Intelligent cooking control method and system, electronic equipment and storage medium

Abstract

The invention provides an intelligent cooking control method and system, electronic equipment and a storage medium, and belongs to the technical field of automatic cooking equipment. The method comprises the following steps: regulating a dish text library and a standard cooking program library corresponding to dishes in a predetermined range; obtaining a target dish text corresponding to the dish ordering information based on the dish ordering information issued by the restaurant ordering system; matching the target dish text with a preset dish text in a dish text library; calling a target cooking program corresponding to the successfully matched target dish text from a standard cooking program library; controlling the real-time temperature of the wok by adopting a temperature control algorithm when the target cooking program is executed; based on the real-time temperature, the stir-frying state of the stir-frying pan in the stir-frying mode is controlled in combination with a stir-frying control algorithm; when an execution completion instruction is obtained, the dishes in the wok are controlled to be taken out of the wok, and the wok is cleaned. By means of the cooking robot, the defects that an existing cooking robot is poor in frying pan temperature control capacity, poor in stir-frying mode and stir-frying frequency control capacity and the like can be overcome.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic cooking equipment, and specifically relates to an intelligent cooking control method, system, electronic equipment and storage medium. Background Art

[0002] A large number of modern kitchen equipment have appeared at home and abroad, such as rice cookers, soybean milk machines, juicers, meat grinders, etc. The emergence of these equipment has reduced people's kitchen workload to a certain extent; but the preparation of dishes is a complicated process, and it is even more difficult to achieve automatic cooking with machines, such as some different forms of automatic cooking machines on the market.

[0003] At present, the successful development of cooking robots has made further improvements in intelligence compared to automatic cooking machines. However, their intelligence cannot meet people's needs. They are simple in function, complex in operation, and low in intelligence. These cooking robots still need to manually control the time and amount of seasonings during the cooking process. The dishes made by stirring or simply slicing during the cooking process are mashed and have uneven tastes. In particular, it is difficult to effectively control the temperature of the wok, the way of frying, and the frequency of frying, which directly affects the uniformity of the overall dishes during the frying process, resulting in poor quality and taste of the stir-fried dishes.

[0004] Therefore, how to improve the shortcomings of existing cooking robots, such as poor wok temperature control ability, poor control ability of stir-frying mode and stir-frying frequency, and thus improve the intelligence level of cooking robots, is an urgent issue to be solved by technical personnel in this field. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an intelligent cooking control method, system, electronic device and storage medium.

[0006] In the first aspect, the invention provides an intelligent cooking control method, comprising: Regulate the dish text library and standard cooking program library corresponding to the dishes within the predetermined range; Obtain the target dish text corresponding to the order information issued by the restaurant ordering system; Matching the target dish text with the preset dish text in the dish text library; Retrieving the target cooking program corresponding to the target dish text after successful matching from the standard cooking program library; When executing the target cooking program, a temperature control algorithm is used to control the real-time temperature of the wok; Based on the real-time temperature, the stir-frying state of the wok in the stir-frying mode is controlled in combination with a stir-frying control algorithm; wherein the stir-frying mode includes a manual spatula, a pan flipping, a stirring and a roller stir-frying mode; When the execution completion instruction is obtained, the dishes in the wok are controlled to be taken out of the wok and the wok is cleaned.

[0007] Preferably, the step of regulating the dish text library and the standard cooking program library corresponding to the dishes within the predetermined range specifically includes: Encode the dishes within the predetermined range into corresponding preset dish texts according to the text encoding rules, and compile them into a dish text library; Compile standard cooking programs for dishes within a predetermined range according to the corresponding conventional cooking processes to form a standard cooking program library; The dish text library is associated with the standardized cooking process library according to the preset dish text, so that the preset dish text and the standard cooking process corresponding to each dish correspond to each other one by one.

[0008] Preferably, the step of acquiring the target dish text corresponding to the ordering information issued by the restaurant ordering system specifically includes: Scan and receive the ordering information sent by the restaurant's ordering system; Parsing the ordering information to obtain coded text; The coded texts are sorted based on the coding order to obtain target dish texts corresponding to the ordering information.

[0009] Preferably, the step of matching the target dish text with the preset dish text in the dish text library specifically includes: Input the target dish text into the two-stream self-attention model for vectorization processing; The trained XLNet model is used to extract the vectorized target dish text to obtain the corresponding text vector; Based on the text vector, the preset dish text in the dish text library is matched synchronously, and the preset dish text matching the target dish text is retrieved.

[0010] Preferably, the step of using a temperature control algorithm to control the real-time temperature of the wok when executing the target cooking program specifically includes: The real-time deviation between the temperature control enabling parameter and the wok temperature is processed by an analog-to-digital conversion circuit to obtain an accurate deviation value; Inputting the accurate value of the deviation into a fuzzy controller to generate a control variable; Fuzzifying the control variables to obtain a subset of real-time deviations; Performing fuzzy decision making on the subset according to the inference rule to determine the output value of the fuzzy controller; Defuzzification and digital-to-analog conversion are performed on the output quantity to obtain a temperature-corrected analog quantity to control the real-time temperature of the wok.

