Cooking robot control system based on voice processing

Abstract

The invention discloses a cooking robot control system based on voice processing, and relates to the technical field of intelligent robotics.The cooking robot control system comprises a steady-state locking subsystem, a dynamic detection subsystem, a change analysis subsystem and a feedback subsystem, firstly, a cooking robot sends a rotating speed control instruction to a motor corresponding to a stirring paddle by recognizing a voice command of a user, and the rotating speed control instruction is sent to the motor corresponding to the stirring paddle; the method comprises the following steps: collecting pot wall temperature data in a pot cavity, monitoring operation data of a motor of a stirring paddle in the pot cavity to determine a steady-state control mode, identifying the time for performing centrifugal inrush flow on food in the pot cavity based on the steady-state control mode and the push resistance rate monitored in real time to form centrifugal inrush flow and sauce redistribution operation, and then acquiring the pot wall temperature data in real time to obtain a sauce redistribution result. The temperature difference change before and after the short-time pulse acceleration is executed is analyzed to obtain an execution change temperature difference, an adjustment mechanism is triggered based on the execution change temperature difference, and finally, after the adjustment mechanism is executed, the temperature difference change is continuously fed back until the cooking operation is completed.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent robots, and in particular to a cooking robot control system based on voice processing. Background Art

[0002] In modern households, kitchen appliances, especially cooking appliances, have become an integral part of daily life. With the accelerating pace of life, people are increasingly demanding automation in the kitchen. Traditional cooking methods struggle to meet the demands for efficiency, convenience, and personalization. Smart kitchen devices, particularly cooking robots, have become an ideal choice for improving efficiency in both homes and the restaurant industry.

[0003] Although existing cooking robots can automatically stir-fry ingredients and control the mixing and heat distribution of ingredients through motors, in practice, they have some shortcomings that make it difficult to fully adapt to user needs and kitchen environments. Existing systems often produce uneven cooking results when faced with complex distribution of ingredients and sauces within the pot cavity. In particular, the system often fails to provide timely and effective feedback and adjustments regarding sauce distribution and pot wall temperature balance. Using traditional temperature- or speed-based control systems, cooking robots struggle to precisely control local temperature differences or changes in thrust resistance on the pot wall, resulting in unstable cooking results. Summary of the Invention

[0004] In view of the deficiencies in the prior art, the present invention provides a cooking robot control system based on voice processing, which solves the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a cooking robot control system based on voice processing, comprising: The steady-state locking subsystem recognizes the user's voice commands and issues speed control instructions to the motor corresponding to the stirring paddle. The system also monitors the operating data of the stirring paddle motor in the pot cavity, obtains several sets of thrust-drag ratios, and determines the steady-state control mode based on the difference between the thrust-drag ratios. The dynamic detection subsystem, based on the steady-state control mode and real-time monitoring of the thrust-drag ratio, identifies when to centrifuge the food in the pot cavity. It then drives the motor corresponding to the stirring paddle to perform short-term pulse acceleration to create centrifugal flow and sauce redistribution. The change analysis subsystem collects pot wall temperature data in real time and analyzes the temperature difference before and after the short-time pulse acceleration to obtain the execution change temperature difference. Based on the execution change temperature difference, the adjustment mechanism is triggered. After executing the adjustment mechanism, the feedback subsystem continuously feeds back the temperature difference changes, so that the pot cavity thermal state adjustment and resistance adjustment form an interactive coupling closed loop until the cooking operation is completed.

