A cargo identification and adaptive control method for automatic loading and unloading vehicles

Abstract

The present invention discloses a cargo identification and adaptive control method for an automatic loading and unloading vehicle, which relates to the field of automation technology, including: obtaining cargo information of a to-be-loaded and unloaded task, analyzing the attribute preference of each cargo information of the to-be-loaded and unloaded task, and evaluating the cargo priority index of each to-be-loaded and unloaded task; obtaining the appearance and posture of the cargo of the to-be-loaded and unloaded task and the cargo information in the loading and unloading task, and analyzing the cargo loading and unloading requirements of the to-be-loaded and unloaded task; obtaining several loading and unloading methods of the automatic loading and unloading vehicle, analyzing the matching degree between the cargo loading and unloading requirements of the to-be-loaded and unloaded task and the several loading and unloading methods, and evaluating the cargo loading and unloading method matching index of each to-be-loaded and unloaded task; uploading the cargo priority index of each to-be-loaded and unloaded task and the cargo loading and unloading method matching index of each to-be-loaded and unloaded task to the task scheduling table of the automatic loading and unloading vehicle, and generating an adaptive control optimal solution. The advantages of the present invention are: improving cargo loading and unloading efficiency and reducing the risk of cargo damage during loading and unloading.

Description

Technical Field

[0001] The present invention relates to the field of automation technology, and in particular to a cargo identification and adaptive control method for an automatic loading and unloading vehicle. Background Art

[0002] The adaptive control of cargo identification of automatic loading and unloading vehicles means that the automatic loading and unloading vehicle system realizes functions such as automatic identification, positioning, grasping and placement of cargo by integrating key components such as sensors, control systems and actuators. The system can also be adaptively adjusted according to actual needs to accommodate cargo of different sizes, shapes and weights.

[0003] Due to the complexity of the cargo currently handled by logistics transfer stations, the classification and loading and unloading of cargo still rely on manpower, and the loading and unloading of cargo cannot be fully monitored, resulting in damage to cargo during manual handling. In addition, the efficiency of manual handling depends on the proficiency of the workers, and the efficiency of cargo loading and unloading is low. Summary of the Invention

[0004] In order to solve the above technical problems, a method for adaptive control of cargo identification of automatic loading and unloading vehicles is provided. This technical solution solves the above-mentioned problem that due to the complexity of the cargo currently faced by logistics transfer stations, the classification and loading and unloading of cargo still rely on manpower, and the loading and unloading of cargo cannot be monitored throughout the process, resulting in frequent damage to the cargo during manual handling, and the efficiency of manual handling depends on the proficiency of the workers, resulting in low efficiency of cargo loading and unloading.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A method for adaptively controlling cargo identification of an automatic loading and unloading vehicle, comprising:

[0007] Based on the background loading and unloading task table, obtain the cargo information of the tasks to be loaded and unloaded, analyze the attribute preferences of the cargo information of each task to be loaded and unloaded, and evaluate the cargo priority index of each task to be loaded and unloaded;

[0008] Using edge scanning sensors, the appearance and posture of the cargo to be loaded and unloaded, as well as cargo information during the loading and unloading task, are acquired to analyze the cargo loading and unloading requirements of the task.

[0009] Obtain several loading and unloading methods for automatic loading and unloading vehicles, analyze the matching degree between the cargo loading and unloading requirements of the loading and unloading tasks and the several loading and unloading methods, and evaluate the matching index of the cargo loading and unloading methods for each loading and unloading task;

[0010] The cargo priority index of each task to be loaded and unloaded and the cargo loading and unloading method matching index of each task to be loaded and unloaded are uploaded to the task scheduling table of the automatic loading and unloading vehicle to generate the optimal adaptive control plan.

