CN116069043A - Unmanned agricultural machinery operation speed autonomous decision-making method - Google Patents

Unmanned agricultural machinery operation speed autonomous decision-making method Download PDF

Info

Publication number
CN116069043A
CN116069043A CN202310293891.3A CN202310293891A CN116069043A CN 116069043 A CN116069043 A CN 116069043A CN 202310293891 A CN202310293891 A CN 202310293891A CN 116069043 A CN116069043 A CN 116069043A
Authority
CN
China
Prior art keywords
operation speed
speed
unmanned agricultural
agricultural machinery
gaussian process
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.)
Granted
Application number
CN202310293891.3A
Other languages
Chinese (zh)
Other versions
CN116069043B (en
Inventor
何杰
汪沛
胡炼
黄培奎
李明锦
黄钰峰
丁帅奇
曾思晓
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China Agricultural University
Original Assignee
South China Agricultural University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China Agricultural University filed Critical South China Agricultural University
Priority to CN202310293891.3A priority Critical patent/CN116069043B/en
Publication of CN116069043A publication Critical patent/CN116069043A/en
Application granted granted Critical
Publication of CN116069043B publication Critical patent/CN116069043B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
    • 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
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
    • 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

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses an autonomous decision-making method for the operation speed of an unmanned agricultural machine, which comprises the following steps: s1, acquiring movement state data of an agricultural machine, screening dependent variables influencing the operation speed of the agricultural machine, and establishing a target operation speed training sample; s2, establishing a Gaussian process regression prediction model of the agricultural machinery target operation speed; s3, training a Gaussian process regression prediction model of the target operation speed; s4, evaluating and checking the regression prediction result of the Gaussian process during training to obtain a training completion prediction model; s5, acquiring an agricultural machinery test data sample, acquiring a corresponding unmanned agricultural machinery prediction speed by adopting a Gaussian process regression prediction model, and taking the average value of the prediction speed as a target decision speed; and S6, automatically adjusting the operation speed of the unmanned agricultural machinery according to the obtained target decision speed. The method of the invention realizes that the unmanned agricultural machinery autonomously decides and adjusts the operation speed according to the fluctuation condition of the farmland topography, and improves the operation efficiency while guaranteeing the operation quality.