[0011] Preferably, the step of controlling the stir-frying state of the wok in the stir-frying mode based on the real-time temperature and in combination with a stir-frying control algorithm specifically includes: Selecting a stir-frying mode suitable for the ordering information according to the target cooking program; Adopting adaptive control algorithm to control the speed and position of the wok in the selected stir-fry mode; Using a preset optimization algorithm to adjust the predetermined parameters in the adaptive control algorithm to optimize the speed and position of the wok in the selected stir-fry mode; Adjust the real-time status of frying in the wok according to the optimized speed and position of the wok.

[0012] Preferably, the preset optimization algorithm is specifically an improved genetic algorithm, which includes parameter encoding, initial population setting, fitness function design, and genetic operator design.

[0013] In the second aspect, an intelligent cooking control system includes: The regulation module regulates the dish text library and standard cooking program library corresponding to the dishes within the predetermined range; The parsing module obtains the corresponding target dish text based on the order information issued by the restaurant ordering system; A matching module, used for matching the target dish text with the preset dish text in the dish text library; A calling module, used for calling the target cooking program corresponding to the target dish text after successful matching from the standard cooking program library; A temperature control module, used to control the real-time temperature of the wok using a temperature control algorithm when executing the target cooking program; A flipping control module, for controlling the flipping state of the wok in the flipping mode based on the real-time temperature and in combination with a flipping control algorithm; wherein the flipping mode includes a manual spatula, flipping the wok, stirring and a roller flipping mode; The overlay control module is used to control the dishes in the wok to be taken out of the wok and clean the wok when the execution is completed.

[0014] Preferably, the regulatory module specifically includes: An encoding unit, used to encode dishes within a predetermined range into corresponding preset dish texts according to text encoding rules, and compile them into a dish text library; A compilation unit compiles standard cooking programs for dishes within a predetermined range according to corresponding conventional cooking processes to form a standard cooking program library; The association unit is used to associate the dish text library with the standardized cooking process library according to the preset dish text, so that the preset dish text and the standard cooking process corresponding to each dish correspond one to one.

[0015] Preferably, the parsing module includes: A scanning unit, used to scan and receive the ordering information sent by the restaurant ordering system; A parsing unit, used for parsing the ordering information to obtain a coded text; The sorting unit is used to sort the coded text based on the coding order to obtain the target dish text corresponding to the ordering information.

[0016] Preferably, the matching module specifically includes: A vectorization unit, used for inputting the target dish text into a two-stream self-attention model for vectorization processing; An extraction unit, used for extracting the vectorized target dish text using the trained XLNet model to obtain a text vector corresponding thereto; A matching unit is used to match the preset dish text in the dish text library based on the text vector synchronization, and retrieve the preset dish text that matches the target dish text.

[0017] Preferably, the temperature control module specifically includes: A conversion unit, used for processing the real-time deviation between the temperature control enabling parameter and the wok temperature through an analog-to-digital conversion circuit to obtain an accurate deviation value; An input unit, used for inputting the accurate value of the deviation into the fuzzy controller to generate a control variable; A fuzzification unit, used for performing fuzzification processing on the control variables to obtain a subset of real-time deviations; An inference unit, used for making a fuzzy decision for the subset according to an inference rule to determine an output value of the fuzzy controller; The reverse unit is used to perform defuzzification and digital-to-analog conversion processing on the output quantity to obtain the analog quantity after temperature correction so as to control the real-time temperature of the wok.

[0018] Preferably, the flip control module specifically includes: A selection unit, configured to select a stir-frying mode suitable for the ordering information according to the target cooking program; An adaptive unit, used to control the speed and position of the wok in the selected stir-fry mode by using an adaptive control algorithm; A setting unit, used for setting predetermined parameters in the adaptive control algorithm by using a preset optimization algorithm, so as to optimize the speed and position of the wok in the selected stir-fry mode; The adjustment unit is used to adjust the real-time state of frying in the wok according to the optimized speed and position of the wok.

[0019] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the intelligent cooking control method as described in the first aspect is implemented.

[0020] In a fourth aspect, the present application provides a storage medium on which a computer program is stored, and when the program is executed by a processor, the intelligent cooking control method as described in the first aspect is implemented.

[0021] Compared with the prior art, the present invention has the following beneficial effects: 1. By performing text encoding and standard cooking program processing on each dish to be stir-fried within a predetermined range, a dish text library and a standard cooking program library are obtained; then, based on the preset dish text in the dish text library, the dish text library is associated with the standard cooking program library. When any dish needs to be stir-fried, the dish can be quickly matched with the stir-frying program, thereby improving the processing efficiency of the cooking robot.

[0022] 2. By processing the target dish text of the dishes to be stir-fried through the dual-stream self-attention model and the XLNet model, the target dish text matching task can be achieved with higher accuracy and robustness, and the matching task of long texts can be better handled, thus avoiding the impact of long-term matching on the operating efficiency of the cooking robot.

[0023] 3. By using fuzzy and CNC conversion methods to control the temperature of the cooking robot, the temperature control accuracy and response speed of the wok can be improved, the degree of uniform frying of dishes during the frying process can be improved, and the quality of frying of dishes can be improved.