[0006] Preferably, the steady-state locking subsystem includes a speech recognition module, a motor operation monitoring module and a pattern building module; The voice recognition module: When the cooking robot starts cooking, the voice recognition unit deployed in the controller recognizes the user's voice commands in real time, extracts keywords, and indexes the control parameter configuration file of the corresponding dish according to the voice command, including the target speed range and related threshold settings corresponding to the corresponding dish; The motor operation monitoring module indexes the control parameter configuration file corresponding to the dish based on the voice command, and sends a speed control command to the motor corresponding to the stirring paddle through the controller, so that the motor is driven by a PWM signal with a preset duty cycle, directly entering the target speed range, and monitoring the operation data of the stirring paddle motor in the pot cavity in real time. The operation data includes real-time sampling of the motor phase current and angular velocity; In the mode construction module, the controller calculates the output torque of the agitator in real time based on the known torque constant of the motor, and obtains a set of thrust-drag rate sequences by calculating the ratio of the output torque to the angular velocity within the sampling period. The thrust-drag rate sequence includes the thrust-drag rate corresponding to each sampling point, and monitors the size difference of several groups of thrust-drag rates in the thrust-drag rate sequence to determine the steady-state control mode.

[0007] Preferably, monitoring the difference between the thrust-drag ratios of several groups in the thrust-drag ratio sequence and determining the steady-state control mode includes the following steps: Real-time monitoring of the difference between adjacent thrust-drag rates in the thrust-drag rate sequence; If the difference does not exceed the preset difference threshold, it indicates that the current pot is in steady-state control mode. At this time, the push-resistance rate will be continuously monitored and the push-resistance rate sequence will be updated; If the difference exceeds the preset difference threshold, it indicates that the current pot is not in steady-state control mode. At this time, the push-resistance rate will be continuously monitored and the update of the push-resistance rate sequence will be stopped to form the final updated push-resistance rate sequence; When the steady-state control mode is maintained, the thrust-resistance rate is integrated and averaged to form a steady-state reference thrust-resistance rate, which is used to form a standardized resistance reference value in the initial state of the pot cavity.

[0008] Preferably, the dynamic detection subsystem includes a calculation module and a dynamic execution module; The calculation module, after determining the final updated thrust-drag rate sequence, the cooking robot enters the dynamic detection phase, and obtains the thrust-drag deviation by analyzing and calculating the deviation between the real-time monitored thrust-drag rate and the steady-state reference thrust-drag rate; The dynamic execution module pre-sets the deviation threshold based on the mean-standard deviation method and calculates the absolute value of the deviation threshold. If the push-resistance deviation is greater than 0 and exceeds the deviation threshold, the motor corresponding to the stirring paddle is driven to perform short-time pulse acceleration to form centrifugal flow and sauce redistribution operations; otherwise, the motor corresponding to the stirring paddle is not driven to perform short-time pulse acceleration, and the current steady-state speed stir-frying operation is continued. At the same time, the detection cycle of the next sampling point is entered, and the changes in the internal resistance of the pot cavity are continuously monitored until the push-resistance deviation is detected to exceed the deviation threshold, and then the short-time pulse acceleration is triggered.

[0009] Preferably, the dynamic execution module is further configured to drive the motor corresponding to the stirring paddle to perform short-time pulse acceleration when the thrust-resistance deviation is greater than 0 and exceeds the deviation threshold, so as to form a centrifugal flow and sauce redistribution operation, and the specific steps include: According to the currently monitored thrust-resistance deviation, the amplitude of the pulse acceleration is calculated as follows: ,in, is the amplitude of pulse acceleration, is the proportional gain coefficient, is the deviation threshold, It is the push resistance deviation; When executing short-time pulse acceleration on the motor corresponding to the stirring paddle, the limiting operation is performed, specifically: ,in, is the amplitude of the short-time pulse acceleration of the motor corresponding to the stirring paddle. is the upper limit of the pulse acceleration amplitude; Based on the amplitude of the short-time pulse acceleration executed on the motor corresponding to the stirring paddle, the food in the pot cavity is centrifugally surged to achieve the sauce redistribution operation.

[0010] Preferably, the change analysis subsystem includes a pot temperature change module and a pot temperature comparison module; The pot temperature change module collects pot wall temperature data in real time from multiple temperature sensors arranged around the pot cavity during the process of performing short-time pulse acceleration to form centrifugal flow for sauce redistribution. This module calculates the maximum pot cavity temperature gradient after execution. Specifically: ,in, is the maximum value of the pot cavity temperature gradient after execution, To obtain the maximum value, is the temperature value at the i-th temperature sampling point, is the average temperature of each temperature sampling point, i is the number of the temperature sampling point; The pot temperature comparison module extracts the temperature sampling point corresponding to the maximum value of the pot cavity temperature gradient after execution to obtain the difference between the temperature at the temperature sampling point before the short-time pulse acceleration is executed and the temperature average value at the corresponding moment, marked as the temperature difference value. By comparing the temperature difference value with the maximum value of the pot cavity temperature gradient after execution at the corresponding temperature sampling point, the execution change temperature difference is calculated.

