measured by the angle sensor,athe acceleration value is the acceleration value of the object in the lifting process, and the acceleration value is changed in the lifting process;F ax andF ay is thatF a Decomposition in the horizontal direction and in the vertical direction,F ax =PL 3 /h 1 =m(a-g)L 3 /h 1 ，F ay =P(1+L 3 ctgα/h 1 ) =m(a-g)(1+L 3 ctgα/h 1 )，α=kθ+b. As can be seen in FIGS. 10 and 11It is known that, due to the elastic deformation of the mechanical structure and the large acceleration generated when the mechanical structure starts to rise, the data of the hydraulic sensor at the initial rise show a vibration change and then tend to be stable.
At this point the sensor readingss 2 And the force applied by the sameF 2 Has the following functional relationship:
s 2= g 2 (F 2 )=g 2 (f 2 (m,α,a))
wherein the method comprises the steps ofg 2 Representing hydraulic sensor readingss 2 And the force applied by the sameF 2 A functional relationship between them.
When the tail plate is static and the tail plate forms an included angle with the groundθNot equal to 0, the hydraulic sensor readss 3 The method comprises the following steps:
s 3 =s 1 +ε 1
ε 1 =σ(s,θ,α,a)=σ(s 3 ,θ 0 ,α 0 ,0)
wherein the method comprises the steps ofε 1 Is the included angle between the tail board and the ground when the tail board is in a static stateθWhen not equal to 0, the error caused by the measured value of the weighing sensor,θ 0 can be measured by a turnover angle sensor when the tail plate is in a static state,α 0 the lifting angle sensor can measure the lifting angle of the tail plate in a static state.
When the tail plate ascends or descends and forms an included angle with the groundθNot equal to 0, reading by hydraulic sensor when the object is overweight or weightlesss 4 The method comprises the following steps:
s 4 =s 1 +ε 2
ε 2 =σ(s,θ,α,a)=σ(s 4 ,θ,α,a)=s 4 -s 1
wherein the method comprises the steps ofε 2 Is the included angle between the tail board and the ground when the tail board ascends or descendsθWhen not equal to 0, the error caused by the measured value of the weighing sensor,θcan be measured by a turnover angle sensor when the tail plate ascends or descends,αcan be measured by a lifting angle sensor when the tail plate is lifted or lowered.
The compensation operation is as follows:
1. firstly, determining the readings of hydraulic sensors of all standard weights of the tail board under the condition of static inclination-free weightlessness/overweight, and selecting accelerationa=0 and included angleθWhen the pressure is approximately equal to 0, the average value of the readings of the hydraulic sensor in a period of time is taken as the standard reading of the hydraulic sensor of the standard weight under the static conditions 1 。
2. Collected using hydraulic sensors measuring lifting pressures 4 And corresponding to the flip angle sensorθAnd the lifting angle measured by the inertial sensorαAcceleration measured by inertial sensoraConstructing a dataset
S={[s 4 ,θ,α,a];ε 2 }
ε 2 =s 4 -s 1
Wherein [ among others ]s 4 ,θ,α,a]Is an input value;ε 2 is the output value, i.e. the error.
3. Constructing a neural network, and using the data set pair functionσFitting was performed.
4. Obtaining an error functionσIt can then be used to correct the hydraulic sensor readings:
Data after hydraulic sensor correctionŝThe calculation is as follows:
ŝ=s 4 -ε 2 =s 4 -σ(s 4 ,θ,α,a)
in a specific embodiment, as shown in fig. 2, the process of performing step (2) may specifically include the following steps:
s201, respectively extracting lifting hydraulic pressure sensor data, overturning hydraulic pressure sensor data, lifting angle sensor data and overturning angle sensor data corresponding to the compensated data;
s202, respectively calculating the average value corresponding to the lifting hydraulic pressure sensor data, the overturning hydraulic pressure sensor data, the lifting angle sensor data and the overturning angle sensor data to obtain the average value of the lifting hydraulic pressure sensor, the average value of the overturning hydraulic pressure sensor, the average value of the lifting angle sensor and the average value of the overturning angle sensor;
and S203, carrying out centering and standardization operation on the lifting hydraulic pressure sensor mean value, the overturning hydraulic pressure sensor mean value, the lifting angle sensor mean value and the overturning angle sensor mean value to obtain first standard data.
