Cooking equipment control method and device, cooking equipment and storage medium

Abstract

The invention relates to a cooking equipment control method and device, cooking equipment and a storage medium. The method comprises the following steps: detecting odor tainting conditions of different food materials in the cooking equipment in different cooking stages through a reinforcement learning model, outputting a cooking scheme corresponding to the odor tainting conditions, and performing zoning control according to different cooking areas where the different food materials are located, namely, respectively controlling cooking parameters of the cooking areas where the different food materials are located, therefore, the problem that the odor tainting risk cannot be effectively reduced due to the fact that the cooking equipment cooks different food materials at the same time is solved.

Description

Technical Field

[0001] The present application relates to the field of cooking equipment control, and in particular to a cooking equipment control method, device, cooking equipment and storage medium. Background Art

[0002] With the continuous advancement of science and technology and the improvement of people's living standards, the requirements for the intelligence and refinement of cooking equipment have become crucial. As a cooking device that integrates multiple functions such as steaming, baking, and stewing, cooking equipment plays an increasingly important role in the family. However, when cooking equipment cooks multiple ingredients at the same time, the odor molecules released by different ingredients during the heating process penetrate each other, which often leads to odor problems and affects the taste and quality of the food. Traditional cooking equipment control methods are mostly based on preset cooking programs and fixed parameter settings. It is difficult to adapt to changes in different ingredients and cooking conditions, and therefore cannot effectively reduce the risk of odor cross-contamination. Summary of the invention

[0003] The present application provides a cooking equipment control method, device, cooking equipment and storage medium to solve the problem that the cooking equipment cannot effectively reduce the risk of cross-flavor when cooking different ingredients at the same time.

[0004] In a first aspect, the present application provides a cooking device control method, the method comprising:

[0005] Obtain detection data for different ingredients in cooking equipment;

[0006] Inputting the detection data into a reinforcement learning model, and determining the output of the reinforcement learning model as a cooking plan for the ingredients corresponding to the detection data, wherein the reinforcement learning model is used to determine the odor contamination between different ingredients according to the detection data of different ingredients, and output a cooking plan for reducing the probability of odor contamination;

[0007] The cooking parameters of the cooking area where the ingredients are located are controlled according to the cooking plans corresponding to the different ingredients.

[0008] Optionally, the acquiring detection data of different ingredients in the cooking device includes:

[0009] Obtain environmental data and food data for different food ingredients in the cooking device, wherein the detection data includes the environmental data and the food ingredient data, the environmental data includes temperature, humidity and odor factor distribution area, and the food ingredient data includes food type and food quantity.

[0010] Optionally, controlling the cooking parameters of the cooking area where the ingredients are located according to the cooking schemes corresponding to the different ingredients includes:

[0011] Sending the cooking plans corresponding to the ingredients to the agents corresponding to the cooking areas where the ingredients are located, wherein each agent is used to control the cooking parameters of a cooking area;

[0012] Each of the intelligent agents is utilized to control the cooking parameters of the corresponding cooking area according to the corresponding cooking plan.

[0013] Optionally, using each of the intelligent agents to control cooking parameters of the corresponding cooking area according to the corresponding cooking plan includes:

[0014] Determining the mutually exclusive status between the cooking plans according to the communication results between the intelligent agents;

[0015] When the mutually exclusive state between the cooking schemes is collaborative, each of the intelligent agents is utilized to control the cooking parameters of the corresponding cooking area according to the corresponding cooking scheme.

[0016] Optionally, after determining the mutually exclusive states between the cooking schemes according to the communication results between the agents, the method further comprises:

[0017] When the mutually exclusive states between at least two of the cooking schemes are mutually exclusive, determining the execution priority between the mutually exclusive schemes, wherein the mutually exclusive schemes refer to the cooking schemes whose mutually exclusive states are mutually exclusive;

[0018] According to the execution priorities between the mutually exclusive schemes, the agents corresponding to the mutually exclusive schemes are used in turn to control the cooking parameters of the corresponding cooking areas according to the corresponding mutually exclusive schemes.