[0024] 4. By improving the genetic algorithm and combining it with the real-time temperature of the wok, the frequency, speed and position of the wok during frying are continuously corrected, thereby improving the uniformity of heating of the dishes during the frying process and improving the quality of the dishes. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0026] Figure 1 A flow chart of the intelligent cooking control method provided in Example 1 of the present invention; Figure 2 is a structural block diagram of an intelligent cooking control system corresponding to the method of embodiment 1 provided in embodiment 2 of the present invention; Figure 3 It is a schematic diagram of the hardware structure of the computer provided in Example 3 of the present invention.

[0027] Description of reference numerals: 10-regulatory module, 11-coding unit, 12-compilation unit, 13-association unit; 20-analysis module, 21-scanning unit, 22-analysis unit, 23-sorting unit; 30-matching module, 31-vectorization unit, 32-extraction unit, 33-matching unit; 40-Call module; 50-temperature control module, 51-conversion unit, 52-input unit, 53-fuzzification unit, 54-reasoning unit, 55-reverse unit; 60-turn control module, 61-selection unit, 62-adaptive unit, 63-setting unit, 64-adjustment unit; 70- stacking control module; 80 - bus, 81 - processor, 82 - memory, 83 - communication interface. DETAILED DESCRIPTION

[0028] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more comprehensive and complete and will fully convey the concept of the example embodiments to those skilled in the art.

[0029] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0030] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0031] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0032] Example 1 Specifically, Figure 1 Shown is a flow chart of the intelligent cooking control method provided in this embodiment.

[0033] like Figure 1 As shown, the intelligent cooking control method of this embodiment includes the following steps: S101, regulating the dish text library and standard cooking program library corresponding to the dishes within a predetermined range.

[0034] Specifically, for a cooking robot to achieve full automation from ordering to cooking and serving, textual, digital, and intelligent processing of dishes in the restaurant it operates is an indispensable prerequisite. Whether it is a WeChat ordering applet, a restaurant backend management system, or a smart cooking machine control system, a standardized dish and the corresponding cooking program can be used to achieve the intelligent implementation of the above operations.

[0035] Furthermore, the specific steps of step S101 include: S1011, encode the dishes within the predetermined range into corresponding preset dish texts according to text encoding rules, and compile them into a dish text library.

[0036] Specifically, in restaurant operations, consumers' choices of dishes are not limited to the requirements on the menu. For example, there are mild, medium, and extra spicy options for spiciness; there are 3-well, 5-well, 7-well, and 10-well options for doneness; there are also taboos on dish condiments, such as no spicy food, no onions, no ginger, and no garlic. In order to facilitate the intelligent processing of special requirements for dishes by the cooking robot, it is necessary to encode the dishes commonly available on the market in text type so that the cooking robot can know the consumer's requirements for dishes. This embodiment sets a text encoding rule, which is specifically: popular dishes + spiciness level + degree of doneness + special requirements; among which, the spiciness level is 01-slightly spicy, 02-medium spicy, 03-super spicy; the degree of doneness is 13-3 well-done, 15-5 well-done, 17-7 well-done, 10-10 well-done; special requirements are 21-no spicy, 22-no onion, 23-no ginger, 24-no garlic, 05-no cooking wine. If a consumer orders a portion of fish-flavored shredded pork, and asks for slightly spicy, 10 well-done, and no ginger, then the dish ordered by the consumer is encoded by the encoding rule, and it is presented as "fish-flavored shredded pork + 01 + 10 + 23". All dishes are encoded into corresponding preset dish texts according to the above text encoding rules as needed, thereby forming a dish text library.

[0037] S1012, compiling standard cooking programs for dishes within a predetermined range according to corresponding conventional cooking processes to form a standard cooking program library.

[0038] Specifically, the cooking experience is digitized, that is, under the guidance of the chef, quantitative and qualitative analysis of the dishes is carried out, and the data of the best quality of the dishes is found through a large number of cooking experiments. These data are solidified by the program to control the operation of the equipment and form a standard cooking procedure. The specific process includes: First, the ingredients of the dishes are standardized. The amount of main ingredients, auxiliary ingredients, and condiments corresponding to each dish is fixed to ensure the stability of the quality of the dishes. Second, the standardization of the dish processing process, the processing process of the dishes is compiled into a program, and the order of processing cannot be changed at will, so that each dish has its own fixed processing procedure. Third, the process parameters are digitized, and the firepower selected for each dish, the duration of each action, the amount of liquid seasoning injected, etc. are precisely controlled.

[0039] S1013, associating the dish text library with the standardized cooking process library according to the preset dish text, so that the preset dish text and the standard cooking process corresponding to each dish correspond one to one.

[0040] Specifically, before the dish text library is associated with the standard cooking process library, this embodiment uses the TextSeek text search method to first search for popular dishes, and then search for the spicy level, degree of doneness, and special requirements in turn. Although the TextSeek software takes the most time to search, its search accuracy is very high, the indexing time is the least, and the search accuracy rate greatly exceeds other software. Avoid repeated search operations caused by poor search accuracy of other software, which affects the overall running time. In this embodiment, the dish text library is associated with the standard cooking process library, such as using PyCharm software. When opening a file in the editor, PyCharm will select the file type and the corresponding language service according to the file name pattern. If the file name does not match any pattern registered for the file type, the file name pattern can be associated with a specific file type.