[0011] Preferably, based on the background loading and unloading task table, the cargo information of the to-be-loaded and unloaded tasks is obtained, the attribute preference of each cargo information of the to-be-loaded and unloaded tasks is analyzed, and the cargo priority index of each to-be-loaded and unloaded tasks is evaluated. Specifically, the following are included:

[0012] Obtain the current real-time external ambient temperature and record it as the real-time ambient temperature data for the loading and unloading task;

[0013] Based on the cargo information in the task to be loaded and unloaded, the storage temperature attribute of the cargo information is normalized to initialize the temperature attribute preference range of the cargo information in the task to be loaded and unloaded;

[0014] Using the entropy weight method, the temperature attribute preference range of the cargo information to be initialized for loading and unloading tasks is calculated, and the temperature change information entropy when the temperature rises and falls is considered;

[0015] Based on the temperature change information entropy, the difference coefficient of the temperature attribute preference range of the cargo information to be initialized for loading and unloading tasks is calculated, and the temperature attribute preference weight of the cargo information to be loaded and unloaded tasks is evaluated;

[0016] Performing a difference operation based on the temperature attribute preference range of the initialized cargo information to be loaded and unloaded and the real-time ambient temperature data of the cargo information to be loaded and unloaded, thereby obtaining the ambient temperature deviation of the cargo information to be loaded and unloaded;

[0017] Calculate the cargo priority index of each loading and unloading task based on the optimal value of the temperature attribute preference of the cargo information of the loading and unloading task, the temperature attribute preference weight and the ambient temperature deviation of the cargo information of the loading and unloading task;

[0018] The entropy weight method is specifically as follows:

[0019] ;

[0020] Where H j To initialize the temperature attribute preference range value of the cargo information to be loaded and unloaded, the temperature change information entropy when the temperature rises and falls is: j is the difference coefficient of the jth temperature attribute preference range value of the cargo information to be initialized for the loading and unloading task, w j is the temperature attribute preference weight of the cargo information to be loaded and unloaded, P ij is the jth temperature attribute preference range value of the i-th cargo information to be loaded and unloaded, k is the normalized information entropy constant, ln() is the logarithmic function, n is the total number of cargo information to be loaded and unloaded, and m is the temperature attribute preference range;

[0021] The calculation of the cargo priority index for each loading and unloading task is specifically as follows:

[0022] ;

[0023] Where G i is the priority index of the i-th cargo to be loaded and unloaded, ∆T i is the ambient temperature deviation of the i-th cargo information to be loaded and unloaded.

[0024] Preferably, several loading and unloading methods of the automatic loading and unloading vehicle are obtained, and the matching degree between the cargo loading and unloading requirements of the loading and unloading task and the several loading and unloading methods is analyzed. The matching index of the cargo loading and unloading method of the automatic loading and unloading vehicle is evaluated, specifically including:

[0025] Based on edge scanning sensors, the appearance data of the cargo is collected and a 3D posture model of the cargo to be loaded and unloaded is established;

[0026] Based on the cargo information of the to-be-loaded and unloaded task, the physical attributes of the cargo information are obtained, and data simulation fusion is performed with the three-dimensional posture model of the cargo to be loaded and unloaded to obtain a physical simulation three-dimensional posture model of the cargo to be loaded and unloaded; the physical attributes of the cargo information include: cargo type and cargo weight;

[0027] Determining cargo loading and unloading influencing characteristic parameters of the cargo to be loaded and unloaded based on a physical simulation three-dimensional posture model of the cargo to be loaded and unloaded; the cargo loading and unloading influencing characteristic parameters include: size characteristics, shape characteristics, stress characteristics, and dynamic characteristics;

[0028] Obtain several loading and unloading methods of the automatic loading and unloading vehicle, and mark and obtain the loading and unloading action parameters of the several loading and unloading methods of the automatic loading and unloading vehicle according to the output method of the mechanical equipment force of each loading and unloading method;

[0029] Based on random forests, a cargo loading and unloading demand decision tree is constructed for each pending loading and unloading task. The cargo loading and unloading influencing characteristic parameters of the pending loading and unloading task are used as the partitioning features of each leaf node of each cargo loading and unloading demand decision tree. The loading and unloading action parameters of several loading and unloading methods of the automatic loading and unloading vehicle are substituted as the root node. The nodes are divided according to the maximum information gain value to generate the cargo loading and unloading method matching index of the automatic loading and unloading vehicle.