[0019] Optionally, before obtaining the detection data of different ingredients in the cooking device, the method further includes:

[0020] Acquire data to be trained, wherein the data to be trained includes a plurality of different food material data and environmental data;

[0021] Preprocessing the data to be trained to obtain preprocessed data;

[0022] Inputting the preprocessed data into an initial learning model, learning the preprocessed data in the initial learning model to determine the cooking progress of different ingredients, and outputting a test plan according to an evaluation result of the cooking progress by a preset reward function, wherein the preset reward function combines multiple reference dimensions, and the multiple reference dimensions include an odor factor distribution area;

[0023] Adjusting the model parameters of the initial learning model according to the cooking effect of the test solution, and iteratively performing the learning and training tasks from acquiring the data to be trained to inputting the preprocessed data into the initial learning model using the updated initial learning model;

[0024] When the initial learning model meets the test requirements, the initial learning model that meets the test requirements is determined as the reinforcement learning model.

[0025] Optionally, after controlling the cooking parameters of the cooking area where the ingredients are located according to the cooking schemes corresponding to the different ingredients, the method further includes:

[0026] When the cooking feedback information is received, the model parameters of the reinforcement learning model are adjusted according to the cooking feedback information and the cooking plans corresponding to the different ingredients.

[0027] In a second aspect, the present application provides a cooking equipment control device, the device comprising:

[0028] An acquisition module is used to acquire detection data of different ingredients in the cooking equipment;

[0029] a processing module, configured to input the detection data into a reinforcement learning model, and determine the output of the reinforcement learning model as a cooking plan for the ingredients corresponding to the detection data, wherein the reinforcement learning model is configured to determine the odor mixing between different ingredients according to the detection data of different ingredients, and output a cooking plan for reducing the probability of odor mixing;

[0030] The control module is used to control the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to the different ingredients.

[0031] In a third aspect, the present application provides a cooking device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned cooking device control method when executing the computer program.

[0032] In a fourth aspect, the present application also provides a computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the above cooking device control method.

[0033] The above-mentioned technical solution provided by the embodiment of the present application has the following advantages over the prior art: the method provided by the embodiment of the present application obtains detection data for different ingredients in the cooking equipment; inputs the detection data into a reinforcement learning model, and determines the output of the reinforcement learning model as a cooking plan for the ingredients corresponding to the detection data, wherein the reinforcement learning model is used to determine the odor contamination between different ingredients based on the detection data of different ingredients, and output a cooking plan for reducing the probability of odor contamination; and controls the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to different ingredients.

[0034] Based on the above method, a reinforcement learning model is used to detect the odor cross-talk of different ingredients in the cooking equipment at different cooking stages, and a cooking plan to deal with the odor cross-talk is output. The cooking areas where different ingredients are located are zoned for control, that is, the cooking parameters of the cooking areas where different ingredients are located are controlled separately to reduce the risk of odor cross-talk between different ingredients, thereby solving the problem that the cooking equipment cannot effectively reduce the risk of odor cross-talk when cooking different ingredients at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0037] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0038] Figure 1 A flowchart of a cooking device control method provided in an embodiment of the present application;

[0039] Figure 2 A structural block diagram of a cooking equipment control device provided in an embodiment of the present application;

[0040] Figure 3 A schematic diagram of the internal structure of a cooking device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0042] The disclosure below provides many different embodiments or examples to implement different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0043] In one embodiment, Figure 1 FIG. 1 is a flow chart of a cooking device control method in an embodiment, referring to FIG. Figure 1 , a cooking device control method is provided. This embodiment mainly uses the method applied to a cooking device control device in a cooking device as an example for explanation. The cooking device may specifically be a steamer, an oven or a steam oven. In this embodiment, the cooking device is taken as an example for explanation. The cooking device control method specifically includes the following steps:

[0044] Step S210, obtaining detection data of different ingredients in the cooking equipment.

[0045] Specifically, the cooking device supports cooking different ingredients at the same time. Different ingredients are placed in different steaming and baking areas in the cooking device and receive different cooking methods. Detection data refers to the data collected from ingredients at different cooking stages in the cooking device. The detection data is used to indicate the cooking progress and cooking effect of the ingredients. The detection data is collected by multiple different sensors set at different positions in the cooking device. The camera in the steam oven collects images of dishes / ingredients placed in the steam oven by the user, and the sensor collects data such as temperature distribution, humidity changes, and physical and chemical changes of different ingredients during the cooking process in real time in the steam oven to understand the cooking characteristics requirements of the ingredients.

[0046] Step S220, input the detection data into a reinforcement learning model, and determine the output of the reinforcement learning model as a cooking plan for the ingredients corresponding to the detection data, wherein the reinforcement learning model is used to determine the odor contamination between different ingredients based on the detection data of different ingredients, and output a cooking plan for reducing the probability of odor contamination.