[0011] Preferably, the pot temperature comparison module is also used to compare the executed temperature difference with a preset temperature threshold. If the executed temperature difference exceeds the temperature threshold, it is determined that a local overheating area appears inside the current pot cavity and the adjustment mechanism is triggered. Otherwise, the mechanism is not triggered.

[0012] Preferably, the adjustment mechanism is triggered to perform dynamic temperature balancing adjustment, specifically including: reducing the amplitude of the short-time pulse acceleration performed on the motor corresponding to the stirring paddle, and reducing it to the rotation state when the short-time pulse acceleration is not performed, so as to promote local heat conduction and diffusion of the pot wall.

[0013] Preferably, the feedback subsystem re-collects the pot wall temperature data in real time during the stage of promoting local heat conduction and diffusion of the pot wall to re-obtain the execution change temperature difference and re-judge whether to perform the mechanism trigger processing. If the mechanism trigger processing is not performed, it indicates that the local overheating area in the pot wall has diffused, and the controller immediately re-enters the push-resistance deviation monitoring stage to form centrifugal flow and sauce redistribution operations again until the user participates in the adjustment function to stop the cooking robot operation. The adjustment function includes voice mode and pause button mode.

[0014] The present invention provides a cooking robot control system based on voice processing, which has the following beneficial effects: (1) The steady-state locking subsystem monitors the push-resistance rate in the pot cavity in real time and dynamically adjusts the working mode of the system by comparing the change difference of each push-resistance rate. This enables the cooking robot to maintain a relatively stable control mode in the initial stage. Regardless of the amount of ingredients, sauce concentration or temperature changes in the pot, it can ensure a stable and efficient cooking process. The dynamic detection subsystem accurately identifies the relatively optimal time for centrifugal flow based on the real-time push-resistance rate data and performs short-time pulse acceleration. This process effectively improves the uniform stirring of ingredients, allowing the sauce to be quickly and evenly distributed on the surface of each piece of ingredient, thereby improving the taste and color of the dish and solving the problem of uneven sauce distribution in traditional cooking. However, uneven throwing will occur when stirring or throwing the sauce. At this time, the change analysis subsystem collects the pot wall temperature data in real time, calculates the temperature difference change and triggers the adjustment mechanism based on the temperature difference. This function prepares for the next full stirring of the ingredients, effectively avoiding the problem of local overheating while ensuring that the sauce in the pot wall falls back to the center of the pot body to prepare for the next throwing, thereby optimizing the cooking effect. The feedback subsystem continuously monitors the temperature changes of the pot wall and adjusts the thermal state regulation and resistance regulation in real time to form an interactively coupled closed-loop control mechanism, thereby improving the intelligence and automation level of the cooking robot.

[0015] (2) By continuously monitoring the changes in the thrust-resistance ratio in the pot cavity and performing real-time analysis, the system can automatically adjust the pulse acceleration amplitude, speed and other parameters according to the changes in the ingredients and sauce, and dynamically adjust the pot wall temperature and stirring method based on the feedback mechanism. Ultimately, the system can adaptively adjust the cooking process according to the characteristics of the dish and the current temperature and thrust-resistance status in the pot cavity, ensuring that each dish is completed under relatively good conditions, thereby reducing human intervention.