[0030] The maximum value of the information gain of each node is divided into: G(X,Y)=H(Y)-H(Y|X);

[0031] Where G(X,Y) is the information gain between the loading and unloading action parameter Y of several loading and unloading methods of the automatic loading and unloading vehicle and the partition feature X of the leaf node, H(Y) is the information entropy of the loading and unloading action parameter Y, and H(Y|X) is the conditional information entropy of the loading and unloading action parameter Y under the condition of the partition feature X of the leaf node.

[0032] Preferably, the cargo priority index of each to-be-loaded and unloaded task and the cargo loading and unloading mode matching index of each to-be-loaded and unloaded task are uploaded to the task scheduling table of the automatic loading and unloading vehicle, and the adaptive control optimal solution is generated, specifically including:

[0033] Based on multi-objective optimization and machine learning, a decision model for loading and unloading methods of automatic loading and unloading vehicles is constructed;

[0034] The cargo priority index of each loading and unloading task is used as a constraint condition;

[0035] Based on the matching index of the cargo loading and unloading method of each loading and unloading task, the optimal efficiency objective function of the cargo loading and unloading method of the loading and unloading task is constructed;

[0036] The decision model for loading and unloading methods of the tasks to be loaded and unloaded by the automatic loading and unloading vehicle is based on the optimal efficiency objective function under the constraints, taking the cargo information of the tasks to be loaded and unloaded as input and the optimal loading and unloading method of the cargo to be loaded and unloaded as output;

[0037] The decision model for the loading and unloading mode of the automatic loading and unloading vehicle for the task to be loaded and unloaded is specifically as follows:

[0038] ;

[0039] Where, is the vth optimal loading and unloading method for the i-th cargo to be loaded and unloaded, G i is the priority index of the i-th cargo to be loaded and unloaded, R iv To match the vth optimal loading and unloading method for the i-th cargo to be loaded and unloaded, S is the set of loading and unloading methods of the automatic loading and unloading vehicle.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] This invention proposes an adaptive cargo recognition and control solution for automated loading and unloading vehicles. This solution uses a backend loading and unloading task table to obtain cargo information and assess priorities. It then analyzes loading and unloading requirements based on cargo appearance and posture information captured in real time by edge scanning sensors. This solution then compares and evaluates the optimal loading and unloading method with the automated loading and unloading vehicle's various loading and unloading methods. Ultimately, the optimal loading and unloading method is selected and uploaded to the task schedule, forming an adaptive control solution. This solution has the beneficial effects of improving cargo loading and unloading efficiency and reducing the risk of cargo damage during loading and unloading. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a flow chart of a method for adaptively controlling cargo identification of an automatic loading and unloading vehicle;

[0043] Figure 2 A flow chart of cargo priority indicators for evaluating each cargo loading and unloading task;

[0044] Figure 3 A flow chart of the matching indicator method for evaluating the cargo handling method for each loading and unloading task;

[0045] Figure 4 Flowchart of the method for generating the optimal solution for adaptive control. DETAILED DESCRIPTION

[0046] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0047] Reference Figure 1 As shown, a cargo recognition and adaptive control method for an automatic loading and unloading vehicle includes:

[0048] Based on the background loading and unloading task table, obtain the cargo information of the tasks to be loaded and unloaded, analyze the attribute preferences of the cargo information of each task to be loaded and unloaded, and evaluate the cargo priority index of each task to be loaded and unloaded;

[0049] Using edge scanning sensors, the appearance and posture of the cargo to be loaded and unloaded, as well as cargo information during the loading and unloading task, are acquired to analyze the cargo loading and unloading requirements of the task.