[0047] Specifically, the reinforcement learning model is used to output a cooking plan for a period of time in the future based on all the detection data of the current cooking stage of the cooking device. In the reinforcement learning model, the odor contamination between different ingredients will be judged based on the input detection data, and a cooking plan for reducing the probability of odor contamination will be output based on the odor contamination situation. That is, different ingredients correspond to different cooking plans, and the cooking plans corresponding to different ingredients are not only used to achieve the cooking purpose of each ingredient, but also to reduce the probability of odor contamination as much as possible during the cooking process. The cooking plan is used to indicate the cooking strategy to be executed by the cooking device, and the cooking parameters involved in the cooking plan include at least one of temperature, humidity, ventilation mode, cooking time, etc.

[0048] Step S230: controlling the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to the different ingredients.

[0049] Specifically, the cooking parameters of the cooking areas where the different ingredients are located are controlled separately according to the cooking plans corresponding to the different ingredients to achieve regional cooking control. While controlling the cooking of different ingredients separately, the probability of tainting of different ingredients can be reduced, thereby solving the problem that the cooking equipment cannot effectively reduce the risk of tainting when cooking different ingredients at the same time. In addition, cooking different ingredients at the same time can also improve cooking efficiency.

[0050] In one embodiment, the step of obtaining detection data of different ingredients in the cooking device includes:

[0051] Obtain environmental data and food data for different food ingredients in the cooking device, wherein the detection data includes the environmental data and the food ingredient data, the environmental data includes temperature, humidity and odor factor distribution area, and the food ingredient data includes food type and food quantity.

[0052] Specifically, environmental data and food data are collected respectively through sensors with different functions arranged at different positions inside the cooking equipment. The environmental data include temperature, humidity, odor factor distribution area, position, volume, gas concentration, ventilation speed, temperature change rate, etc. Among them, the odor factor distribution area is obtained by collecting the odor factors emitted by different ingredients during the cooking process through the odor factor sensor. The odor factor distribution area is used to indicate the distribution of odor factors released by an ingredient during the cooking process. Different ingredients will emit different odor factors under different cooking methods. Therefore, the distribution areas of different odor factors can be used to judge whether there is odor cross-talk between different ingredients and the degree of odor cross-talk.

[0053] For example, beef will release odor factors such as sulfur compounds, aldehydes, and ketone compounds during the steaming and baking process, while the odor factors released by fish during the steaming and baking process mainly include trimethylamine and piperidine. When roasting beef and steaming fish at the same time, the degree of odor cross-talk between the two different ingredients during the cooking process can be determined based on the odor factor distribution area. If there is overlapping area between the odor factor distribution areas corresponding to different ingredients, it means that there is odor cross-talk between the different ingredients. The degree of odor cross-talk is determined specifically according to the size of the overlapping area. That is, the larger the overlapping area, the higher the degree of odor cross-talk, and the smaller the overlapping area, the lower the degree of odor cross-talk.

[0054] The location in the environmental data is used to indicate the placement and occupied area of ​​different ingredients in the cooking equipment. During the steaming and baking process, the position is detected to determine whether the ingredients have been deformed or moved, because some ingredients will expand or shrink during the steaming and baking process. The cooking progress and cooking effect of the ingredients can be determined by observing the degree of deformation. The ingredient data includes the type and quantity of ingredients, and the ingredient data can be collected and identified by the camera module installed inside the cooking equipment.

[0055] In one embodiment, controlling the cooking parameters of the cooking area where the ingredients are located according to the cooking schemes corresponding to the different ingredients includes:

[0056] Sending the cooking plans corresponding to the ingredients to the agents corresponding to the cooking areas where the ingredients are located, wherein each agent is used to control the cooking parameters of a cooking area;

[0057] Each of the intelligent agents is utilized to control the cooking parameters of the corresponding cooking area according to the corresponding cooking plan.

[0058] Specifically, a plurality of intelligent agents are provided in the cooking device, and each intelligent agent is used to control the cooking parameters of one or more cooking areas, that is, different intelligent agents are used to adjust the cooking parameters according to different cooking schemes to realize cooking control in different areas. The cooking scheme corresponding to each ingredient is sent to the intelligent agent corresponding to the cooking area where the ingredient is located, so that each intelligent agent controls the cooking parameters of the corresponding cooking area according to the received cooking scheme, that is, a mapping relationship is established between each cooking area and the intelligent agent, so when the cooking area where the ingredient is located is determined, the intelligent agent used to control the cooking parameters in the cooking area can be determined based on the mapping relationship. The cooking parameters include at least one of temperature, humidity, cooking time, and ventilation mode. The ventilation mode is determined by ventilation speed and ventilation intensity. The ventilation speed is used to indicate the speed of ventilation, and the ventilation intensity is used to indicate the size of the ventilation wind force. That is, the intelligent agent corresponding to each cooking area can control the temperature, humidity, cooking time, ventilation speed and ventilation intensity in the cooking area to adjust the cooking method of the ingredients in the cooking area.