Specifically, the server normalizes the compensated data. In the process of training the time sequence model, in order to ensure the reliability of the result, the original index data needs to be subjected to centering and standardization processing. Furthermore, since the distribution of the data measured by the sensors of different vehicles is not uniform due to the difference in mechanical structure, it is necessary to perform centering and normalization processing on the data. Recording lifting hydraulic pressure sensor data, overturning hydraulic pressure sensor data, lifting angle sensor data and overturning angle sensor data as P 1 =ŝ，P 2 ，A 1 ，A 2 The specific processing mode is to calculate the average value
，/>
，/>
，
Standard deviationσ 1 =std(P 1 )，σ 2 =std(P 2 )，σ 3 =std(A 1 )，σ 4 =std(A 2 ) Finally, the data is centered and standardized as follows:
wherein,,P 1 ，P 2 ，A 1 ，A 2 for the sensor data before the centralization,P 1 * ，P 2 * ，A 1 * ，A 2 * is the sensor data after centralization. This processing effectively reduces the complexity of the data, making the data distribution more consistent.
In a specific embodiment, as shown in fig. 3, the process of executing step S103 may specifically include the following steps:
s301, extracting time sequence data information corresponding to first standard data;
s302, acquiring time sequence data of a rising process and time sequence data of a falling process of first standard data based on time sequence data information;
s303, adding Gaussian noise into time sequence data of a rising process and time sequence data of a falling process respectively to obtain a first rising data set and a first falling data set corresponding to a tail plate of the vehicle.
Specifically, a dataset is constructed based on the normalized data. Since during the ascent of an actual scene a person may need to ascend with the object being weighed and only during the descent, it is necessary to construct an ascending data set and a descending data set. The normalized data requires further extraction of the data sets of the rising and falling portions. During the rising or falling of the weight, the response frequency of the hydraulic sensor f 1 10kHz or less, and the response frequency of the angle sensorf 2 10Hz or more and 10kHz or less, so that the time sequence data information of one lifting measured by the hydraulic sensor and the time sequence of one lifting measured by the angle sensor can be extracted according to the frequency of 10HzData information
For each sensor, a set of T time data is obtainedP 1 =[p 11 ,p 22 ,...,p T1 ]
，P 2 =[p 21 ,p 22 ,...,p T2 ]，A 1 =[a 11 ,a 22 ,...,a T1 ]，A 2 =[a 21 ,a 22 ,...,a T2 ]T is generally not fixed due to weight and differences in hydraulic system and mechanical structure of different vehicles.
Fig. 12 shows a process of ascending and descending a certain vehicle type tail board, wherein the curve in the red frame is ascending and descending once, and the ascending part and the descending part are data to be extracted. The values of the hydraulic sensor extracting the ascending and descending processes are shown in fig. 13. Extracting data of a certain rising process and a certain falling process as
Wherein,,X up for a certain rise process sensor timing data,X down sensor time sequence data for a certain descent process;
，/>
，/>
，/>
respectively acquiring lifting hydraulic pressure sensor data, overturning hydraulic pressure sensor data, lifting angle sensor data and overturning angle sensor data in the lifting process at the first moment; />
，/>
，/>
，/>
The lifting hydraulic sensor data, the overturning hydraulic sensor data, the lifting angle sensor data and the overturning angle sensor data in the descending process acquired at the first moment are respectively.