Description

Unmanned agricultural machinery operation speed autonomous decision-making method
Technical Field
The invention belongs to the technical field of intelligent agricultural machinery, and particularly relates to an autonomous decision-making method for the operation speed of an unmanned agricultural machinery.
Background
In a wet and slippery slurry and hard bottom layer uneven farmland environment, the unmanned agricultural machine is easy to cause the reduction of the linear running precision of the agricultural machine, so that the operation effect is reduced. At present, unmanned agricultural machinery usually sets up fixed speed operation by people, or adjusts operation speed by human intervention, and experienced agricultural machinery drivers generally meet uneven farmland areas, generally independently decide to slow down and pass through the hollow areas, so that operation quality is ensured, operation speed is increased in flat areas, and operation efficiency is improved.
The Chinese patent No. 202011598475.7 proposes a method and a system for regulating and controlling the advancing speed of an unmanned agricultural machine, which can automatically regulate and stabilize the operating speed of the agricultural machine by judging the operating load of the agricultural machine; the Chinese patent No. 201811075147.1 proposes an unmanned intelligent control system of agricultural machinery and a control method thereof, and designs a debugging device and a method capable of realizing electric control; however, the inventive solution only achieves the speed-controllable function of the unmanned agricultural machine, which still cannot autonomously decide and adjust the operation speed.
Disclosure of Invention
The invention mainly aims to overcome the defects and shortcomings of the prior art, and provides an autonomous decision-making method for the operation speed of an unmanned agricultural machine.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
an autonomous decision-making method for the operation speed of an unmanned agricultural machine comprises the following steps:
s1, acquiring movement state data of an unmanned agricultural machine, screening dependent variables affecting the operation speed of the unmanned agricultural machine, and establishing a target operation speed training sample;
s2, establishing a Gaussian process regression prediction model of the target operation speed of the unmanned agricultural machinery;
s3, training a Gaussian process regression prediction model of the target operation speed;
s4, evaluating and checking Gaussian process regression prediction results during training to obtain a trained Gaussian process regression prediction model;
s5, acquiring an unmanned agricultural machinery test data sample, acquiring a corresponding unmanned agricultural machinery prediction speed by adopting a Gaussian process regression prediction model, and taking the average value of the prediction speeds as a target decision speed;
and S6, automatically adjusting the operation speed of the unmanned agricultural machinery according to the obtained target decision speed.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. according to the invention, the unmanned agricultural machine operation speed is predicted by using the information such as elevation difference, roll angle, transverse position deviation, course deviation and the like in the movement process of the agricultural machine, which is acquired by the unmanned agricultural machine pose sensing device, so that the operation speed is independently decided and regulated, and the operation efficiency is improved while the operation quality is ensured; the method has strong adaptability and can effectively improve the intelligent degree and applicability of the unmanned agricultural machinery.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a mapping diagram of a Gaussian process regression prediction model of the present invention;
FIG. 3 is a Gaussian process regression flow of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but embodiments of the present invention are not limited thereto.
Examples
As shown in fig. 1, the invention relates to an autonomous decision-making method for the operation speed of an unmanned agricultural machine, which comprises the following steps:
s1, acquiring unmanned agricultural machinery motion state data through an airborne terminal of the unmanned agricultural machinery, screening dependent variables influencing the operation speed of the unmanned agricultural machinery, and establishing a target operation speed training sample; the method comprises the following steps:
the operation speed of the unmanned agricultural machinery is influenced by a plurality of dependent variables, and the Spearman rank correlation coefficient is respectively calculated for each dependent variable by adopting a correlation method:
Figure SMS_1
（1）
wherein ,
Figure SMS_2
for the rank correlation coefficient,nfor data length +.>
Figure SMS_3
and />
Figure SMS_4
Original data, respectively->
Figure SMS_5
and />
Figure SMS_6
Respectively, the data are arranged in order from small to large;
calculating each dependent variable and speed according to the formula (1)vRank correlation coefficient
Figure SMS_7
，/>
Figure SMS_8
The larger the absolute value is, the dependent variable affects the speed of the unmanned agricultural machineryvThe more pronounced.
In the embodiment, the target operation speed of the unmanned agricultural machinery is influencedvThe most significant 4 factors: difference of elevation
Figure SMS_9
Roll angle->
Figure SMS_10
Lateral position deviation->
Figure SMS_11
Heading angle->
Figure SMS_12
Composing 4-dimensional input vectorsm：
Figure SMS_13
（2）
Figure SMS_14
（3）
wherein ,
Figure SMS_15
is the firstiInput vector->
Figure SMS_16
Respectively represent the firstiDifference in elevation of the individual input vectors>
Figure SMS_17
Roll angle->
Figure SMS_18
Lateral position deviation->
Figure SMS_19
Heading angle->
Figure SMS_20
；
Collecting a plurality of input vectors and output scalar quantities, and establishing a target operation speed training sample D:
Figure SMS_21
（4）
wherein ,
Figure SMS_22
is the firstiOutput scalar of the individual samples; />
Figure SMS_23
Inputting a matrix formed by vectors for all training samples; />
Figure SMS_24
Outputting a scalar-composed matrix for all training samples, wherein the training sample set is noisy and obeys a standard gaussian distribution
Figure SMS_25
。
S2, establishing a Gaussian process regression prediction model of the target operation speed of the unmanned agricultural machinery; the method comprises the following steps:
defining a Gaussian process
Figure SMS_26
Is +.about.>
Figure SMS_27
and />
Figure SMS_28
：/>
Figure SMS_29
（5）
Figure SMS_30
（6）
The gaussian process is:
Figure SMS_31
（7）
wherein, the random variable parameterxOutput value of (2)
Figure SMS_32
Is a random variable;
for the followingnInput number
Figure SMS_33
Output->
Figure SMS_34
Obeying a joint gaussian distribution:
Figure SMS_35
（8）
wherein ,
Figure SMS_36
，/>
Figure SMS_37
is->
Figure SMS_38
A dimension matrix;
for decision regression of target operation speed, a noise-containing function model is adopted:
Figure SMS_39
（9）
wherein ,yin order to observe the value of the value,ffor the function value of the function,
Figure SMS_40
is thatxA corresponding true value; />
Figure SMS_41
Obeying normal distribution, wherein the average value is 0;
observations ofyIs:
Figure SMS_42
（10）
given training set
Figure SMS_43
Test set->
Figure SMS_44
Let->
Figure SMS_45
，
Figure SMS_46
Observations ofyAnd predictive value->
Figure SMS_47
Is:
Figure SMS_48
（11）
wherein ,
Figure SMS_49
、/>
Figure SMS_51
、/>
Figure SMS_53
and />
Figure SMS_54
Are all the covariance matrices of the two images,
Figure SMS_55
is->
Figure SMS_57
Symmetric positive definite covariance matrix, matrix element +.>
Figure SMS_59
Representation->
Figure SMS_50
and />
Figure SMS_52
Correlation between; />
Figure SMS_56
Is->
Figure SMS_58
A unit matrix; />
Figure SMS_60
and />
Figure SMS_61
The method comprises the following steps:
Figure SMS_62
（12）
Figure SMS_63
（13）
the key prediction equation for obtaining Gaussian process regression is obtained by adopting Bayesian theory:
Figure SMS_64
（14）/>
wherein ,
Figure SMS_65
data point +.>
Figure SMS_66
Corresponding mean value->
Figure SMS_67
For the corresponding variance:
Figure SMS_68
（15）
Figure SMS_69
（16）
wherein ,
Figure SMS_70
the predicted value of the target working speed is obtained.
As shown in fig. 2, a gaussian process regression prediction model map is provided.
S3, training a Gaussian process regression prediction model of the target operation speed; as shown in fig. 3, includes:
s31, determining a covariance function; the method comprises the following steps:
the Gaussian process regression prediction model adopts a square index covariance kernel function:
Figure SMS_71
（17）
wherein ,
Figure SMS_72
、lis a hyper-parameter of kernel function, +.>
Figure SMS_73
For the signal variance to be a function of the signal variance,ldescribed are scale parameters of the complexity of the kernel function in the parameter space.
S32, setting initial super parameters and optimizing marginal likelihood super parameters; the method comprises the following steps:
input vector of Gaussian process regression prediction model of unmanned agricultural machinery target operation speedmBy difference in elevation
Figure SMS_74
Roll angle->
Figure SMS_75
Lateral position deviation->
Figure SMS_76