[0016] (3) Through the calculation module, the system can monitor the deviation between the thrust resistance rate in the pot cavity and the steady-state reference thrust resistance rate in real time. The module uses the thrust resistance deviation to accurately judge the distribution of ingredients and sauces in the pot during the cooking process, thereby dynamically adjusting the acceleration or deceleration strategy to ensure that the ingredients in the pot are always in a relatively good stir-frying state. Based on this real-time feedback mechanism, the system can promptly detect and adjust any cooking deviation caused by factors such as uneven stirring, uneven distribution of ingredients, or uneven pot cavity temperature. When the thrust resistance deviation is detected to exceed the preset threshold, the dynamic execution module accurately calculates the pulse acceleration amplitude. Through this pulse acceleration mechanism, the cooking robot can control the centrifugal flow in the pot cavity to promote uniform stirring of ingredients, especially in the case of uneven sauce coverage, effectively realize sauce redistribution, ensure that the surface of the ingredients is evenly coated with sauce, and thus improve the taste and appearance of the dish.

[0017] (4) Through the pot temperature change module, the system can collect data from multiple temperature sensors around the pot cavity in real time, calculate the maximum value of the pot cavity temperature gradient after execution, and accurately evaluate the thermal state of the pot wall. This process effectively monitors the temperature distribution of the pot cavity, enabling the system to identify and record local overheating areas, thereby providing a basis for further heat regulation. This real-time monitoring and analysis capability improves the response speed and accuracy to the thermal state changes of the pot wall during cooking. The pot temperature comparison module can extract and compare the maximum value of the pot cavity temperature gradient after execution and the temperature difference between the temperature sampling points. By judging whether the temperature difference exceeds the preset temperature threshold, it can intelligently determine whether there is a local overheating area inside the pot cavity. If the temperature difference exceeds the threshold, the system triggers the adjustment mechanism and immediately adjusts the local overheating area of ​​the pot wall to avoid overcooking or burning of the ingredients due to local overheating, while preparing for the next fusion of sauce and ingredients. When the system detects local overheating, it triggers the adjustment mechanism and performs dynamic temperature balance adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a block diagram of the cooking robot control system based on voice processing of the present invention; Figure 2 This is a partial logic diagram of the cooking robot control system based on voice processing of the present invention. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0020] Example 1

[0021] See also Figure 1 and Figure 2 The present invention provides a cooking robot control system based on voice processing, comprising: The steady-state locking subsystem recognizes the user's voice commands and issues speed control instructions to the motor corresponding to the stirring paddle. The system also monitors the operating data of the stirring paddle motor in the pot cavity, obtains several sets of thrust-drag ratios, and determines the steady-state control mode based on the difference between the thrust-drag ratios. The dynamic detection subsystem, based on the steady-state control mode and real-time monitoring of the thrust-drag ratio, identifies when to centrifuge the food in the pot cavity. It then drives the motor corresponding to the stirring paddle to perform short-term pulse acceleration to create centrifugal flow and sauce redistribution. The change analysis subsystem collects pot wall temperature data in real time and analyzes the temperature difference before and after the short-time pulse acceleration to obtain the execution change temperature difference. Based on the execution change temperature difference, the adjustment mechanism is triggered. After executing the adjustment mechanism, the feedback subsystem continuously feeds back the temperature difference changes, so that the pot cavity thermal state adjustment and resistance adjustment form an interactive coupling closed loop until the cooking operation is completed.

[0022] In this embodiment, the system combines voice recognition with a dynamic adjustment mechanism to adjust various parameters in the cooking process in real time according to user instructions, thereby effectively improving the automation and accuracy of cooking. The system realizes comprehensive monitoring and adjustment of ingredients, sauces and temperature in the pot cavity through the collaborative work of multiple subsystems, further improving the quality and efficiency of cooking operations.

[0023] Specifically, the steady-state locking subsystem uses real-time push-to-resistance data to ensure the stirring paddles inside the wok cavity operate in a steady-state mode during cooking, further ensuring stability. The dynamic detection subsystem identifies the dynamic changes in the ingredients and sauce within the wok cavity and triggers short pulse acceleration, creating a centrifugal surge to evenly distribute the sauce, ensuring the ingredients are fully coated and evenly stir-fried. The change analysis subsystem collects real-time wok wall temperature data and analyzes the temperature difference before and after acceleration to promptly determine whether the wok wall is overheating. This triggers a regulation mechanism to prevent localized overheating that could overcook the ingredients while ensuring that the sauce within the wok cavity falls back to the center of the wok, ready for the next drop. After the feedback subsystem executes the regulation mechanism, it continuously provides feedback on the temperature difference, ensuring an effective interactive closed-loop between thermal state regulation and resistance regulation within the wok cavity, ensuring that the ingredients achieve the desired texture and color upon completion.