[0050] Obtain several loading and unloading methods for automatic loading and unloading vehicles, analyze the matching degree between the cargo loading and unloading requirements of the loading and unloading tasks and the several loading and unloading methods, and evaluate the matching index of the cargo loading and unloading methods for each loading and unloading task;

[0051] The cargo priority index of each task to be loaded and unloaded and the cargo loading and unloading method matching index of each task to be loaded and unloaded are uploaded to the task scheduling table of the automatic loading and unloading vehicle to generate the optimal adaptive control plan.

[0052] This solution uses a backend loading and unloading task table to obtain cargo information and assess priorities. It then analyzes loading and unloading requirements using real-time cargo appearance and posture information captured by edge scanning sensors. This analysis then compares these with various loading and unloading methods available for automated loading and unloading vehicles. Ultimately, the optimal loading and unloading method is selected and uploaded to the task schedule, forming an adaptive control plan. This solution has the beneficial effects of improving cargo loading and unloading efficiency and reducing the risk of cargo damage during loading and unloading.

[0053] Reference Figure 2 As shown, based on the background loading and unloading task table, the cargo information of the loading and unloading tasks to be obtained, the attribute preferences of the cargo information of each loading and unloading task to be analyzed, and the cargo priority indicators of each loading and unloading task to be evaluated include:

[0054] Obtain the current real-time external ambient temperature and record it as the real-time ambient temperature data for the loading and unloading task;

[0055] Based on the cargo information in the task to be loaded and unloaded, the storage temperature attribute of the cargo information is normalized to initialize the temperature attribute preference range of the cargo information in the task to be loaded and unloaded;

[0056] Using the entropy weight method, the temperature attribute preference range of the cargo information to be initialized for loading and unloading tasks is calculated, and the temperature change information entropy when the temperature rises and falls is considered;

[0057] Based on the temperature change information entropy, the difference coefficient of the temperature attribute preference range of the cargo information to be initialized for loading and unloading tasks is calculated, and the temperature attribute preference weight of the cargo information to be loaded and unloaded tasks is evaluated;

[0058] Performing a difference operation based on the temperature attribute preference range of the initialized cargo information to be loaded and unloaded and the real-time ambient temperature data of the cargo information to be loaded and unloaded, thereby obtaining the ambient temperature deviation of the cargo information to be loaded and unloaded;

[0059] Calculate the cargo priority index of each loading and unloading task based on the optimal value of the temperature attribute preference of the cargo information of the loading and unloading task, the temperature attribute preference weight and the ambient temperature deviation of the cargo information of the loading and unloading task;

[0060] The entropy weight method is specifically:

[0061] ;

[0062] Where H j To initialize the temperature attribute preference range value of the cargo information to be loaded and unloaded, the temperature change information entropy when the temperature rises and falls is: j is the difference coefficient of the jth temperature attribute preference range value of the cargo information to be initialized for the loading and unloading task, w j is the temperature attribute preference weight of the cargo information to be loaded and unloaded, P ij is the jth temperature attribute preference range value of the i-th cargo information to be loaded and unloaded, k is the normalized information entropy constant, ln() is the logarithmic function, n is the total number of cargo information to be loaded and unloaded, and m is the temperature attribute preference range;

[0063] The calculation of the cargo priority index for each loading and unloading task is specifically as follows:

[0064] ;

[0065] Where G i is the priority index of the i-th cargo to be loaded and unloaded, ∆T i is the ambient temperature deviation of the i-th cargo information to be loaded and unloaded.

[0066] This solution obtains cargo information and real-time ambient temperature data for the cargo to be loaded and unloaded, normalizes the cargo's storage temperature attributes, and uses the entropy weight method to assess the cargo's temperature attribute preference weights. This method then combines ambient temperature deviations with the optimal temperature attribute preference values ​​to comprehensively calculate each cargo's priority index. This approach offers the potential to scientifically and rationally determine the loading and unloading sequence based on the cargo's temperature sensitivity and real-time environmental conditions, optimize resource allocation, ensure cargo safety and quality, and improve the efficiency and accuracy of loading and unloading operations.