And heading angle->
Figure SMS_77
Composition, therefore, for the spatial length scale of equation (17)lThe determination of (1) employs an automatic determination of the correlation, expressed as:
Figure SMS_78
（18）
wherein ,
Figure SMS_79
respectively represent the height difference +.>
Figure SMS_80
Roll angle->
Figure SMS_81
Lateral position deviation->
Figure SMS_82
And heading angle
Figure SMS_83
A length scale of each dimension of (a);
the training process of the Gaussian process regression prediction model is regarded as a problem for solving nonlinear numerical optimization, and super-parameter vectors are defined:
Figure SMS_84
（19）
by using
Figure SMS_85
Representation->
Figure SMS_86
The element in (a), namely the super parameter;
optimizing the super-parameters by adopting a Maximum Likelihood Estimation (MLE) method, wherein the marginal likelihood function is as follows:
Figure SMS_87
（20）
likelihood obeys a gaussian distribution:
Figure SMS_88
（21）
the method is obtained by a Gaussian process regression prediction model:
Figure SMS_89
（22）/>
equation (22) still obeys the gaussian distribution after the integration operation:
Figure SMS_90
（23）
wherein ,
Figure SMS_91
，/>
Figure SMS_92
a covariance matrix which does not consider Gaussian noise;
solving the negative logarithm of the formula (20) to obtain a negative log-boundary likelihood (Negative log marginal likelihood, NLML) function:
Figure SMS_93
（24）
solving the minimum value of the formula (24) to obtain optimized super-parameters, namely converting training of a Gaussian process regression prediction model into an optimization problem of the formula (24), wherein the objective function is thatLThe optimization objective is to solveLIs the minimum of (2);
solving the formula (24) for the super-parameters
Figure SMS_94
Is a partial derivative of:
Figure SMS_95
（25）
wherein ,
Figure SMS_96
，/>
Figure SMS_97
representing the trace of the matrix;
the conjugate gradient method is adopted to minimize the partial derivative, so as to obtain the optimal solution of the super parameter, namely, the altitude difference obtained according to the measurement of the unmanned agricultural machinery
Figure SMS_98
Roll angle->
Figure SMS_99
Lateral position deviation->
Figure SMS_100
And heading angle->
Figure SMS_101
And predicting the target operation speed to be adjusted through a Gaussian regression process.
S4, evaluating and checking Gaussian process regression prediction results during training to obtain a trained Gaussian process regression prediction model; in this embodiment, the mean absolute error is specifically adopted for the evaluation of the regression prediction result in the gaussian processMAEAnd judging the deviation between the predicted value and the true value of the sample:
Figure SMS_102
（26）
wherein ,
Figure SMS_103
and />
Figure SMS_104
Respectively represent the firstiA true value and a predicted value,Nto predict the number of samples;MAEthe smaller the value, the better the model predictive performance;
autoregressive training samples using a trained Gaussian process regression prediction model according to equation (15), and calculating according to equation (26)MAEJudging whether the training is successful or not, and if the training is unsuccessful, returning to the step S3; otherwise, step S5 is entered.
S5, acquiring an unmanned agricultural machinery test data sample, acquiring a corresponding unmanned agricultural machinery prediction speed by adopting a Gaussian process regression prediction model, and taking the average value of the prediction speeds as a target decision speed; the method comprises the following steps:
deploying a trained Gaussian process regression prediction model on an unmanned agricultural machine terminal controller, and acquiring data of the unmanned agricultural machine in a period of time as a test data sample
Figure SMS_105
Obtaining a corresponding unmanned agricultural machinery prediction speed by adopting a Gaussian process regression prediction model>
Figure SMS_106
With this mean value of the predicted speed +.>
Figure SMS_107
Decision speed as target->
Figure SMS_108
。
S6, automatically adjusting the operation speed of the unmanned agricultural machinery according to the obtained target decision speed; the method comprises the following steps:
will obtain a target decision speed
Figure SMS_109
An on-board speed controller transmitted to the unmanned agricultural machinery, the speed controller determining a speed according to the current speed and the target>
Figure SMS_110
And the difference value is used for automatically adjusting the operation speed of the unmanned agricultural machinery.
It should also be noted that in this specification, terms such as "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (10)