[0024] When the user specifies a dish, for example, stir-fried shredded potatoes, the control system recognizes the command and loads the appropriate speed parameters (e.g., 20-40 rpm). During this process, the system monitors the thrust-drag ratio within the pot cavity in real time and determines whether it is in steady-state control mode by calculating the difference between adjacent thrust-drag ratios. If the thrust-drag ratio varies minimally, the system is operating stably and the agitator speed remains within the steady-state control range. Based on the steady-state control mode, the dynamic detection subsystem triggers short bursts of acceleration at appropriate times to create centrifugal flow, evenly distributing the sauce and enhancing the coating of the ingredients. For example, when stir-frying sweet and sour pork, sauce accumulates at the bottom of the pot. The system detects the difference in thrust-drag ratio within the pot cavity and determines that a short burst of acceleration is appropriate, causing the agitator to rotate rapidly. This centrifugal force then throws the sauce toward the ingredients, ensuring even coating. The change analysis subsystem monitors the pot wall temperature in real time and determines whether localized overheating is occurring by analyzing the temperature difference before and after acceleration. If overheating is detected, the control mechanism is triggered, and the system monitors the pot wall temperature in real time using a temperature sensor. If it is detected that the temperature difference of the pot wall increases after acceleration (the temperature is too high), the system will trigger the adjustment mechanism to reduce the speed of the stirring paddle to ensure that the sauce in the pot wall falls back to the center of the pot body in preparation for the next throw-off.

[0025] The thrust resistance ratio refers to the mechanical resistance generated by the interaction between the stirring paddle and the ingredients or sauce in the pot cavity. It is expressed by the ratio of the motor's output torque to the angular velocity. It reflects the physical resistance during the stirring process in the pot cavity and is used to judge the state changes of the ingredients in the pot cavity.

[0026] Short-term pulse acceleration refers to applying a short period of high speed operation to the stirring paddle motor at a specific moment, which is used to accelerate the mixing of ingredients, distribution of sauces, etc. This process helps to evenly distribute the ingredients and sauces in the pot by accelerating the centrifugal force.

[0027] Temperature difference refers to the temperature change in different areas of the pot wall before and after pulse acceleration. The temperature difference can reflect the uniformity of heat distribution in the pot cavity. If the temperature difference in a certain area changes too much, it means that the area may be locally overheated, and there may be uneven mixing or throwing of ingredients or sauces, which requires adjustment.

[0028] The feedback closed loop means that during the entire cooking process, the system monitors various data in the pot cavity (such as temperature and thrust resistance) in real time, continuously adjusts the control parameters, and forms a continuous adjustment feedback loop to ensure the quality and stability of the cooking operation.

[0029] Through the coordinated operation of four subsystems, this invention ensures that every operation in the cooking process can be controlled, enabling the cooking robot to provide a more personalized and stable cooking experience based on different ingredients and user needs. Each subsystem provides feedback and adjustments from different dimensions, such as temperature, thrust resistance, and speed, to ensure stable and improved dish quality.