[0041] S102, obtaining the target dish text corresponding to the ordering information sent by the restaurant ordering system.

[0042] Specifically, the intelligent cooking machine uses the main control chip as the core to control various peripherals. The main control MCU communicates with the touch screen, barcode scanner, and 4G module through the RS232 serial port. The lower-level main control MCU and the 4G module realize serial port communication, and can obtain the user's ordering menu data transmitted from the upper-level restaurant information management system, and can upload the machine's operating status and other information to the restaurant information management system. In this embodiment, the target dish text ordered by the consumer is obtained by cooperating with the barcode scanner, RS232 serial port, and main control chip.

[0043] Furthermore, the specific steps of step S102 include: S1021, scan and receive the ordering information sent by the restaurant ordering system.

[0044] Specifically, the main control MCU communicates with the barcode scanner through the RS232 serial port. When the feeding slide moves to the barcode scanner position, the RS232 serial port interrupt is triggered, and the barcode scanner data processing task is enabled in the interrupt service function of the RS232 serial port.

[0045] S1022, parsing the ordering information to obtain a coded text.

[0046] Specifically, the data sent back by the scanner is parsed and matched one by one with the ordering information issued by the restaurant information management system. When the match is successful, the cooking process is started. If the match fails, the feeding slide continues to move until the match is successful. If the feeding slide has not been matched when it moves to the limit, a prompt command is issued and the slide quickly returns to its position.

[0047] S1023, sorting the coded text based on the coding order to obtain the target dish text corresponding to the ordering information.

[0048] Specifically, according to the above-mentioned text encoding rules (popular dishes + spiciness level + doneness + special requirements), the obtained different encoded texts are sorted according to such encoding order, and the ordering information is converted into the corresponding target dish text.

[0049] S103, matching the target dish text with the preset dish texts in the dish text library.

[0050] Specifically, early matching methods explored matching problems from the perspective of character similarity, such as matching algorithms such as function method, whole word matching method, fuzzy query method, etc. Such methods involve a large number of geometric operations and have low execution efficiency. This embodiment adopts a matching framework based on the XLNet algorithm. XLNet uses an AR method from left to right to input data, but inside the model, AE is implemented through an attention mask method, thereby making full use of the bidirectional information of the sequence.

[0051] Furthermore, the specific steps of step S103 include: S1031, input the target dish text into the dual-stream self-attention model for vectorization processing.

[0052] Specifically, the use of the traditional Transformer structure in the specific implementation of the arrangement language will lead to the problem of ignoring the word order of the previous text during prediction. This defect may cause the model to obtain incorrect representation results. To solve this problem, XLNet adopts a dual-stream self-attention model to simultaneously perceive the position and content information of the predicted word. The "dual-stream self-attention" in the dual-stream self-attention model refers to the content stream self-attention and the query stream attention. The content information stream is a representation of contextual information, including the current word, and the query information stream contains contextual information other than the current word and the position information of the current word, but does not contain the content information of the current word. When predicting other words, the information itself can be seen in the content information stream. In this embodiment, the target dish text must first undergo data cleaning preprocessing and then vectorization before it can be input into the XLNet matching model for matching.

[0053] S1032, using the trained XLNet model to extract the vectorized target dish text to obtain the corresponding text vector.

[0054] Specifically, the traditional vectorization method uses the Word2vec model, but it cannot handle long text problems, which will cause the phenomenon that different requirements of the same name cannot be distinguished during the matching process. Therefore, this embodiment uses the XLNet model for text vector representation. The XLNet model is based on the AR model and the AE model. It retains the advantages of the AR model and combines the principles of the BERT model to make up for the shortcomings of the autoregressive model that only considers unilateral semantics, so that the model can take into account the problem of long text.

[0055] S1033, based on the text vector synchronization, matching is performed with the preset dish text in the dish text library, and the preset dish text matching the target dish text is retrieved.

[0056] Specifically, this embodiment uses the XLNet model to first clean the target dish text pair of the input text, then input the processed text to be matched into the two-stream self-attention model for feature extraction to generate a text vector, and then trains the XLNet model using the prepared training text, and uses the Softmax function and loss function to optimize the model parameters so that it can accurately predict the long text matching situation. The experimental and evaluation results show that the XLNet model has a more comprehensive understanding of long texts than other models, can obtain more accurate semantic representations, has higher accuracy and robustness in matching tasks, and can better handle long text matching tasks.

[0057] S104, retrieving the target cooking program corresponding to the target dish text after successful matching from the standard cooking program library.

[0058] Specifically, from the above steps, it can be seen that the standard cooking program includes: standardization of ingredients for dishes, standardization of dish processing, and dataization of process parameters. The order information issued by the restaurant ordering system is encoded from the target dish text through text encoding rules. Based on the target dish text, the XLNet model is used to retrieve the preset dish text that matches the target dish text, so that the target cooking program can be retrieved from the standard cooking program according to the matching preset dish text. The entire retrieval process is accurate, fast, and efficient.

[0059] S105, using a temperature control algorithm to control the real-time temperature of the wok when executing the target cooking program.