1. An autonomous decision-making method for the operation speed of an unmanned agricultural machine is characterized by comprising the following steps:
s1, acquiring movement state data of an unmanned agricultural machine, screening dependent variables affecting the operation speed of the unmanned agricultural machine, and establishing a target operation speed training sample;
s2, establishing a Gaussian process regression prediction model of the target operation speed of the unmanned agricultural machinery;
s3, training a Gaussian process regression prediction model of the target operation speed;
s4, evaluating and checking Gaussian process regression prediction results during training to obtain a trained Gaussian process regression prediction model;
s5, acquiring an unmanned agricultural machinery test data sample, acquiring a corresponding unmanned agricultural machinery prediction speed by adopting a Gaussian process regression prediction model, and taking the average value of the prediction speeds as a target decision speed;
and S6, automatically adjusting the operation speed of the unmanned agricultural machinery according to the obtained target decision speed.
2. The autonomous decision making method for the operation speed of the unmanned agricultural machine according to claim 1, wherein in the step S1, the screening of the dependent variables affecting the operation speed of the unmanned agricultural machine is specifically:
the operation speed of the unmanned agricultural machinery is influenced by a plurality of dependent variables, and the Spearman rank correlation coefficient is respectively calculated for each dependent variable by adopting a correlation method:
Figure QLYQS_1
（1）
wherein ,
Figure QLYQS_2
for the rank correlation coefficient,nfor data length +.>
Figure QLYQS_3
and />
Figure QLYQS_4
Original data, respectively->
Figure QLYQS_5
and />
Figure QLYQS_6
Respectively, the data are arranged in order from small to large;
calculating each dependent variable and speed according to the formula (1)vRank correlation coefficient
Figure QLYQS_7
，/>
Figure QLYQS_8
The larger the absolute value is, the dependent variable affects the speed of the unmanned agricultural machineryvThe more pronounced.
3. An autonomous decision making method for the operation speed of an unmanned agricultural machine according to claim 2, wherein the operation speed of the unmanned agricultural machine is influenced according to formula (1)vThe most significant 4 factors: difference of elevation
Figure QLYQS_9
Roll angle->
Figure QLYQS_10
Lateral position deviation->
Figure QLYQS_11
Heading angle->
Figure QLYQS_12
Composing 4-dimensional input vectorsm：
Figure QLYQS_13
（2）
Figure QLYQS_14
（3）
wherein ,
Figure QLYQS_15
is the firstiInput vector->
Figure QLYQS_16
Respectively represent the firstiDifference in elevation of the individual input vectors>
Figure QLYQS_17
Roll angle->
Figure QLYQS_18
Lateral position deviation->
Figure QLYQS_19
Heading angle->
Figure QLYQS_20
；
Collecting a plurality of input vectors and output scalar quantities, and establishing a target operation speed training sample D:
Figure QLYQS_21
（4）
wherein ,
Figure QLYQS_22
is the firstiOutput scalar of the individual samples; />
Figure QLYQS_23
Inputting a matrix formed by vectors for all training samples; />
Figure QLYQS_24
Outputting a scalar matrix for all training samples, wherein the training sample set is noisy and obeys a standard Gaussian distribution>
Figure QLYQS_25
。
4. The autonomous decision making method for the operation speed of the unmanned agricultural machinery according to claim 1, wherein the step S2 is specifically:
defining a Gaussian process
Figure QLYQS_26
Is +.about.>
Figure QLYQS_27
and />
Figure QLYQS_28
：
Figure QLYQS_29
（5）
Figure QLYQS_30
（6）
The gaussian process is:
Figure QLYQS_31
（7）
wherein, the random variable parameterxOutput value of (2)
Figure QLYQS_32
Is a random variable;
for the followingnInput number
Figure QLYQS_33
Output->
Figure QLYQS_34
Obeying a joint gaussian distribution:
Figure QLYQS_35
（8）
wherein ,
Figure QLYQS_36
，/>
Figure QLYQS_37
is->
Figure QLYQS_38
A dimension matrix;
for decision regression of target operation speed, a noise-containing function model is adopted:
Figure QLYQS_39
（9）
wherein ,yin order to observe the value of the value,ffor the function value of the function,
Figure QLYQS_40
is thatxA corresponding true value; />
Figure QLYQS_41
Obeying normal distribution, wherein the average value is 0;
observations ofyIs:
Figure QLYQS_42
（10）
given training set
Figure QLYQS_43
Test set->
Figure QLYQS_44
Order-making
Figure QLYQS_45
，/>
Figure QLYQS_46
Observations ofyAnd predictive value->
Figure QLYQS_47
Is:
Figure QLYQS_48
（11）
wherein ,
Figure QLYQS_50
、/>
Figure QLYQS_52
、/>
Figure QLYQS_54
and />
Figure QLYQS_56
Are all the covariance matrices of the two images,
Figure QLYQS_58
is->
Figure QLYQS_60
Symmetric positive definite covariance matrix, matrix element +.>
Figure QLYQS_61
Representation->
Figure QLYQS_49
And
Figure QLYQS_51
correlation between; />
Figure QLYQS_53
Is->
Figure QLYQS_55
A unit matrix; />
Figure QLYQS_57
and />
Figure QLYQS_59
The method comprises the following steps:
Figure QLYQS_62
（12）/>
Figure QLYQS_63
（13）
the key prediction equation for obtaining Gaussian process regression is obtained by adopting Bayesian theory:
Figure QLYQS_64
（14）
wherein ,
Figure QLYQS_65
data point +.>
Figure QLYQS_66
Corresponding mean value->
Figure QLYQS_67
For the corresponding variance:
Figure QLYQS_68
（15）
Figure QLYQS_69
（16）
wherein ,
Figure QLYQS_70
is a predicted value of the target work speed.
5. The autonomous decision making method for the operation speed of the unmanned agricultural machinery according to claim 1, wherein the step S3 is specifically:
s31, determining a covariance function;
s32, setting initial super parameters and optimizing marginal likelihood super parameters.