[0059] For example, different suitable temperatures can be set for grilled steak and steamed fish according to the characteristics of the ingredients. The temperature of the inner wall of the cooking equipment box can adjust the corresponding temperature. According to the position where the user places it, its temperature control system sets the temperature value of the four walls of the oven to steam and bake the corresponding ingredients. If it is steamed, its humidity value is guaranteed to be appropriate. At the same time, its various sensors can detect values, so that its cooking equipment can accurately control temperature and humidity. And dynamically adjust the cooking time according to the type of ingredients and the obtained values. Different odor factor sensors are configured in the cooking equipment, which can detect the odor factor values ​​emitted by the corresponding ingredients during the steaming and baking process. It can feedback the precise control system based on the obtained values ​​and adjust its gas flowability to prevent odor contamination. The cooking equipment is usually equipped with a ventilation system inside for circulating air and discharging moisture. During the cooking process, the wind speed and wind direction of the ventilation system can be adjusted to discharge the odors emitted by different ingredients out of the cooking equipment in time to reduce the occurrence of odor contamination.

[0060] When the concentration of the smell of a certain ingredient is detected to be too high, the ventilation volume in that area can be automatically increased to accelerate the dissipation of the smell. At the same time, the ventilation volume in other areas is maintained at an appropriate level to maintain the stability of the cooking environment.

[0061] In one embodiment, using each of the intelligent agents to control cooking parameters of the corresponding cooking area according to the corresponding cooking plan includes:

[0062] Determining the mutually exclusive status between the cooking plans according to the communication results between the intelligent agents;

[0063] When the mutually exclusive state between the cooking schemes is collaborative, each of the intelligent agents is utilized to control the cooking parameters of the corresponding cooking area according to the corresponding cooking scheme.

[0064] Specifically, different intelligent agents can communicate with each other to communicate the cooking plans and detection data they have received, so that different intelligent agents can know the working mode and purpose of other intelligent agents. Based on the communication results between the various intelligent agents, the mutually exclusive state between the various cooking plans can be determined, that is, different cooking plans may occupy the same cooking resources at the same time, but due to the limited cooking resources, resource conflicts occur, that is, two mutually exclusive cooking plans cannot be implemented at the same time. Therefore, before each intelligent agent controls according to the cooking plan, it is necessary to first determine whether there is a logical mutual exclusion between the respective cooking plans, that is, to determine the mutually exclusive state between the various cooking plans.

[0065] If it is determined that the mutually exclusive states between various cooking plans are collaborative, it means that there are no two mutually exclusive cooking plans, and all cooking plans can be implemented at the same time. Therefore, each intelligent agent can control the cooking parameters of the corresponding cooking area according to the cooking plan it receives, so as to achieve simultaneous cooking of different ingredients and reduce the probability of cross-flavoring between different ingredients.

[0066] In one embodiment, after determining the mutually exclusive status between the cooking schemes according to the communication results between the agents, the method further comprises:

[0067] When the mutually exclusive states between at least two of the cooking schemes are mutually exclusive, determining the execution priority between the mutually exclusive schemes, wherein the mutually exclusive schemes refer to the cooking schemes whose mutually exclusive states are mutually exclusive;

[0068] According to the execution priorities between the mutually exclusive schemes, the agents corresponding to the mutually exclusive schemes are used in turn to control the cooking parameters of the corresponding cooking areas according to the corresponding mutually exclusive schemes.