In each ascending and descending process, the weight and the weight of the person in the ascending process and the weight of the person in the descending process are recorded as the label values of the training set
In order to make the model more robust, gaussian noise is added to the time series data, and the data set of the rising process is constructed as
The data set of the descent process is +.>
。
In a specific embodiment, as shown in fig. 4, the process of executing step S104 may specifically include the following steps:
s401, respectively constructing an ascending model and a descending model according to a first ascending data set and a first descending data set;
s402, respectively setting a loss function of a rising model and a loss function of a falling model;
s403, updating network parameters based on the loss function through a gradient descent method, and performing model training on the ascending model and the descending model to obtain a target ascending model and a target descending model.
Specifically, based on the constructed dataset, a corresponding model is constructed. The time sequence characteristics are provided because the data sets are up and down and updated with the passage of time. As shown in fig. 13, when the data is amplified, if the data has nonlinear characteristics and time series characteristics, it is necessary to construct a model for processing the corresponding characteristics, that is, a model for measuring rising and a model for measuring falling. The timing model constructed may be RNN, LSTM, GRU, etc., in this example, LSTM timing model, as shown in fig. 13, where at the first time of a certain rise and fall, x 1 Is thatx 1 up Or (b)x 1 down While
，
. Further, based on time sequence data with different lengths, the time sequence data can be input into the LSTM time sequence model, so that corresponding predicted values are output in different units. The time sequence model fully utilizes the data and the related characteristics thereof, fits the time sequence characteristics and the nonlinear characteristics thereof, is suitable for different vehicles, has strong robustness and generalization, does not need repeated calibration of field installers, and improves the efficiency.
And training a model. Setting the loss function as
The following is an illustration of one specific embodiment of the present invention:
the embodiment is applied to the tail board of the sanitation truck.
For sanitation vehicles, the transportation mode is lifting and transporting the garbage can, and the garbage can is pushed into the sanitation vehicle in a manual mode: the garbage can is placed on the tail plate lifting and overturning mechanism at the tail part of the vehicle, and the manual control mechanism is used for lifting and overturning the tail plate and pushing the garbage can into the carriage. And weighing the garbage can in the ascending and descending process of the tail plate.
The first step is to collect time series data of sensors and the weight of the weighed object under the same vehicle model and different vehicle models. The sensors used in this embodiment are a hydraulic sensor and an angle sensor. Under different vehicle types or the same vehicle type, the installed sensor data have different values, and the same sensor data can not be measured because of the same vehicle type. In the lifting process, the sensor data and the set weight Y of the weighed object are required to be recorded, and the data recorded by the terminal are uploaded and stored.
The second step is to pre-process the data set to integrate into appropriate time series training data. Based on historical data in recent years, the data are compensated, centered and standardized. Extracting each time sequence data information measured by the current sensor in the process of lifting the object to be weighedx i Andy i the weight of the currently weighed object is recorded. Then n=6000 time sequence training data in each rising and falling process are obtained
、
. And finally, storing the time sequence training data and the label data row by row.
And thirdly, inputting training data into a time sequence model, and training the time sequence model to carry out weighing prediction. The time series data is added with Gaussian noise and then is input into the time series model for training, and then a value y is output, wherein the value y is used for predicting the weight of the object to be weighed, as shown in fig. 13. Setting a loss function based on the set loss function
And updating network parameters by using a gradient descent method, so as to obtain a trained time sequence model. The time sequence model adopts an LSTM model. The trained time sequence model is deployed on the terminal equipment, the constructed data set is input into the time sequence model, and accordingly the model calculates the weight of the object.
In this example, the test was performed using standard weight simulation sanitation tanks of 59kg, 109kg, 209kg, 309kg, 409kg, 509kg, and 6000 pieces of available data were used. The relative error of the final model test was less than 2.2% and the absolute error was less than 4kg, as shown in fig. 14 and 15. The method can achieve high data utilization rate, improves the efficiency of manual installation, has robustness and generalization of the model, and can be deployed to different vehicles of the same vehicle type without recalibration and training.