6. The autonomous decision making method for the operation speed of the unmanned agricultural machinery according to claim 5, wherein the step S31 is specifically:
the Gaussian process regression prediction model adopts a square index covariance kernel function:
Figure QLYQS_71
（17）
wherein ,
Figure QLYQS_72
、lis a hyper-parameter of kernel function, +.>
Figure QLYQS_73
For the signal variance to be a function of the signal variance,ldescribed are scale parameters of the complexity of the kernel function in the parameter space.
7. The autonomous decision making method for the operation speed of the unmanned agricultural machinery according to claim 6, wherein the step S32 is specifically:
input vector of Gaussian process regression prediction model of unmanned agricultural machinery target operation speedmBy difference in elevation
Figure QLYQS_74
Transverse crossRoll angle->
Figure QLYQS_75
Lateral position deviation->
Figure QLYQS_76
And heading angle->
Figure QLYQS_77
Composition, thus for formula (17)lThe determination of (1) employs an automatic determination of the correlation, expressed as:
Figure QLYQS_78
（18）
wherein ,
Figure QLYQS_79
respectively represent the height difference +.>
Figure QLYQS_80
Roll angle->
Figure QLYQS_81
Lateral position deviation->
Figure QLYQS_82
And heading angle->
Figure QLYQS_83
A length scale of each dimension of (a);
the training process of the Gaussian process regression prediction model is regarded as a problem for solving nonlinear numerical optimization, and super-parameter vectors are defined:
Figure QLYQS_84
（19）
by using
Figure QLYQS_85
Representation->
Figure QLYQS_86
The element in (a), namely the super parameter; />
Optimizing the super-parameters by adopting a Maximum Likelihood Estimation (MLE) method, wherein the marginal likelihood function is as follows:
Figure QLYQS_87
（20）
likelihood obeys a gaussian distribution:
Figure QLYQS_88
（21）
the method is obtained by a Gaussian process regression prediction model:
Figure QLYQS_89
（22）
the integration of equation (22) still obeys the gaussian distribution:
Figure QLYQS_90
（23）
wherein ,
Figure QLYQS_91
，/>
Figure QLYQS_92
a covariance matrix which does not consider Gaussian noise;
solving the negative logarithm of the formula (20) to obtain a negative logarithm boundary likelihood function:
Figure QLYQS_93
（24）
solving the minimum of equation (24) to obtain optimized hyper-parameters, i.e. training of Gaussian process regression prediction model is converted into the optimization of equation (24)Problem of chemical transformation, objective function isLThe optimization objective is to solveLIs the minimum of (2);
solving the formula (24) for the super-parameters
Figure QLYQS_94
Is a partial derivative of:
Figure QLYQS_95
（25）
wherein ,
Figure QLYQS_96
，/>
Figure QLYQS_97
representing the trace of the matrix;
the conjugate gradient method is adopted to minimize the partial derivative, so as to obtain the optimal solution of the super parameter, namely, the altitude difference obtained according to the measurement of the unmanned agricultural machinery
Figure QLYQS_98
Roll angle->
Figure QLYQS_99
Lateral position deviation->
Figure QLYQS_100
And heading angle->
Figure QLYQS_101
And predicting the target operation speed to be adjusted through a Gaussian regression process.
8. The autonomous decision making method for the operation speed of the unmanned agricultural machinery according to claim 4, wherein in the step S4, the evaluation of the regression prediction result of the Gaussian process specifically adopts an average absolute errorMAEAnd judging the deviation between the predicted value and the true value of the sample:
Figure QLYQS_102
（26）
wherein ,
Figure QLYQS_103
and />
Figure QLYQS_104
Respectively represent the firstiA true value and a predicted value,Nto predict the number of samples;MAEthe smaller the value, the better the model predictive performance;
autoregressive training samples using a trained Gaussian process regression prediction model according to equation (15), and calculating according to equation (26)MAEJudging whether the training is successful or not, and if the training is unsuccessful, returning to the step S3; otherwise, step S5 is entered.
9. The autonomous decision making method for the operation speed of the unmanned agricultural machinery according to claim 1, wherein the step S5 is specifically:
the trained Gaussian process regression prediction model is deployed on a terminal controller of the unmanned agricultural machine, and data of the unmanned agricultural machine in a period of time is obtained to serve as test data samples
Figure QLYQS_105
Obtaining a corresponding unmanned agricultural machinery prediction speed by adopting a Gaussian process regression prediction model>
Figure QLYQS_106
With this mean value of the predicted speed +.>
Figure QLYQS_107
Decision speed as target->
Figure QLYQS_108
。
10. The autonomous decision making method for the operation speed of the unmanned agricultural machinery according to claim 9, wherein the step S6 is specifically:
will obtain a target decision speed
Figure QLYQS_109
An on-board speed controller transmitted to the unmanned agricultural machinery, the speed controller determining a speed according to the current speed and the target>
Figure QLYQS_110
And the difference value is used for automatically adjusting the operation speed of the unmanned agricultural machinery. />
CN202310293891.3A 2023-03-24 2023-03-24 Unmanned agricultural machinery operation speed autonomous decision-making method Active CN116069043B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310293891.3A CN116069043B (en) 2023-03-24 2023-03-24 Unmanned agricultural machinery operation speed autonomous decision-making method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310293891.3A CN116069043B (en) 2023-03-24 2023-03-24 Unmanned agricultural machinery operation speed autonomous decision-making method