[0069] Specifically, if it is determined that the mutually exclusive status between at least two cooking schemes is mutually exclusive, it means that there is a resource conflict between these mutually exclusive cooking schemes, that is, these cooking schemes cannot be implemented at the same time, then the mutually exclusive cooking schemes are determined as mutually exclusive schemes, and the execution priority between the mutually exclusive schemes is determined. Specifically, the execution priority of the mutually exclusive schemes can be determined based on the preset priority of the intelligent body corresponding to the mutually exclusive schemes; the execution priority of the mutually exclusive schemes can also be determined according to the cooking time in the mutually exclusive schemes. The longer the cooking time, the lower the corresponding execution priority, that is, the cooking scheme with a shorter time is executed first, or the longer the cooking time, the higher the corresponding execution priority, that is, the cooking scheme with a longer time is executed first; the execution priority of the mutually exclusive schemes can also be determined according to the cooking priority of the ingredients corresponding to the mutually exclusive schemes. For example, the cooking priority of desserts is higher than the cooking priority of meat ingredients. Because the cooking requirements of desserts are higher than those of meat ingredients, the cooking scheme corresponding to dessert ingredients needs to be executed first; the execution priority of the mutually exclusive schemes can also be determined according to the execution urgency of the mutually exclusive schemes. The execution urgency is determined by the characteristics of the ingredients and the user's personal taste preferences.

[0070] After determining the execution priorities of different mutually exclusive schemes, the intelligent agents corresponding to the mutually exclusive schemes are used in turn in the order of the determined execution priorities to control the cooking parameters of the corresponding cooking areas according to the mutually exclusive schemes, thereby eliminating the mutual exclusion problem of the mutually exclusive schemes.

[0071] Multiple intelligent agents are set up in the cooking equipment, each of which is responsible for controlling the cooking process of ingredients in one or more cooking areas, realizing independent control of cooking in different areas or ingredients. At the same time, multiple intelligent agents can exchange information through communication protocols to share environmental information and strategy selection to work together, ensuring that the entire cooking equipment reaches the best state during the cooking process.

[0072] For example, Agent 1 (for beef processing): is responsible for monitoring and controlling parameters such as temperature and humidity in the beef cooking area. During the cooking process, Agent 1 adjusts the heating power and cooking time according to the cooking requirements of the beef and the current environmental status. Agent 2 (for grilled fish): is responsible for monitoring and controlling parameters such as temperature and wind direction in the grilled fish cooking area. Since grilled fish usually requires higher temperatures and appropriate wind directions to promote crispy skin and cooked interior, Agent 2 will perform corresponding control operations. Collaborative work: At the beginning of cooking, Agent 1 and Agent 2 share the initial environmental parameters of the cooking equipment and their respective control targets. During the cooking process, Agent 1 and Agent 2 continuously monitor the parameter changes in their respective control areas and share this information in real time through the communication protocol. If it is found that the temperature change in one area has an impact on another area (such as the steam generated by beef cooking affects the temperature of the grilled fish area), they will negotiate to adjust the control strategy to reduce this impact. If an agent finds a resource conflict (such as the allocation problem of the hot air circulation path), it will negotiate with another agent and reach a consensus to ensure the stability and coordination of the internal environment of the entire cooking equipment. After cooking is finished, Agent 1 and Agent 2 will evaluate their cooking effects and feed back the evaluation results to the system. The cooking equipment control device will optimize and improve the entire collaborative workflow based on these feedbacks to improve future cooking efficiency and quality.

[0073] In one embodiment, before obtaining detection data of different ingredients in the cooking device, the method further includes:

[0074] Acquire data to be trained, wherein the data to be trained includes a plurality of different food material data and environmental data;

[0075] Preprocessing the data to be trained to obtain preprocessed data;

[0076] Inputting the preprocessed data into an initial learning model, learning the preprocessed data in the initial learning model to determine the cooking progress of different ingredients, and outputting a test plan according to an evaluation result of the cooking progress by a preset reward function, wherein the preset reward function combines multiple reference dimensions, and the multiple reference dimensions include an odor factor distribution area;

[0077] Adjusting the model parameters of the initial learning model according to the cooking effect of the test solution, and iteratively performing the learning and training tasks from acquiring the data to be trained to inputting the preprocessed data into the initial learning model using the updated initial learning model;

[0078] When the initial learning model meets the test requirements, the initial learning model that meets the test requirements is determined as the reinforcement learning model.

[0079] Specifically, before using the reinforcement learning model, it is necessary to use the data to be trained to perform deep learning training on the initialized initial learning model. The initial learning model can be implemented using PPO (Proximal Policy Optimization) or A3C (Asynchronous Advantage Actor-Critic Algorithm) reinforcement learning algorithms. These algorithms can find the optimal strategy in the continuous action space and adapt to complex environmental changes.

[0080] The preprocessed data is used as input and placed in the initial learning model. At the same time, the food information obtained by the target detection algorithm is also used as the input of the reinforcement learning model. The cooking plan output by the model is used as the current control strategy and output to the cooking equipment control device for execution. During the cooking process, the initial learning model will dynamically adjust the control strategy according to the reward function to optimize the cooking effect. When designing the reward function, it is necessary to consider the distribution area of ​​odor factors, cooking efficiency, food taste, cooking quality evaluation, timely and long-term rewards, and the combination of reward function adjustment and user feedback suggestions.