[0030] Example 2

[0031] Please refer to Figure 1 ,Specifically: the steady-state locking subsystem includes a speech recognition module, a motor ,operation monitoring module and a pattern building module; The voice recognition module: When the cooking robot begins the cooking process, the voice recognition unit deployed in the controller recognizes the user's voice command in real time, extracts keywords in the voice command, such as "stir-fry shredded potatoes," "quick stir-fry," and "slow stir-fry," and then indexes the control parameter configuration file for the corresponding dish based on the voice command, including the target speed range and related threshold settings for the corresponding dish; First, acoustic feature extraction is used to convert the user's voice (analog sound waves) into a digital feature stream. These features are then mapped into a phoneme sequence using an acoustic model (typically an HMM-GMM or a deep neural network (DNN). A language model (N-Gram or RNN) is then used to infer the phoneme / syllable sequence into a more meaningful word string to recognize the voice command. Finally, semantic understanding (SLU) is used to extract the key words in the voice command. Among them, relevant threshold settings include gap threshold, deviation threshold and temperature threshold; The control parameter configuration file refers to a set of specific control parameters pre-saved in the cooking robot's control system (controller MCU or mainboard memory) for different dishes and different taste requirements. Each dish has a control parameter description table; The motor operation monitoring module uses voice commands to index the control parameter configuration file corresponding to the dish. It then issues a speed control command to the motor corresponding to the stirring paddle through the controller, driving the motor with a PWM (pulse width modulation) signal with a preset duty cycle, directly entering the target speed range (e.g., 20-40 rpm) to maintain the subsequent steady-state control mode. It also monitors the operating data of the stirring paddle motor in the pot cavity in real time, including real-time sampling of the motor's phase current and angular velocity. The so-called preset duty cycle PWM drive means that when the controller is just started, it does not start from 0 and gradually increase, but directly issues a fixed value PWM drive signal. For example, the controller directly outputs a 50% duty cycle PWM output, so that the motor immediately receives an average drive equivalent to half the voltage. This is driving the motor with a preset duty cycle PWM signal, which is used to allow the motor to directly reach a desired steady-state speed, rather than slowly rising from a very low speed. By using PWM with a preset duty cycle, static friction can be broken through immediately, the target speed can be quickly reached, and local sticking of the pan due to long-term friction heating can be avoided.

[0032] In the mode construction module, the controller calculates the output torque of the agitator in real time based on the known torque constant of the motor, and obtains a set of thrust-drag rate sequences by calculating the ratio of the output torque to the angular velocity within the sampling period. The thrust-drag rate sequence includes the thrust-drag rate corresponding to each sampling point, and monitors the size difference of several groups of thrust-drag rates in the thrust-drag rate sequence to determine the steady-state control mode.

[0033] The specific method of obtaining the output torque of the stirring paddle is as follows: ,in, is the output torque, is a known torque constant, is the phase current; The known torque constant refers to the theoretical torque that the motor can output when the unit current flows through the winding. The unit is Nm / A, for example =0.05 Nm / A, which means that when the motor is supplied with 1A current, it will output a torque of 0.05Nm. The known torque constant can be obtained from the motor manufacturer's data sheet; Output torque is the torque that can be output at the shaft end of the motor when it rotates. The greater the torque, the greater the density of the food and the more local accumulation. The push-resistance rate sequence includes the push-resistance rate corresponding to each sampling point, where the push-resistance rate is obtained by the following formula: ,in, is the thrust resistance, is the angular velocity; The thrust-drag ratio refers to the torque that the stirring paddle needs to apply per unit angular velocity. It can be understood as follows: if the ingredients in the pot are very thin (for example, only a small amount of oil), the faster the rotation, the smaller the torque required. If the ingredients in the pot are very thick and heavy, the faster the stirring paddle rotates, the greater the torque required. Therefore, the larger the thrust-drag ratio, the more torque needs to be applied while maintaining a certain angular velocity, and the greater the resistance in the pot (viscosity, density, local accumulation). This is like the concept of "viscous resistance" in viscous fluids, except that this is replaced with a rotating scenario. Output torque refers to the torque that can be output at the shaft end of the motor when it rotates; The controller refers to the central control unit (main control unit) inside the cooking robot that is used to coordinate the overall drive, detection, data processing and execution of control strategies.

[0034] The stirring paddle (or stirring rod) is the actuator located inside the pot cavity, which is in direct contact with the ingredients and sauce.

[0035] The motor that drives the stirring paddle (usually a brushless DC motor or a stepper motor) is the source of output power. When the resistance of the food increases, the stirring paddle becomes more difficult to rotate, and the motor needs to provide a larger current. Therefore, the actual output torque can be calculated by detecting its current, voltage or back electromotive force, and the thrust-resistance ratio can be indirectly obtained.