The method for weighing the vehicle tail board in the embodiment of the present invention is described above, and the vehicle tail board weighing device in the embodiment of the present invention is described below, referring to fig. 5, where an embodiment of the vehicle tail board weighing device in the embodiment of the present invention includes:
the acquisition module 501 is configured to acquire training data of a vehicle tail board based on a preset sensor point location, so as to obtain first sensor data;
the processing module 502 is configured to perform data compensation and data normalization processing on the first sensor data to obtain first standard data;
a construction module 503, configured to construct a first ascending data set and a first descending data set corresponding to the tail board of the vehicle according to the first standard data;
The training module 504 is configured to construct a lifting model and a descending model according to the first lifting data set and the first descending data set, and perform model training on the lifting model and the descending model to obtain a target lifting model and a target descending model;
the preprocessing module 505 is configured to collect second sensor data to be processed, and preprocess the second sensor data to obtain second standard data;
and a calculating module 506, configured to construct a second ascending data set and a second descending data set according to the second standard data, input the second ascending data set and the second descending data set into the target ascending model and the target descending model respectively, and subtract the output result of the target ascending model and the output result of the target descending model to obtain the object weight.
Through the cooperative cooperation of the components, the invention acquires training data of the vehicle tail board based on the preset sensor point positions to obtain first sensor data, can be suitable for a plurality of different application scenes, more comprehensively analyzes, processes and calculates errors of the data, analyzes the dynamic loading and unloading process of the tail board, enriches the characteristics of input data, respectively builds an ascending model and a descending model according to the first ascending data set and the first descending data set, and carries out model training on the ascending model and the descending model to obtain a target ascending model and a target descending model; by training by using more different vehicle data, the model has generalization, can be deployed to different vehicles of the same vehicle type without recalibration and training, improves the efficiency of manual installation, and obtains low-noise and high-authenticity data based on an error compensation algorithm for data preprocessing, thereby reducing the influence of symmetrical weight precision; the invention considers the dynamic process of lifting the tail plate, fully utilizes the time sequence characteristics of the data, and better accords with the characteristics of the data, thereby improving the weighing precision of the tail plate of the vehicle.
Referring to fig. 6, another embodiment of a vehicle tail plate weighing apparatus according to an embodiment of the present invention includes:
the acquisition module 501 is configured to acquire training data of a vehicle tail board based on a preset sensor point location, so as to obtain first sensor data;
the processing module 502 is configured to perform data compensation and data normalization processing on the first sensor data to obtain first standard data;
a construction module 503, configured to construct a first ascending data set and a first descending data set corresponding to the tail board of the vehicle according to the first standard data;
the training module 504 is configured to construct a lifting model and a descending model according to the first lifting data set and the first descending data set, and perform model training on the lifting model and the descending model to obtain a target lifting model and a target descending model;
the preprocessing module 505 is configured to collect second sensor data to be processed, and preprocess the second sensor data to obtain second standard data;
and a calculating module 506, configured to construct a second ascending data set and a second descending data set according to the second standard data, input the second ascending data set and the second descending data set into the target ascending model and the target descending model respectively, and subtract the output result of the target ascending model and the output result of the target descending model to obtain the object weight.
Optionally, the collecting module 501 is specifically configured to: installing a plurality of sensors corresponding to a vehicle tail board based on preset sensor points, wherein the plurality of sensors comprise a hydraulic sensor for measuring lifting pressure, a hydraulic sensor for measuring overturning pressure, an inertial sensor for measuring lifting angle and acceleration and an inertial sensor for measuring overturning angle; and acquiring sensor data of the vehicle tail board in the multiple tail board lifting processes based on the plurality of sensors to obtain first sensor data.
Optionally, the processing module 502 further includes:
the compensation unit 5021 is configured to perform data compensation on the first sensor data to obtain compensated data;
and the normalizing unit 5022 is configured to perform data normalization processing on the compensated data to obtain first standard data.