[0081] The performance of the current cooking strategy (i.e., the cooking plan) is evaluated using the collected data, and the cooking strategy is updated based on the evaluation results. This may involve optimizing the heating curve, adjusting the cooking time, etc., to better adapt to different ingredients and cooking needs. As the cooking equipment is continuously used, the cooking equipment control device will accumulate more experience data and continuously optimize the cooking strategy based on this data. Even in the face of new ingredients or cooking needs, the trained reinforcement learning model can quickly adapt and give a suitable cooking plan. Among them, the deep reinforcement learning model combines the perception ability of deep learning with the decision-making ability of reinforcement learning. During the model building process, the control device uses historical data and experience to train the neural network to learn how to make the best control decision based on the current environmental data and food status. During the operation of the cooking equipment, the cooking equipment control device will continuously receive new environmental data and food status information. This information will be input into the deep reinforcement learning model, and the model will calculate the optimal control strategy based on the current situation. The cooking equipment control device will continue to collect new environmental data and food status information, and compare and evaluate this information with the previous control strategy. If the new data shows that the control strategy is effective, the cooking equipment control device will continue to execute the strategy; if the effect is not good, the cooking equipment control device will recalculate the control strategy based on the new data and make adjustments. By continuously obtaining feedback signals from the environment and learning and optimizing based on these signals, the cooking equipment control device can gradually improve its adaptability and robustness, so as to better cope with various complex situations.

[0082] The optimal strategy ensures that each ingredient is cooked in the most suitable environment by precisely controlling parameters such as temperature and humidity inside the cooking equipment. During cooking, the optimal strategy monitors the state of ingredients and environmental changes in real time and makes dynamic adjustments. If a certain ingredient is found to produce an unpleasant odor or may affect the taste of other ingredients, the cooking equipment control device can immediately take measures to intervene, such as adjusting cooking parameters, increasing ventilation, etc. The optimal strategy search process optimizes multiple objectives (such as cooking efficiency, food taste, avoiding odor contamination, etc.) and finds a balance point, which not only ensures cooking efficiency and taste, but also effectively avoids the problem of odor contamination.

[0083] According to the above content, the initial learning model is trained. During the training, the initial learning model will output actions according to the current state information, receive a reward value, and adjust parameters to optimize performance. After the training is completed, the initial learning model is evaluated and tuned, such as adjusting the network structure, learning rate and other model parameters. The deep reinforcement learning model can analyze the environmental data inside the cooking equipment and the real-time status of the ingredients, such as temperature, humidity, color, etc., in real time, and then judge the cooking progress and effect of the ingredients. At the same time, the cooking equipment control device will evaluate the current cooking strategy according to the preset reward function, such as cooking efficiency, food taste, degree of odor, etc., and select and execute cooking actions by trial and error.

[0084] In one embodiment, after controlling the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to the different ingredients, the method further includes:

[0085] When the cooking feedback information is received, the model parameters of the reinforcement learning model are adjusted according to the cooking feedback information and the cooking plans corresponding to the different ingredients.

[0086] Specifically, the cooking feedback information can be actively triggered by the user, or it can be determined by the cooking device based on the difference between the actual steaming and baking results and the expected steaming and baking results. The cooking feedback information is combined with the cooking plan for learning and the model parameters of the reinforcement learning model are adjusted in reverse. That is, the trained reinforcement learning model is deployed in the cooking device and actual tests are carried out, and suggestions from experienced users are collected to further optimize the reinforcement learning model. As the cooking device is used and data is accumulated, the reinforcement learning model is regularly updated to improve its performance and adaptability.

[0087] In addition, the reinforcement learning model also introduces an online self-learning mechanism to enable the model to continuously learn and optimize to adapt to the optimal cooking environment and conditions of different ingredients. At the same time, it introduces user feedback mechanism and satisfaction data to continuously tune the model learning parameters and strategy selection, improve the robustness and adaptability of the cooking equipment control device, and realize intelligent control.