[0036] Monitoring the difference between several groups of thrust-drag rates within the thrust-drag rate sequence and determining the steady-state control mode includes the following steps: Real-time monitoring of the difference between adjacent thrust-drag rates in the thrust-drag rate sequence; The size gap refers to the difference calculation between adjacent thrust resistance rates; If the difference does not exceed the preset difference threshold, it indicates that the current pot is in steady-state control mode. At this time, the push-resistance rate will be continuously monitored and the push-resistance rate sequence will be updated; Steady-state control mode refers to the mechanical system driven by the motor corresponding to the stirring paddle in the pot cavity during the operation of the cooking robot. After preliminary adjustment (for example, driving it to the target speed range with a PWM signal with a preset duty cycle), and in the continuous sequence of real-time monitoring of the thrust-resistance rate, the difference between adjacent thrust-resistance rates does not exceed the preset gap threshold, thereby determining that the process of stirring ingredients and stir-frying sauces in the pot cavity is in a relatively stable and uniform working state. This mode is used as a reference benchmark for subsequent thrust-resistance deviation detection, short-time pulse acceleration and temperature gradient adjustment to ensure the accuracy of the entire cooking control.

[0037] If the difference exceeds the preset difference threshold, it indicates that the current pot is not in steady-state control mode. At this time, the push-resistance rate will be continuously monitored and the update of the push-resistance rate sequence will be stopped to form the final updated push-resistance rate sequence; The final updated thrust-drag rate sequence refers to the thrust-drag rate at each sampling point corresponding to the time when the difference does not exceed the preset difference threshold, and the time period for maintaining the steady-state control mode is determined based on the final updated thrust-drag rate sequence; When the steady-state control mode is maintained, the thrust-resistance rate is integrated and averaged to form a steady-state reference thrust-resistance rate, which is used to form a standardized resistance reference value in the initial state of the pot cavity.

[0038] The steady-state reference thrust resistance rate is obtained by the following formula: ,in, is the steady-state reference thrust resistance rate, T is the sampling period, is the first sampling point.

[0039] By quantifying the steady-state baseline thrust resistance rate, it is easy to compare the thrust resistance rate detected each time with this benchmark to determine whether there is new accumulation or sauce accumulation in the pot, so as to trigger short-cycle pulse acceleration to form centrifugal tossing (simulating manual tossing). In this embodiment, the voice recognition module can automatically index the corresponding dish control parameter configuration file by real-time recognition of user voice commands, such as "stir-fry shredded potatoes," and load it into the system. This allows users to complete dish selection through simple voice commands, further enhancing the personalized and convenient user experience. When the user says "quickly stir-fry shredded potatoes," the voice recognition module captures the two keywords "quickly stir-fry" and "shredded potatoes" and then loads relevant parameters from a pre-set configuration file, such as the target speed (30 rpm) and the corresponding heat control. Stir-fry refers to the heat type in the voice command, and the control system loads the corresponding control parameters accordingly.

[0040] The motor operation monitoring module ensures stable rotation of the stirring paddle by monitoring the motor's phase current and angular velocity in real time. It also calculates the thrust-to-drag ratio, enabling the system to dynamically adjust the speed based on the physical resistance of the ingredients in the pot. This module ensures stable operation of the stirring paddle within the target speed range, avoiding overload or low speed conditions. For example, when cooking, the ingredients in the pot are relatively hard, requiring a higher speed to overcome the physical resistance. The motor operation monitoring module collects current and angular velocity in real time and calculates the thrust-to-drag ratio. If the thrust-to-drag ratio indicates that the ingredients in the pot are too compact, the system automatically increases the speed to maintain a stable stirring effect.