Optionally, the compensation unit 5021 is specifically configured to: constructing a compensation dataset based on the first sensor data; fitting the objective function according to the compensation data set to obtain an error function; and correcting the reading of the hydraulic sensor according to the error function to obtain compensated data.
Optionally, the normalizing unit 5022 is specifically configured to: respectively extracting lifting hydraulic pressure sensor data, overturning hydraulic pressure sensor data, lifting angle sensor data and overturning angle sensor data corresponding to the compensated data; respectively calculating the lifting hydraulic pressure sensor data, the overturning hydraulic pressure sensor data, the lifting angle sensor data and the average value corresponding to the overturning angle sensor data to obtain a lifting hydraulic pressure sensor average value, an overturning hydraulic pressure sensor average value, a lifting angle sensor average value and an overturning angle sensor average value; and carrying out centering and standardization operation on the lifting hydraulic sensor average value, the overturning hydraulic sensor average value, the lifting angle sensor average value and the overturning angle sensor average value to obtain first standard data.
Optionally, the building module 503 is specifically configured to: extracting time sequence data information corresponding to the first standard data; acquiring time sequence data of a rising process and time sequence data of a falling process of the first standard data based on the time sequence data information; and adding Gaussian noise into the time sequence data of the ascending process and the time sequence data of the descending process respectively to obtain a first ascending data set and a first descending data set corresponding to the vehicle tail board.
Optionally, the training module 504 is specifically configured to: respectively constructing an ascending model and a descending model according to the first ascending data set and the first descending data set; respectively setting loss functions of the rising model and the falling model; and updating network parameters through a gradient descent method based on the loss function, and performing model training on the ascending model and the descending model to obtain a target ascending model and a target descending model.
According to the embodiment of the invention, training data acquisition is carried out on the tail board of the vehicle based on the preset sensor point positions to obtain first sensor data, the method can be suitable for a plurality of different application scenes, the errors of the data are more comprehensively analyzed, processed and calculated, the dynamic loading and unloading processes of the tail board are analyzed, so that the characteristics of input data are enriched, then an ascending model and a descending model are respectively constructed according to the first ascending data set and the first descending data set, and model training is carried out on the ascending model and the descending model to obtain a target ascending model and a target descending model; by training by using more different vehicle data, the model has generalization, can be deployed to different vehicles of the same vehicle type without recalibration and training, improves the efficiency of manual installation, and obtains low-noise and high-authenticity data based on an error compensation algorithm for data preprocessing, thereby reducing the influence of symmetrical weight precision; the invention considers the dynamic process of lifting the tail plate, fully utilizes the time sequence characteristics of the data, and better accords with the characteristics of the data, thereby improving the weighing precision of the tail plate of the vehicle.
The vehicle tailgate weighing apparatus in the embodiment of the present invention is described in detail above in terms of the modularized functional entity in fig. 5 and 6, and the vehicle tailgate weighing device in the embodiment of the present invention is described in detail below in terms of hardware processing.
Fig. 7 is a schematic structural diagram of a vehicle tail weighing apparatus 600 according to an embodiment of the present invention, where the vehicle tail weighing apparatus 600 may have a relatively large difference due to different configurations or performances, and may include one or more processors (central processing units, CPU) 610 (e.g., one or more processors) and a memory 620, and one or more storage media 630 (e.g., one or more mass storage devices) storing application programs 633 or data 632. Wherein the memory 620 and the storage medium 630 may be transitory or persistent storage. The program stored in the storage medium 630 may include one or more modules (not shown), each of which may include a series of instruction operations for the vehicle tailgate weighing device 600. Still further, the processor 610 may be configured to communicate with the storage medium 630 to execute a series of instruction operations in the storage medium 630 on the vehicle tailgate weighing device 600.