[0067] Reference Figure 3 As shown in the figure, several loading and unloading methods of the automatic loading and unloading vehicle are obtained, and the matching degree between the cargo loading and unloading requirements of the loading and unloading task and the several loading and unloading methods is analyzed. The matching index of the cargo loading and unloading method of the automatic loading and unloading vehicle is evaluated, which specifically includes:

[0068] Based on edge scanning sensors, the appearance data of the cargo is collected and a 3D posture model of the cargo to be loaded and unloaded is established;

[0069] Based on the cargo information of the to-be-loaded and unloaded task, the physical attributes of the cargo information are obtained, and data simulation fusion is performed with the three-dimensional posture model of the cargo to be loaded and unloaded to obtain a physical simulation three-dimensional posture model of the cargo to be loaded and unloaded; the physical attributes of the cargo information include: cargo type and cargo weight;

[0070] Determining cargo loading and unloading influencing characteristic parameters of the cargo to be loaded and unloaded based on a physical simulation three-dimensional posture model of the cargo to be loaded and unloaded; the cargo loading and unloading influencing characteristic parameters include: size characteristics, shape characteristics, stress characteristics, and dynamic characteristics;

[0071] Obtain several loading and unloading methods of the automatic loading and unloading vehicle, and mark and obtain the loading and unloading action parameters of the several loading and unloading methods of the automatic loading and unloading vehicle according to the output method of the mechanical equipment force of each loading and unloading method;

[0072] Based on random forests, a cargo loading and unloading demand decision tree is constructed for each pending loading and unloading task. The cargo loading and unloading influencing characteristic parameters of the pending loading and unloading task are used as the partitioning features of each leaf node of each cargo loading and unloading demand decision tree. The loading and unloading action parameters of several loading and unloading methods of the automatic loading and unloading vehicle are substituted as the root node. The nodes are divided according to the maximum information gain value to generate the cargo loading and unloading method matching index of the automatic loading and unloading vehicle.

[0073] The maximum value of the information gain of each node is divided into: G(X,Y)=H(Y)-H(Y|X);

[0074] Where G(X,Y) is the information gain between the loading and unloading action parameter Y of several loading and unloading methods of the automatic loading and unloading vehicle and the partition feature X of the leaf node, H(Y) is the information entropy of the loading and unloading action parameter Y, and H(Y|X) is the conditional information entropy of the loading and unloading action parameter Y under the condition of the partition feature X of the leaf node.

[0075] This solution collects data on the cargo's appearance and combines it with its physical properties to create a three-dimensional physical simulation model of the cargo, thereby determining the characteristic parameters that influence the loading and unloading method. Subsequently, the loading and unloading action parameters of the automated loading and unloading vehicle are labeled based on the force output of the mechanical equipment used for different loading and unloading methods. A decision tree is constructed using the random forest algorithm, using the characteristic parameters that influence cargo loading and unloading as the basis for classification. The loading and unloading action parameters are then substituted into the decision tree, ultimately generating matching indicators for the automated loading and unloading vehicle's cargo loading and unloading methods. This beneficial effect is the ability to accurately match the cargo to be loaded and unloaded with the most appropriate loading and unloading method, thereby improving loading and unloading efficiency.

[0076] Reference Figure 4 As shown in the figure, the cargo priority index of each to-be-loaded and unloaded task and the cargo loading and unloading method matching index of each to-be-loaded and unloaded task are uploaded to the task scheduling table of the automatic loading and unloading vehicle, and the optimal adaptive control scheme is generated, which specifically includes:

[0077] Based on multi-objective optimization and machine learning, a decision model for loading and unloading methods of automatic loading and unloading vehicles is constructed;

[0078] The cargo priority index of each loading and unloading task is used as a constraint condition;

[0079] Based on the matching index of the cargo loading and unloading method of each loading and unloading task, the optimal efficiency objective function of the cargo loading and unloading method of the loading and unloading task is constructed;