Publications (2)

Publication Number Publication Date
CN116069043A true CN116069043A (en) 2023-05-05
CN116069043B CN116069043B (en) 2023-08-15

Family

ID=86175256

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310293891.3A Active CN116069043B (en) 2023-03-24 2023-03-24 Unmanned agricultural machinery operation speed autonomous decision-making method

Country Status (1)

Country Link
CN (1) CN116069043B (en)

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105701572A (en) * 2016-01-13 2016-06-22 国网甘肃省电力公司电力科学研究院 Photovoltaic short-term output prediction method based on improved Gaussian process regression
CN107315345A (en) * 2017-06-23 2017-11-03 华南农业大学 Take aim at based on double antenna GNSS and in advance the agricultural machinery self-navigation control method of tracing model
CN109978025A (en) * 2019-03-11 2019-07-05 浙江工业大学 A kind of intelligent network connection vehicle front truck acceleration prediction technique returned based on Gaussian process
CN110298122A (en) * 2019-07-03 2019-10-01 北京理工大学 Automatic driving vehicle urban intersection left-hand rotation decision-making technique based on conflict resolution
CN111460380A (en) * 2020-03-30 2020-07-28 上海交通大学 Multi-working-condition driving range prediction method and system based on Gaussian process regression
CN113110526A (en) * 2021-06-15 2021-07-13 北京三快在线科技有限公司 Model training method, unmanned equipment control method and device
CN114162127A (en) * 2021-12-28 2022-03-11 华南农业大学 Paddy field unmanned agricultural machine path tracking control method based on machine tool pose estimation
CN115047207A (en) * 2022-04-14 2022-09-13 华南农业大学 Early warning method for vehicle sinking of unmanned agricultural machine
US20220363259A1 (en) * 2019-11-27 2022-11-17 Momenta (suzhou) Technology Co., Ltd. Method for generating lane changing decision-making model, method for lane changing decision-making of unmanned vehicle and electronic device
CN115407656A (en) * 2022-08-25 2022-11-29 中国科学院计算技术研究所 Intelligent agricultural machinery autonomous operation control method and system based on dynamic self-adaptive LQR
CN115542719A (en) * 2022-09-23 2022-12-30 江苏大学 Combine harvester operation speed control system and method based on multi-operation-parameter reward
CN115635509A (en) * 2022-11-07 2023-01-24 上海大学 Industrial robot position error and confidence interval prediction method based on Gaussian process regression