The vehicle tailgate weighing device 600 may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input/output interfaces 660, and/or one or more operating systems 631, such as Windows Serve, mac OS X, unix, linux, freeBSD, and the like. It will be appreciated by those skilled in the art that the vehicle tailgate weighing device structure shown in fig. 7 is not limiting and may include more or fewer components than illustrated, or may combine certain components, or a different arrangement of components.
The invention also provides a vehicle tail plate weighing device which comprises a memory and a processor, wherein the memory stores computer readable instructions which, when executed by the processor, cause the processor to execute the steps of the vehicle tail plate weighing method in the above embodiments.
The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and which may also be a volatile computer readable storage medium, the computer readable storage medium having instructions stored therein that, when executed on a computer, cause the computer to perform the steps of the vehicle tailgate weighing method.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random acceS memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A vehicle tailgate weighing method, characterized by comprising:
training data acquisition is carried out on a vehicle tail board based on preset sensor point positions, so that first sensor data are obtained;
performing data compensation and data standardization processing on the first sensor data to obtain first standard data;
constructing a first ascending data set and a first descending data set corresponding to the vehicle tail board according to the first standard data;
respectively constructing a rising model and a falling model according to the first rising data set and the first falling data set, and carrying out model training on the rising model and the falling model to obtain a target rising model and a target falling model;
Collecting second sensor data to be processed, and preprocessing the second sensor data to obtain second standard data;
and constructing a second ascending data set and a second descending data set according to the second standard data, inputting the second ascending data set and the second descending data set into the target ascending model and the target descending model respectively, and subtracting the output result of the target ascending model from the output result of the target descending model to obtain the weight of the object.
2. The vehicle tail board weighing method according to claim 1, wherein the training data acquisition is performed on the vehicle tail board based on the preset sensor point positions to obtain first sensor data, and the method comprises the following steps:
installing a plurality of sensors corresponding to a vehicle tail board based on preset sensor points, wherein the plurality of sensors comprise a hydraulic sensor for measuring lifting pressure, a hydraulic sensor for measuring overturning pressure, an inertial sensor for measuring lifting angle and acceleration and an inertial sensor for measuring overturning angle;
and acquiring sensor data of the vehicle tail board in the multiple tail board lifting processes based on the plurality of sensors to obtain first sensor data.
3. The vehicle tailgate weighing method according to claim 1, wherein said performing data compensation and data normalization processing on said first sensor data to obtain first standard data comprises:
performing data compensation on the first sensor data to obtain compensated data;
and carrying out data standardization processing on the compensated data to obtain first standard data.
4. A vehicle tailgate weighing method according to claim 3 wherein said data compensating said first sensor data to obtain compensated data comprises:
constructing a compensation dataset based on the first sensor data;
fitting the objective function according to the compensation data set to obtain an error function;
and correcting the reading of the hydraulic sensor according to the error function to obtain compensated data.
5. A vehicle tailgate weighing method according to claim 3 wherein said data normalization processing of said compensated data to obtain first standard data comprises:
respectively extracting lifting hydraulic pressure sensor data, overturning hydraulic pressure sensor data, lifting angle sensor data and overturning angle sensor data corresponding to the compensated data;
Respectively calculating the lifting hydraulic pressure sensor data, the overturning hydraulic pressure sensor data, the lifting angle sensor data and the average value corresponding to the overturning angle sensor data to obtain a lifting hydraulic pressure sensor average value, an overturning hydraulic pressure sensor average value, a lifting angle sensor average value and an overturning angle sensor average value;
and carrying out centering and standardization operation on the lifting hydraulic sensor average value, the overturning hydraulic sensor average value, the lifting angle sensor average value and the overturning angle sensor average value to obtain first standard data.