[0060] Specifically, in terms of wok temperature control, this embodiment adopts a temperature control method of fuzzy and analog-to-digital conversion processing. Compared with other temperature control effects, the temperature control algorithm of fuzzy and analog-to-digital conversion processing greatly improves the system temperature rise response speed and control accuracy.

[0061] Furthermore, the specific steps of step S105 include: S1051: The real-time deviation between the temperature control enabling parameter and the wok temperature is processed by an analog-to-digital conversion circuit to obtain an accurate deviation value.

[0062] Specifically, the main control MCU can obtain the size of the real-time temperature control enabling parameter according to the specific time period in the target cooking program. However, in the actual stir-frying process, there is bound to be a difference between the size of the wok temperature and the current temperature control enabling parameter. In order to improve the corresponding heating speed and control accuracy of the wok, it is necessary to adjust the existing difference. This embodiment adopts a fuzzy temperature control method, first inputting the real-time deviation between the temperature control enabling parameter and the wok temperature into the analog-to-digital conversion circuit, and calculating the exact value of the deviation.

[0063] S1052, inputting the accurate value of the deviation into the fuzzy controller to generate a control variable.

[0064] Specifically, in general, according to the number of input variables, the fuzzy controller can be divided into one-dimensional, two-dimensional and three-dimensional fuzzy controllers, corresponding to one, two and three input variables respectively. In this embodiment, considering the control accuracy and complexity, a two-dimensional fuzzy controller is adopted. The input of the two-dimensional fuzzy controller is the deviation value and the deviation change rate. After the input is fuzzified, fuzzy reasoning is performed according to the rules set by the fuzzy system to obtain the corresponding control parameters, which can well reflect the dynamic characteristics of the output of the controlled system and improve the dynamic response capability of the system.

[0065] S1053, fuzzy processing is performed on the control variables to obtain a subset of real-time deviations.

[0066] Specifically, in this embodiment, before the fuzzy controller processes the numerical value, it needs to be discretized, such as setting the fuzzy subsets of the system input and output to {NB, NM, NS, ZO, PS, PM, PB}, and the fuzzy domain uses 13 levels.

[0067] S1054, performing fuzzy decision making on the subset according to the inference rule to determine the output value of the fuzzy controller.

[0068] Specifically, after the input is fuzzified, the satisfaction degree of each inference rule for each part of the premise will be known, and a number representing the result of the premise of the rule will be obtained by applying the fuzzy operator. The input of the fuzzy operator is two or more membership values ​​from the fuzzy input variables, and the output is a single true value. In the heating system of this embodiment, in order to make the system have good dynamic characteristics, the fuzzy operator parameters need to be adjusted dynamically in real time. The fuzzy decision is based on the manual adjustment strategy of engineers with deep experience in electromagnetic heating fuzzy control.

[0069] S1055, performing defuzzification and digital-to-analog conversion processing on the output quantity to obtain a temperature-corrected analog quantity to control the real-time temperature of the wok.

[0070] Specifically, the control quantity formed after fuzzy reasoning cannot be used directly for temperature control. It is necessary to convert the fuzzy quantity of the abstract value into an accurate value that can be recognized by the system, so a defuzzification operation is required. Generally, the defuzzification methods include the maximum membership method, the weighted average method and the center of gravity method. In this embodiment, the center of gravity method is selected as the defuzzification method. The output value of the defuzzification processing of the center of gravity method is the center of gravity of the area formed by the membership function curve and the horizontal coordinate. The output is more stable, and the output will change with the adjustment of the input, so that the real-time temperature of the wok can be accurately controlled.

[0071] S106, based on the real-time temperature, in combination with a stir-fry control algorithm, controlling the stir-frying state of the wok in the stir-frying mode.

[0072] The stir-frying modes include manual spatula, pan flipping, stirring and roller stir-frying modes.

[0073] Specifically, during the operation of the cooking robot, the position and posture of the pot will change continuously with the rotation of the motor, the weight of the ingredients in the pot is often different, and the position and state of the ingredients will change continuously in the pot as they are stir-fried, causing the resistance to the motor to change continuously during startup and stir-frying. Since the motor itself is nonlinear, and these disturbances and changes in system parameters will cause serious nonlinearity and multivariable coupling in the control system, it is impossible to accurately construct the motion model of the motor. In this embodiment, a stir-fry control algorithm is formulated, specifically a model-free adaptive control algorithm, which is simple to control while improving the accuracy of motor control, and can solve the problem that traditional PID control cannot achieve effective control effects in this case.

[0074] Furthermore, the specific steps of step S106 include: S1061, selecting a stir-frying mode suitable for the ordering information according to the target cooking program.

[0075] Specifically, the stir-frying mode of the dishes to be stir-fried according to the standard stir-frying procedure can be learned based on the ordering information sent by the restaurant ordering system. In this embodiment, the cooking techniques of professional chefs are studied, and the cooking effects of four stir-frying modes, namely, manual spatula stir-frying, pan flipping, stirring cooking robot and drum cooking robot, are compared. On this basis, a pot motion scheme of pan flipping plus drum stir-frying mode similar to the cooking effect of chefs is adopted; a three-dimensional model of the cooking robot mechanism is established, and the motion control strategy of the stir-frying mechanism is determined.