[0080] The decision model for loading and unloading methods of the tasks to be loaded and unloaded by the automatic loading and unloading vehicle is based on the optimal efficiency objective function under the constraints, taking the cargo information of the tasks to be loaded and unloaded as input and the optimal loading and unloading method of the cargo to be loaded and unloaded as output;

[0081] The decision model for the loading and unloading mode of the automatic loading and unloading vehicle for the task to be loaded and unloaded is specifically as follows:

[0082] ;

[0083] Where, is the vth optimal loading and unloading method for the i-th cargo to be loaded and unloaded, G i is the priority index of the i-th cargo to be loaded and unloaded, R iv To match the vth optimal loading and unloading method for the i-th cargo to be loaded and unloaded, S is the set of loading and unloading methods of the automatic loading and unloading vehicle.

[0084] As can be understood, a model for automated loading and unloading vehicle decision-making utilizes multi-objective optimization techniques and machine learning algorithms. This model uses the cargo priority indicator for each loading and unloading task as a constraint, and combines it with a cargo loading and unloading method matching indicator to construct an optimal efficiency objective function. By inputting the cargo information for the loading and unloading task, the model outputs the optimal loading and unloading method that satisfies the constraints and achieves the highest efficiency. This beneficial effect is the adaptive optimization of automated loading and unloading vehicle task scheduling, improving the efficiency and accuracy of loading and unloading operations.