[0088] Figure 1 FIG. 1 is a flow chart of a cooking device control method in one embodiment. It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0089] In one embodiment, Figure 2 As shown, a cooking equipment control device is provided, comprising:

[0090] An acquisition module 310 is used to acquire detection data of different ingredients in a cooking device;

[0091] A processing module 320 is used to input the detection data into a reinforcement learning model, and determine the output of the reinforcement learning model as a cooking plan for the ingredients corresponding to the detection data, wherein the reinforcement learning model is used to determine the odor mixing between different ingredients according to the detection data of different ingredients, and output a cooking plan for reducing the probability of odor mixing;

[0092] The control module 330 is used to control the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to the different ingredients.

[0093] In one embodiment, the acquisition module 310 is further used for:

[0094] Obtain environmental data and food data for different food ingredients in the cooking device, wherein the detection data includes the environmental data and the food ingredient data, the environmental data includes temperature, humidity and odor factor distribution area, and the food ingredient data includes food type and food quantity.

[0095] In one embodiment, the control module 330 is further configured to:

[0096] Sending the cooking plans corresponding to the ingredients to the agents corresponding to the cooking areas where the ingredients are located, wherein each agent is used to control the cooking parameters of a cooking area;

[0097] Each of the intelligent agents is utilized to control the cooking parameters of the corresponding cooking area according to the corresponding cooking plan.

[0098] In one embodiment, the control module 330 is further configured to:

[0099] Determining the mutually exclusive status between the cooking plans according to the communication results between the intelligent agents;

[0100] When the mutually exclusive state between the cooking schemes is collaborative, each of the intelligent agents is utilized to control the cooking parameters of the corresponding cooking area according to the corresponding cooking scheme.

[0101] In one embodiment, the control module 330 is further configured to:

[0102] When the mutually exclusive states between at least two of the cooking schemes are mutually exclusive, determining the execution priority between the mutually exclusive schemes, wherein the mutually exclusive schemes refer to the cooking schemes whose mutually exclusive states are mutually exclusive;

[0103] According to the execution priorities between the mutually exclusive schemes, the agents corresponding to the mutually exclusive schemes are used in turn to control the cooking parameters of the corresponding cooking areas according to the corresponding mutually exclusive schemes.

[0104] In one embodiment, the control module 330 is further configured to:

[0105] Acquire data to be trained, wherein the data to be trained includes a plurality of different food material data and environmental data;

[0106] Preprocessing the data to be trained to obtain preprocessed data;

[0107] Inputting the preprocessed data into an initial learning model, learning the preprocessed data in the initial learning model to determine the cooking progress of different ingredients, and outputting a test plan according to an evaluation result of the cooking progress by a preset reward function, wherein the preset reward function combines multiple reference dimensions, and the multiple reference dimensions include an odor factor distribution area;

[0108] Adjusting the model parameters of the initial learning model according to the cooking effect of the test solution, and iteratively performing the learning and training tasks from acquiring the data to be trained to inputting the preprocessed data into the initial learning model using the updated initial learning model;

[0109] When the initial learning model meets the test requirements, the initial learning model that meets the test requirements is determined as the reinforcement learning model.

[0110] In one embodiment, the processing module 320 is further configured to:

[0111] When the cooking feedback information is received, the model parameters of the reinforcement learning model are adjusted according to the cooking feedback information and the cooking plans corresponding to the different ingredients.

[0112] like Figure 3As shown, the embodiment of the present application provides a cooking device, including a processor 711, a communication interface 712, a memory 713 and a communication bus 714, wherein the processor 711, the communication interface 712, and the memory 713 communicate with each other through the communication bus 714;

[0113] Memory 713, used for storing computer programs;

[0114] The processor 711 is used to implement the cooking device control method provided by any one of the aforementioned method embodiments when executing the program stored in the memory 713.

[0115] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the scheme of the present application, and does not constitute a limitation on the cooking device to which the scheme of the present application is applied. The specific cooking device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0116] In one embodiment, the cooking device control device provided by the present application can be implemented in the form of a computer program. The computer program can be Figure 3 The cooking device memory may store various program modules constituting the cooking device control device, such as: Figure 2 The acquisition module 310, the processing module 320 and the control module 330 are shown. The computer program composed of various program modules enables the processor to execute the cooking device control method of each embodiment of the present application described in this specification.

[0117] Figure 3 The cooking device shown can be Figure 2 The acquisition module 310 in the cooking device control device shown in the figure acquires the detection data of different ingredients in the cooking device. The cooking device can input the detection data into the reinforcement learning model through the processing module 320, and determine the output of the reinforcement learning model as the cooking plan corresponding to the detection data of the ingredients, wherein the reinforcement learning model is used to determine the odor mixing between different ingredients according to the detection data of different ingredients, and output a cooking plan for reducing the probability of odor mixing. The cooking device can control the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to the different ingredients through the control module 330.