[0076] S1062, using an adaptive control algorithm to control the speed and position of the wok in the selected stir-fry mode.

[0077] Specifically, the adaptive control algorithm uses a pseudo gradient vector and a pseudo order to replace the general nonlinear system model, estimates the pseudo gradient vector of the controlled system in real time online, and uses a series of dynamic linear models to control the trajectory of the controlled system, thereby achieving adaptive control. In this embodiment, the adaptive control algorithm uses the MFAC control algorithm.

[0078] S1063, using a preset optimization algorithm to adjust the predetermined parameters in the adaptive control algorithm to optimize the speed and position of the wok in the selected stir-fry mode.

[0079] Specifically, several parameters that usually affect the control effect of genetic algorithms include ρ k , η k , λ, μ, but η k The influence on the control effect is small, so this study k , λ, μ are optimized. Wherein, the preset optimization algorithm is specifically an improved genetic algorithm, which includes parameter encoding, initial population setting, fitness function design, and genetic operator design. Specifically, common encoding methods include binary encoding, symbolic encoding, and real number encoding. Because real number encoding has higher precision, this embodiment uses real number encoding to encode the ranges of the three parameters. The initial population setting is specifically achieved by randomly generating 50 individual samples within a given range, and the number of iterations is set to 100 times. The fitness function, as an indicator to measure the quality of individuals in a population, has a great influence on the convergence speed of the algorithm, and even determines whether the algorithm can find the optimal solution; this embodiment adds control quantity and rise time as constraints on the basis of the ITAE indicator to further improve the accuracy and practicality of the fitness function. The three main operations of genetic algorithms include selection, crossover and mutation; the selection operator uses the fitness scaling selection method to select excellent individuals; the crossover operator uses two-point crossover, randomly sets two crossover points in certain individuals, and then performs partial gene exchange; the mutation operator applies the mutation operation to the population, changes the gene values ​​at the gene loci of certain individuals in the population to obtain the next generation.

[0080] S1064, adjusting the real-time state of frying in the wok according to the optimized speed and position of the wok.

[0081] Specifically, this embodiment combines the above-mentioned control of heating uniformity in the cooking process, and through algorithm optimization, realizes precise control of the speed and position of the working motor, improves the operation stability of the cooking robot, and further improves the system control effect.

[0082] S107, when the execution completion instruction is obtained, the dish in the wok is controlled to be taken out of the wok and the wok is cleaned.

[0083] Specifically, after the ordering information sent by the restaurant's ordering system is completed according to the corresponding standard cooking program, the main control MCU triggers the temperature control module and the motor module that drives the wok to stir-fry to shut down, and issues an instruction to complete the stir-frying task. The wok is then started to move and flip to take the stir-fried dishes out of the pot, and the cleaning module is started to clean the wok. After cleaning, the wok is controlled to reset.

[0084] In summary: First, by processing each dish to be stir-fried within a predetermined range, a dish text library and a standard cooking program library are obtained; then, based on the preset dish text in the dish text library, the dish text library is associated with the standard cooking program library, so that the dish can be quickly matched with the stir-frying program, thereby improving the processing efficiency of the cooking robot. Secondly, by processing the target dish text of the dish to be stir-fried by the dual-stream self-attention model and the XLNet model, the target dish text matching task can have a higher accuracy and robustness, and can better handle the matching task of long texts. Secondly, by controlling the temperature of the cooking robot by fuzzification and CNC conversion, the temperature control accuracy and response speed of the wok can be improved, and the degree of uniform stir-frying of the dishes during the stir-frying process can be improved. Thirdly, by improving the genetic algorithm and combining the real-time wok temperature, the frequency, speed and position of the wok in the stir-frying process are continuously corrected, thereby improving the uniformity of the heating of the dishes during the stir-frying process of the wok. Finally, when the execution command is obtained, the dishes in the wok are controlled to be taken out of the wok and the wok is cleaned.

[0085] Example 2 This embodiment provides a structural block diagram of a system corresponding to the method described in Embodiment 1. Figure 3 is a block diagram of the intelligent cooking control system according to this embodiment. Figure 3 As shown, the system includes: Regulation module 10, regulating the dish text library and standard cooking program library corresponding to the dishes within the predetermined range; The parsing module 20 obtains the target dish text corresponding to the ordering information issued by the restaurant ordering system; A matching module 30, used for matching the target dish text with the preset dish text in the dish text library; A calling module 40 is used to call the target cooking program corresponding to the target dish text after successful matching from the standard cooking program library; The temperature control module 50 is used to control the real-time temperature of the wok by using a temperature control algorithm when executing the target cooking program; A flipping control module 60, for controlling the frying state of the wok in a frying mode based on the real-time temperature and in combination with a frying control algorithm; wherein the frying mode includes a manual spatula, a pan flipping, a stirring, and a roller frying mode; The overlapping control module 70 is used to control the dishes in the wok to be taken out of the wok and clean the wok when the execution completion instruction is obtained.