[0085] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for adaptively controlling cargo identification of an automatic loading and unloading vehicle, characterized in that: include: Based on the background loading and unloading task table, obtain the cargo information of the tasks to be loaded and unloaded, analyze the attribute preferences of the cargo information of each task to be loaded and unloaded, and evaluate the cargo priority index of each task to be loaded and unloaded; Using edge scanning sensors, the appearance and posture of the cargo to be loaded and unloaded, as well as cargo information during the loading and unloading task, are acquired to analyze the cargo loading and unloading requirements of the task. Obtain several loading and unloading methods for automatic loading and unloading vehicles, analyze the matching degree between the cargo loading and unloading requirements of the loading and unloading tasks and the several loading and unloading methods, and evaluate the matching index of the cargo loading and unloading methods for each loading and unloading task; Upload the cargo priority index and cargo loading and unloading method matching index of each loading and unloading task to the task scheduling table of the automatic loading and unloading vehicle to generate the optimal adaptive control plan; Among them, based on the background loading and unloading task table, obtain the cargo information of the loading and unloading tasks, analyze the attribute preferences of the cargo information of each loading and unloading task, and evaluate the cargo priority indicators of each loading and unloading task. Specifically, the following are included: Obtain the current real-time external ambient temperature and record it as the real-time ambient temperature data for the loading and unloading task; Based on the cargo information in the task to be loaded and unloaded, the storage temperature attribute of the cargo information is normalized to initialize the temperature attribute preference range of the cargo information in the task to be loaded and unloaded; Using the entropy weight method, the temperature attribute preference range of the cargo information to be initialized for loading and unloading tasks is calculated, and the temperature change information entropy when the temperature rises and falls is considered; Based on the temperature change information entropy, the difference coefficient of the temperature attribute preference range of the cargo information to be initialized for loading and unloading tasks is calculated, and the temperature attribute preference weight of the cargo information to be loaded and unloaded tasks is evaluated; Performing a difference operation based on the temperature attribute preference range of the initialized cargo information to be loaded and unloaded and the real-time ambient temperature data of the cargo information to be loaded and unloaded, thereby obtaining the ambient temperature deviation of the cargo information to be loaded and unloaded; Calculate the cargo priority index of each loading and unloading task based on the optimal value of the temperature attribute preference of the cargo information of the loading and unloading task, the temperature attribute preference weight and the ambient temperature deviation of the cargo information of the loading and unloading task; Among them, several loading and unloading methods of automatic loading and unloading vehicles are obtained, and the matching degree between the cargo loading and unloading requirements of the loading and unloading tasks and the several loading and unloading methods is analyzed. The matching indicators of the cargo loading and unloading methods of automatic loading and unloading vehicles are evaluated, including: Based on edge scanning sensors, the appearance data of the cargo is collected and a 3D posture model of the cargo to be loaded and unloaded is established; Based on the cargo information of the to-be-loaded and unloaded task, the physical attributes of the cargo information are obtained, and data simulation fusion is performed with the three-dimensional posture model of the cargo to be loaded and unloaded to obtain a physical simulation three-dimensional posture model of the cargo to be loaded and unloaded; the physical attributes of the cargo information include: cargo type and cargo weight; Determining cargo loading and unloading influencing characteristic parameters of the cargo to be loaded and unloaded based on a physical simulation three-dimensional posture model of the cargo to be loaded and unloaded; the cargo loading and unloading influencing characteristic parameters include: size characteristics, shape characteristics, stress characteristics, and dynamic characteristics; Obtain several loading and unloading methods of the automatic loading and unloading vehicle, and mark and obtain the loading and unloading action parameters of the several loading and unloading methods of the automatic loading and unloading vehicle according to the output method of the mechanical equipment force of each loading and unloading method; Based on random forests, a cargo loading and unloading demand decision tree is constructed for each pending loading and unloading task. The cargo loading and unloading influencing characteristic parameters of the pending loading and unloading task are used as the partitioning features of each leaf node of each cargo loading and unloading demand decision tree. The loading and unloading action parameters of several loading and unloading methods of the automatic loading and unloading vehicle are substituted as the root node. The nodes are divided according to the maximum information gain value to generate the cargo loading and unloading method matching index of the automatic loading and unloading vehicle. The priority index of each cargo to be loaded and unloaded task and the matching index of each cargo loading and unloading method to be loaded and unloaded are uploaded to the task scheduling table of the automatic loading and unloading vehicle to generate the optimal adaptive control plan. Specifically, the following are included: Based on multi-objective optimization and machine learning, a decision model for loading and unloading methods of automatic loading and unloading vehicles is constructed; The cargo priority index of each loading and unloading task is used as a constraint condition; Based on the matching index of the cargo loading and unloading method of each loading and unloading task, the optimal efficiency objective function of the cargo loading and unloading method of the loading and unloading task is constructed; The decision model for loading and unloading methods of the tasks to be loaded and unloaded by the automatic loading and unloading vehicle is based on the optimal efficiency objective function under the constraints, taking the cargo information of the tasks to be loaded and unloaded as input and the optimal loading and unloading method of the cargo to be loaded and unloaded as output; The decision model for the loading and unloading mode of the automatic loading and unloading vehicle for the task to be loaded and unloaded is specifically as follows: ； Where, is the vth optimal loading and unloading method for the i-th cargo to be loaded and unloaded, G i is the priority index of the i-th cargo to be loaded and unloaded, R iv To match the vth optimal loading and unloading method for the i-th cargo to be loaded and unloaded, S is the set of loading and unloading methods of the automatic loading and unloading vehicle.

2. The method for adaptively controlling cargo identification of an automatic loading and unloading vehicle according to claim 1, characterized in that: The entropy weight method is specifically: ； Where H j To initialize the temperature attribute preference range value of the cargo information to be loaded and unloaded, the temperature change information entropy when the temperature rises and falls is: j is the difference coefficient of the jth temperature attribute preference range value of the cargo information to be initialized for the loading and unloading task, w j is the temperature attribute preference weight of the cargo information to be loaded and unloaded, P ij is the jth temperature attribute preference range value of the i-th cargo information to be loaded and unloaded, k is the normalized information entropy constant, ln() is the logarithmic function, n is the total number of cargo information to be loaded and unloaded, and m is the temperature attribute preference range.

3. The method for cargo identification and adaptive control of an automatic loading and unloading vehicle according to claim 2, characterized in that: The cargo priority index for each loading and unloading task is calculated as follows: ； Where G i is the priority index of the i-th cargo to be loaded and unloaded, ∆T i is the ambient temperature deviation of the i-th cargo information to be loaded and unloaded.

Images