6. The vehicle tailgate weighing method according to claim 1, wherein said constructing a first rising data set and a first falling data set corresponding to said vehicle tailgate from said first standard data comprises:
extracting time sequence data information corresponding to the first standard data;
acquiring time sequence data of a rising process and time sequence data of a falling process of the first standard data based on the time sequence data information;
and adding Gaussian noise into the time sequence data of the ascending process and the time sequence data of the descending process respectively to obtain a first ascending data set and a first descending data set corresponding to the vehicle tail board.
7. The vehicle tail board weighing method according to claim 1, wherein the steps of constructing a rising model and a falling model according to the first rising data set and the first falling data set, and performing model training on the rising model and the falling model to obtain a target rising model and a target falling model respectively include:
respectively constructing an ascending model and a descending model according to the first ascending data set and the first descending data set;
respectively setting loss functions of the rising model and the falling model;
and updating network parameters through a gradient descent method based on the loss function, and performing model training on the ascending model and the descending model to obtain a target ascending model and a target descending model.
8. A vehicle tailgate weighing apparatus, characterized in that the vehicle tailgate weighing apparatus comprises:
the acquisition module is used for acquiring training data of the vehicle tail board based on preset sensor point positions to obtain first sensor data;
the processing module is used for carrying out data compensation and data standardization processing on the first sensor data to obtain first standard data;
the building module is used for building a first ascending data set and a first descending data set corresponding to the vehicle tail board according to the first standard data;
The training module is used for respectively constructing a rising model and a falling model according to the first rising data set and the first falling data set, and carrying out model training on the rising model and the falling model to obtain a target rising model and a target falling model;
the preprocessing module is used for acquiring second sensor data to be processed and preprocessing the second sensor data to obtain second standard data;
and the calculation module is used for constructing a second ascending data set and a second descending data set according to the second standard data, inputting the second ascending data set and the second descending data set into the target ascending model and the target descending model respectively, and subtracting the output result of the target ascending model from the output result of the target descending model to obtain the weight of the object.
9. A vehicle tailgate weighing apparatus, characterized in that the vehicle tailgate weighing apparatus comprises: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invokes the instructions in the memory to cause the vehicle tailgate weighing device to perform the vehicle tailgate weighing method of any one of claims 1-7.
10. A computer readable storage medium having instructions stored thereon, which when executed by a processor, implement the vehicle tailgate weighing method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310141151.8A CN115931095B (en) | 2023-02-21 | 2023-02-21 | Vehicle tail plate weighing method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310141151.8A CN115931095B (en) | 2023-02-21 | 2023-02-21 | Vehicle tail plate weighing method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115931095A CN115931095A (en) | 2023-04-07 |
| CN115931095B true CN115931095B (en) | 2023-05-05 |
Family
ID=85833945
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310141151.8A Active CN115931095B (en) | 2023-02-21 | 2023-02-21 | Vehicle tail plate weighing method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115931095B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116152757B (en) * | 2023-04-18 | 2023-07-07 | 深圳亿维锐创科技股份有限公司 | Weighing data analysis method and related device based on multiple points |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB491668A (en) * | 1937-04-09 | 1938-09-07 | Simon Ltd Henry | Improvements relating to automatic weighing machines |