[0086] Furthermore, the regulation module 10 specifically includes: The encoding unit 11 is used to encode the dishes within the predetermined range into corresponding preset dish texts according to the text encoding rules, and compile them into a dish text library; The compilation unit 12 compiles standard cooking programs for the dishes within the predetermined range according to the corresponding conventional cooking processes to form a standard cooking program library; The associating unit 13 is used to associate the dish text library with the standardized cooking process library according to the preset dish text, so that the preset dish text and the standard cooking process corresponding to each dish correspond one to one.

[0087] Furthermore, the analysis module 20 includes: Scanning unit 21, used for scanning and receiving ordering information sent by the restaurant ordering system; A parsing unit 22, used for parsing the ordering information to obtain a coded text; The sorting unit 23 is used to sort the coded text based on the coding order to obtain the target dish text corresponding to the ordering information.

[0088] Furthermore, the matching module 30 specifically includes: A vectorization unit 31, used for inputting the target dish text into a dual-stream self-attention model for vectorization processing; An extraction unit 32 is used to extract the target dish text processed by vectorization using the trained XLNet model to obtain a text vector corresponding thereto; The matching unit 33 is used to match the preset dish text in the dish text library based on the text vector synchronization, and retrieve the preset dish text that matches the target dish text.

[0089] Furthermore, the temperature control module 50 specifically includes: A conversion unit 51 is used to obtain an accurate value of the deviation by processing the real-time deviation between the temperature control enabling parameter and the wok temperature through an analog-to-digital conversion circuit; An input unit 52, used for inputting the accurate value of the deviation into the fuzzy controller to generate a control variable; A fuzzification unit 53, used for performing fuzzification processing on the control variables to obtain a subset of real-time deviations; An inference unit 54, configured to perform fuzzy decision making on the subset according to an inference rule to determine an output value of the fuzzy controller; The reverse unit 55 is used to perform defuzzification and digital-to-analog conversion processing on the output quantity to obtain the analog quantity after temperature correction to control the real-time temperature of the wok.

[0090] Furthermore, the flip control module 60 specifically includes: A selection unit 61, configured to select a stir-frying mode suitable for the ordering information according to the target cooking program; An adaptive unit 62, for controlling the speed and position of the wok in the selected stir-fry mode by using an adaptive control algorithm; A setting unit 63 is used to use a preset optimization algorithm to set the predetermined parameters in the adaptive control algorithm to optimize the speed and position of the wok in the selected stir-fry mode; wherein the preset optimization algorithm is specifically an improved genetic algorithm, which includes parameter encoding, initial population setting, fitness function design, and genetic operator design; The adjustment unit 64 is used to adjust the real-time state of frying in the wok according to the optimized speed and position of the wok.

[0091] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0092] Example 3 Combination Figure 1 The described intelligent cooking control method can be implemented by a computer. Figure 3 Schematic diagram of the hardware structure of a computer according to this embodiment.

[0093] The computer may include a processor 81 and a memory 82 storing computer program instructions.

[0094] Specifically, the processor 81 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the present application.

[0095] Among them, the memory 82 may include a large-capacity memory for data or instructions. For example, but not limitation, the memory 82 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 82 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 82 may be inside or outside the data processing device. In a specific embodiment, the memory 82 is a non-volatile memory. In a specific embodiment, the memory 82 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, PROM for short), an erasable PROM (Erasable Programmable Read-Only Memory, EPROM for short), an electrically erasable PROM (Electrically Erasable Programmable Read-Only Memory, EEPROM for short), an electrically alterable ROM (Electrically Alterable Read-Only Memory, EAROM for short) or a flash memory (FLASH) or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (Static Random-Access Memory, referred to as SRAM) or a dynamic random access memory (Dynamic Random Access Memory, referred to as DRAM), wherein the DRAM can be a fast page mode dynamic random access memory (Fast Page Mode Dynamic Random Access Memory, referred to as FPMDRAM), an extended data output dynamic random access memory (Extended Date Out Dynamic Random Access Memory, referred to as EDODRAM), a synchronous dynamic random access memory (Synchronous Dynamic Random-Access Memory, referred to as SDRAM), etc.

[0096] The memory 82 may be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 81 .

[0097] The processor 81 implements the intelligent cooking control method of the above-mentioned embodiment 1 by reading and executing the computer program instructions stored in the memory 82.

[0098] In some embodiments, the computer may further include a communication interface 83 and a bus 80. Figure 3 As shown, the processor 81, the memory 82, and the communication interface 83 are connected via a bus 80 and communicate with each other.

[0099] The communication interface 83 is used to realize the communication between the modules, devices, units and / or equipment in the present application. The communication interface 83 can also realize data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage and image / data processing workstations.

[0100] The bus 80 includes hardware, software or both, and couples the components of the computer to each other. The bus 80 includes, but is not limited to, at least one of the following: a data bus (DataBus), an address bus (AddressBus), a control bus (ControlBus), an expansion bus (ExpansionBus), and a local bus (LocalBus). By way of example and not limitation, the bus 80 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB) bus, a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, or a 100% Ethernet bus. Bus 80 may include one or more buses, memory bus, Micro Channel Architecture (MCA) bus, Peripheral Component Interconnect (PCI) bus, PCI-Express (PCI-X) bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local Bus (VLB) bus or other suitable buses or a combination of two or more of these. Where appropriate, bus 80 may include one or more buses. Although the present application describes and illustrates a specific bus, the present application contemplates any suitable bus or interconnect.