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105701572A (en) * 2016-01-13 2016-06-22 国网甘肃省电力公司电力科学研究院 Photovoltaic short-term output prediction method based on improved Gaussian process regression
CN107315345A (en) * 2017-06-23 2017-11-03 华南农业大学 Take aim at based on double antenna GNSS and in advance the agricultural machinery self-navigation control method of tracing model
CN109978025A (en) * 2019-03-11 2019-07-05 浙江工业大学 A kind of intelligent network connection vehicle front truck acceleration prediction technique returned based on Gaussian process
CN110298122A (en) * 2019-07-03 2019-10-01 北京理工大学 Automatic driving vehicle urban intersection left-hand rotation decision-making technique based on conflict resolution
US20220363259A1 (en) * 2019-11-27 2022-11-17 Momenta (suzhou) Technology Co., Ltd. Method for generating lane changing decision-making model, method for lane changing decision-making of unmanned vehicle and electronic device
CN111460380A (en) * 2020-03-30 2020-07-28 上海交通大学 Multi-working-condition driving range prediction method and system based on Gaussian process regression
CN113110526A (en) * 2021-06-15 2021-07-13 北京三快在线科技有限公司 Model training method, unmanned equipment control method and device
CN114162127A (en) * 2021-12-28 2022-03-11 华南农业大学 Paddy field unmanned agricultural machine path tracking control method based on machine tool pose estimation
CN115047207A (en) * 2022-04-14 2022-09-13 华南农业大学 Early warning method for vehicle sinking of unmanned agricultural machine
CN115407656A (en) * 2022-08-25 2022-11-29 中国科学院计算技术研究所 Intelligent agricultural machinery autonomous operation control method and system based on dynamic self-adaptive LQR
CN115542719A (en) * 2022-09-23 2022-12-30 江苏大学 Combine harvester operation speed control system and method based on multi-operation-parameter reward
CN115635509A (en) * 2022-11-07 2023-01-24 上海大学 Industrial robot position error and confidence interval prediction method based on Gaussian process regression

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
QINGKUAN MENG等: "Development of agricultural implement system based on machine vision and fuzzy control", 《COMPUTERS AND ELECTRONICS IN AGRICULTURE》, pages 128 - 149 *
何德峰;彭彬彬;顾煜佳;余世明;: "基于高斯过程回归的车辆巡航系统学习预测控制", 上海交通大学学报, no. 09, pages 28 - 33 *
周亚同;赵翔宇;何峰;石超君;: "基于高斯过程混合模型的大气温湿度预测", 农业工程学报, no. 05, pages 227 - 234 *
张智刚;王进;朱金光;胡炼;罗锡文;: "我国农业机械自动驾驶系统研究进展", 农业工程技术, no. 18, pages 25 - 29 *
赵润茂;胡炼;罗锡文;周浩;袁琦堡;张盟;: "基于ARMA的插秧机田间行驶姿态预测", 农业机械学报, no. 1, pages 8 - 12 *

Also Published As

Publication number Publication date
CN116069043B (en) 2023-08-15

Similar Documents

Publication Publication Date Title
CN113874865A (en) Method and device for determining model parameters of a control strategy of a technical system by means of a Bayesian optimization method
KR102225370B1 (en) Prediction system based on parameter improvement through learning and method thereof
US20170091667A1 (en) Machine learning device and method for optimizing frequency of tool compensation of machine tool, and machine tool having the machine learning device
US20180210406A1 (en) Numerical controller and machine learning device
CN112289034A (en) Deep neural network robust traffic prediction method based on multi-mode space-time data
US11579574B2 (en) Control customization system, control customization method, and control customization program
JP6923484B2 (en) Machining condition adjustment device and machine learning device
DE102017104438A1 (en) Machine learning apparatus, laser apparatus and machine learning method
DE102016008994A1 (en) TOOLING MACHINE, SIMULATING DEVICE AND MACHINE INCOMING DEVICE
CN106292550B (en) The Servocontrol device of function with vehicle air-conditioning gain
CN112418498B (en) Temperature prediction method and system for intelligent greenhouse
Trabelsi et al. Fuzzy identification of a greenhouse
JP6703020B2 (en) Control device and machine learning device
JP2021501433A (en) Generation of control system for target system
CN113939775B (en) Method and device for determining a regulating strategy for a technical system
CN113377018A (en) Intelligent control method and intelligent control equipment of air conditioner
CN116069043B (en) Unmanned agricultural machinery operation speed autonomous decision-making method
CN102393645A (en) Control method of high-speed electro-hydraulic proportional governing system
US20190310115A1 (en) Measurement operation parameter adjustment apparatus, machine learning device, and system
CN112925207A (en) Greenhouse environment temperature self-adaption method based on parameter identification
JP6829271B2 (en) Measurement operation parameter adjustment device, machine learning device and system
CN112036630B (en) Highway pavement rainfall distribution estimation method, storage medium and computing equipment
Korniіenko et al. Method and algorithms of nonlinear dynamic processes identification
DE102021204943A1 (en) META-SCHOOLED EVOLUTIONAL STRATEGY BLACKBOX OPTIMIZATION CLASSIFIER
KR20210138498A (en) Device and method for operating a test bench

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