[0118] An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the cooking device control method provided by any of the aforementioned method embodiments is implemented.

[0119] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, or of course by hardware. Based on this understanding, the above technical solution, in essence or in other words, the part that contributes to the relevant technology, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a cooking device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0121] It should be understood that the terms used herein are only for the purpose of describing specific example embodiments and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative may be used.

[0122] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A cooking equipment control method, characterized in that: The method comprises: Obtain detection data for different ingredients in cooking equipment; Inputting the detection data into a reinforcement learning model, and determining the output of the reinforcement learning model as a cooking plan for the ingredients corresponding to the detection data, wherein the reinforcement learning model is used to determine the odor contamination between different ingredients according to the detection data of different ingredients, and output a cooking plan for reducing the probability of odor contamination; The cooking parameters of the cooking area where the ingredients are located are controlled according to the cooking plans corresponding to the different ingredients.

2. The method according to claim 1, characterized in that The obtaining of detection data of different ingredients in the cooking device includes: Obtain environmental data and food data for different food ingredients in the cooking device, wherein the detection data includes the environmental data and the food ingredient data, the environmental data includes temperature, humidity and odor factor distribution area, and the food ingredient data includes food type and food quantity.

3. The method according to claim 1, characterized in that Controlling the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to the different ingredients includes: Sending the cooking plans corresponding to the ingredients to the agents corresponding to the cooking areas where the ingredients are located, wherein each agent is used to control the cooking parameters of a cooking area; Each of the intelligent agents is utilized to control the cooking parameters of the corresponding cooking area according to the corresponding cooking plan.

4. The method according to claim 3, characterized in that Using each of the intelligent agents to control cooking parameters of the corresponding cooking area according to the corresponding cooking plan includes: Determining the mutually exclusive status between the cooking plans according to the communication results between the intelligent agents; When the mutually exclusive state between the cooking schemes is collaborative, each of the intelligent agents is utilized to control the cooking parameters of the corresponding cooking area according to the corresponding cooking scheme.

5. The method according to claim 4, characterized in that After determining the mutually exclusive states between the cooking schemes according to the communication results between the intelligent agents, the method further comprises: When the mutually exclusive states between at least two of the cooking schemes are mutually exclusive, determining the execution priority between the mutually exclusive schemes, wherein the mutually exclusive schemes refer to the cooking schemes whose mutually exclusive states are mutually exclusive; According to the execution priorities between the mutually exclusive schemes, the agents corresponding to the mutually exclusive schemes are used in turn to control the cooking parameters of the corresponding cooking areas according to the corresponding mutually exclusive schemes.

6. The method according to claim 1, characterized in that Before obtaining the detection data of different ingredients in the cooking device, the method further includes: Acquire data to be trained, wherein the data to be trained includes a plurality of different food material data and environmental data; Preprocessing the data to be trained to obtain preprocessed data; Inputting the preprocessed data into an initial learning model, learning the preprocessed data in the initial learning model to determine the cooking progress of different ingredients, and outputting a test plan according to an evaluation result of the cooking progress by a preset reward function, wherein the preset reward function combines multiple reference dimensions, and the multiple reference dimensions include an odor factor distribution area; Adjusting the model parameters of the initial learning model according to the cooking effect of the test solution, and iteratively performing the learning and training tasks from acquiring the data to be trained to inputting the preprocessed data into the initial learning model using the updated initial learning model; When the initial learning model meets the test requirements, the initial learning model that meets the test requirements is determined as the reinforcement learning model.

7. The method according to claim 1, characterized in that After controlling the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to the different ingredients, the method further includes: When the cooking feedback information is received, the model parameters of the reinforcement learning model are adjusted according to the cooking feedback information and the cooking plans corresponding to the different ingredients.

8. A cooking equipment control device, characterized in that: The device comprises: An acquisition module is used to acquire detection data of different ingredients in the cooking equipment; a processing module, configured to input the detection data into a reinforcement learning model, and determine the output of the reinforcement learning model as a cooking plan for the ingredients corresponding to the detection data, wherein the reinforcement learning model is configured to determine the odor mixing between different ingredients according to the detection data of different ingredients, and output a cooking plan for reducing the probability of odor mixing; The control module is used to control the cooking parameters of the cooking area where the ingredients are located according to the cooking plans corresponding to the different ingredients.

9. A cooking 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 method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images