[0101] The computer can obtain the intelligent cooking control system and execute the intelligent cooking control method of Example 1.

[0102] In addition, in combination with the intelligent cooking control method in the above embodiment 1, the present application can provide a storage medium for implementation. The storage medium stores computer program instructions; when the computer program instructions are executed by the processor, the intelligent cooking control method in the above embodiment 1 is implemented.

[0103] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent cooking control method, characterized in that: include: Regulate the dish text library and standard cooking program library corresponding to the dishes within the predetermined range; Obtain the target dish text corresponding to the order information issued by the restaurant ordering system; Matching the target dish text with the preset dish text in the dish text library; Retrieving the target cooking program corresponding to the target dish text after successful matching from the standard cooking program library; When executing the target cooking program, a temperature control algorithm is used to control the real-time temperature of the wok; Based on the real-time temperature, the stir-frying state of the wok in the stir-frying mode is controlled in combination with a stir-frying control algorithm; wherein the stir-frying mode includes a manual spatula, a pan flipping, a stirring and a roller stir-frying mode; When the execution completion instruction is obtained, the dishes in the wok are controlled to be taken out of the wok and the wok is cleaned.

2. The intelligent cooking control method according to claim 1, characterized in that: The step of regulating the dish text library and the standard cooking program library corresponding to the dishes within the predetermined range specifically includes: Encode the dishes within the predetermined range into corresponding preset dish texts according to the text encoding rules, and compile them into a dish text library; Compile standard cooking programs for dishes within a predetermined range according to the corresponding conventional cooking processes to form a standard cooking program library; The dish text library is associated with the standardized cooking process library according to the preset dish text, so that the preset dish text and the standard cooking process corresponding to each dish correspond to each other one by one.

3. The intelligent cooking control method according to claim 1, characterized in that: The step of obtaining the target dish text corresponding to the ordering information issued by the restaurant ordering system specifically includes: Scan and receive the ordering information sent by the restaurant's ordering system; Parsing the ordering information to obtain coded text; The coded texts are sorted based on the coding order to obtain target dish texts corresponding to the ordering information.

4. The intelligent cooking control method according to claim 1, characterized in that: The step of matching the target dish text with the preset dish text in the dish text library specifically includes: Input the target dish text into the two-stream self-attention model for vectorization processing; The trained XLNet model is used to extract the vectorized target dish text to obtain the corresponding text vector; Based on the text vector, the preset dish text in the dish text library is matched synchronously, and the preset dish text matching the target dish text is retrieved.

5. The intelligent cooking control method according to claim 1, characterized in that: The step of using a temperature control algorithm to control the real-time temperature of the wok when executing the target cooking program specifically includes: The real-time deviation between the temperature control enabling parameter and the wok temperature is processed by an analog-to-digital conversion circuit to obtain an accurate deviation value; Inputting the accurate value of the deviation into a fuzzy controller to generate a control variable; Fuzzifying the control variables to obtain a subset of real-time deviations; Performing fuzzy decision making on the subset according to the inference rule to determine the output value of the fuzzy controller; Defuzzification and digital-to-analog conversion are performed on the output quantity to obtain a temperature-corrected analog quantity to control the real-time temperature of the wok.

6. The intelligent cooking control method according to claim 1, characterized in that: The step of controlling the stir-frying state of the wok in the stir-frying mode based on the real-time temperature and in combination with the stir-frying control algorithm specifically includes: Selecting a stir-frying mode suitable for the ordering information according to the target cooking program; Adopting adaptive control algorithm to control the speed and position of the wok in the selected stir-fry mode; Using a preset optimization algorithm to adjust the predetermined parameters in the adaptive control algorithm to optimize the speed and position of the wok in the selected stir-fry mode; Adjust the real-time status of frying in the wok according to the optimized speed and position of the wok.

7. The intelligent cooking control method according to claim 6, characterized in that: The preset optimization algorithm is specifically an improved genetic algorithm, which includes parameter encoding, initial population setting, fitness function design, and genetic operator design.

8. An intelligent cooking control system, characterized in that: include: The regulation module regulates the dish text library and standard cooking program library corresponding to the dishes within the predetermined range; The parsing module obtains the corresponding target dish text based on the order information issued by the restaurant ordering system; A matching module, used for matching the target dish text with the preset dish text in the dish text library; A calling module, used for calling the target cooking program corresponding to the target dish text after successful matching from the standard cooking program library; A temperature control module, used to control the real-time temperature of the wok using a temperature control algorithm when executing the target cooking program; A flipping control module, for controlling the flipping state of the wok in the flipping mode based on the real-time temperature and in combination with a flipping control algorithm; wherein the flipping mode includes a manual spatula, flipping the wok, stirring and a roller flipping mode; The overlay control module is used to control the dishes in the wok to be taken out of the wok and clean the wok when the execution is completed.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the intelligent cooking control method according to any one of claims 1 to 7 is implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the intelligent cooking control method as described in any one of claims 1 to 7 is implemented.

Images