| EP0032119A1 (en) * | 1980-01-08 | 1981-07-15 | Walter Dipl.-Ing. Dr. Hillesheimer | Hydraulically operated lifting, loading or tilting platform |
| DE4092254T (en) * | 1989-12-18 | 1992-12-10 | ||
| CN106197633A (en) * | 2016-07-05 | 2016-12-07 | 厦门市三友软件有限公司 | Vehicle weighing method, vehicle weighing device and vehicle weighing system |
| CN107727210A (en) * | 2017-09-21 | 2018-02-23 | 深圳市汉德网络科技有限公司 | A kind of vehicle-mounted Weighing method of kitchen waste cart and system |
| CN113091866A (en) * | 2021-04-06 | 2021-07-09 | 深圳市汉德网络科技有限公司 | Method and device for measuring load mass of automobile in real time |
| CN113104455A (en) * | 2021-04-13 | 2021-07-13 | 深圳市汉德网络科技有限公司 | Garbage weighing method and device based on bucket-hanging type garbage truck, electronic equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2798525C (en) * | 2011-05-13 | 2013-07-16 | Deploy Technologies Inc. | Load-measuring, fleet asset tracking and data management system for load-lifting vehicles |
-
2023
- 2023-02-21 CN CN202310141151.8A patent/CN115931095B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB491668A (en) * | 1937-04-09 | 1938-09-07 | Simon Ltd Henry | Improvements relating to automatic weighing machines |
| EP0032119A1 (en) * | 1980-01-08 | 1981-07-15 | Walter Dipl.-Ing. Dr. Hillesheimer | Hydraulically operated lifting, loading or tilting platform |
| DE4092254T (en) * | 1989-12-18 | 1992-12-10 | ||
| CN106197633A (en) * | 2016-07-05 | 2016-12-07 | 厦门市三友软件有限公司 | Vehicle weighing method, vehicle weighing device and vehicle weighing system |
| CN107727210A (en) * | 2017-09-21 | 2018-02-23 | 深圳市汉德网络科技有限公司 | A kind of vehicle-mounted Weighing method of kitchen waste cart and system |
| CN113091866A (en) * | 2021-04-06 | 2021-07-09 | 深圳市汉德网络科技有限公司 | Method and device for measuring load mass of automobile in real time |
| CN113104455A (en) * | 2021-04-13 | 2021-07-13 | 深圳市汉德网络科技有限公司 | Garbage weighing method and device based on bucket-hanging type garbage truck, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115931095A (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115931095B (en) | Vehicle tail plate weighing method, device, equipment and storage medium | |
| CN110689723B (en) | Truck overload identification method based on power distribution and self-learning | |
| US9470574B2 (en) | Slow speed weigh-in-motion system | |
| CN107727210B (en) | A kind of vehicle-mounted weighing method of kitchen waste cart and system | |
| CN111157092B (en) | Vehicle-mounted weighing automatic calibration method and computer readable storage medium | |
| CN111089643B (en) | Dynamic vehicle-mounted weighing method and system | |
| CN113091866B (en) | Method and device for measuring loading quality of automobile in real time | |
| CN102261988B (en) | Method for testing residual life of crane | |
| CN113104455B (en) | Garbage weighing method and device based on bucket-hanging type garbage truck, electronic equipment and storage medium | |
| US10203241B2 (en) | Slow speed weigh-in-motion system with flexure | |
| KR101478760B1 (en) | Apparatus and method for measuring lift axle weight of vehicle | |
| US20180203949A1 (en) | Data transformation for a virtual asset | |
| CN113850244B (en) | Coal conveying quantity monitoring method, device and equipment based on image recognition and storage medium | |
| CN111783326B (en) | Wind speed and dynamic response test-based wind-induced safety assessment method for power transmission tower structure | |
| CN116087042B (en) | Real-time monitoring system and method for dust deposition volume spatial distribution | |
| CN109556687A (en) | Weight measurements, device, equipment and computer readable storage medium | |
| US11215499B1 (en) | Calibration apparatus and method | |
| CN111829639B (en) | Calibration method of combined truck scale | |
| CN112763045A (en) | Vehicle self-load detection cloud calibration prediction method | |
| CN110850043A (en) | Grain temperature and humidity prediction method based on GRA-BPNN | |
| CN113865682A (en) | Truck tire load determining method and device and storage medium | |
| CN116625473B (en) | Method and system for measuring weight of different-density cargoes loaded on mixer truck | |
| CN117755991A (en) | Method and device for detecting abnormal state of hoisting equipment and computer equipment | |
| CN117115796A (en) | Method, device, equipment and storage medium for determining vehicle load | |
| CN117686074A (en) | Vehicle-mounted weighing model training method, device, equipment 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 |