[0041] The thrust-drag ratio measures the firmness of the ingredients within the pot cavity and helps the system determine whether speed adjustment is necessary. The pattern building module analyzes the ratio of motor output torque to angular velocity, calculates a thrust-drag ratio sequence, and compares the difference between adjacent thrust-drag ratios to determine whether the pot cavity is in steady-state control mode. If the thrust-drag ratio difference is less than a preset threshold, the ingredients in the pot are evenly mixed and in steady-state control mode. Otherwise, the system automatically adjusts the mixing parameters, entering the dynamic adjustment phase.

[0042] Assuming that the ingredients in the pot are stir-fried relatively evenly, and the system detects in real time that the difference in the thrust resistance sequence does not exceed the set threshold (such as 0.02), the system will consider that the pot cavity is in steady-state mode and continue to maintain the current speed without making further adjustments.

[0043] The difference in push resistance is used to judge the uniformity of ingredients in the pot cavity and the stability of stirring.

[0044] The dynamic detection subsystem uses thrust-resistance deviation and preset thresholds to determine when to perform short-term pulse acceleration. This mechanism performs short-term acceleration when the sauce or ingredients in the pot cavity are unevenly distributed to ensure even distribution of the sauce.

[0045] Assuming that part of the food in the pot is adhered to the pot wall, the system calculates that the push-resistance deviation is 0.1 and exceeds the threshold. The system will perform short-time pulse acceleration and use centrifugal force to make the food and sauce on the pot wall return to the pot, thereby achieving redistribution of the sauce.

[0046] The thrust-resistance deviation is an indicator for determining whether acceleration is needed. Short-term pulse acceleration is used to speed up the redistribution of ingredients to avoid localized retention of sauce, which can cause the sauce to be delayed in blending with other ingredients. The feedback subsystem forms an interactively coupled closed-loop control by monitoring the pot wall temperature and thrust resistance, ensuring the uniformity of heat distribution in the pot cavity and the stirring of ingredients, further avoiding local overheating or uneven stir-frying of ingredients. The system can automatically adjust the temperature and speed to ensure that the ingredients are cooked in the best condition.

[0047] Assuming that after a short pulse acceleration, the temperature difference of the pot wall is large, the feedback subsystem detects local overheating in the pot cavity through the temperature difference change and triggers the regulation mechanism, such as adjusting the speed of the stirring paddle to promote uniform heat diffusion.

[0048] The temperature change of the pot wall is the key indicator of the feedback mechanism, and the feedback loop ensures that the heat and pushing resistance are always balanced.

[0049] When the stirring paddle rotates in the pot, the resistance it needs to overcome mainly depends on the overall density, viscosity and local accumulation of the ingredients in the pot. At the beginning (all the ingredients have not yet been significantly softened by heat and the sauce consistency remains unchanged), the average thrust resistance rate measured by the system is a characteristic quantity reflecting the initial physical state of the pot. It is equivalent to quantifying the baseline of the resistance field in the pot before obvious phase changes (such as protein coagulation, starch gelatinization) or significant movement of the liquid phase occur during stir-frying. This allows for subsequent judgment of the resistance changes caused by local sauce accumulation or ingredient movement in the pot, thereby determining whether transient pulse centrifugation should be triggered.

[0050] During the continuous stir-frying process, the resistance in the wok will increase for a short period of time as the ingredients gradually slide and accumulate, or as the sauce concentrates locally. When this push-resistance deviation is greater than 0 and exceeds the deviation threshold, it indicates that localized aggregation or accumulation may have occurred in the wok. This is similar to a chef feeling that the wok is becoming heavy. At this time, a short pulse acceleration is needed to achieve a full fusion of the ingredients and the sauce. This is especially important for dishes such as sweet and sour spare ribs, where timely fusion of the ingredients and the sauce is necessary to prevent them from sticking to the wok. Example 3 Please refer to Figure 1 ,Specifically: the dynamic detection subsystem includes a calculation module and a dynamic ,execution module; The calculation module, after determining the final updated thrust-drag rate sequence, the cooking robot enters the dynamic detection phase, and obtains the thrust-drag deviation by analyzing and calculating the deviation between the real-time monitored thrust-drag rate and the steady-state reference thrust-drag rate; The thrust deviation is obtained as follows: ,in, It is the push resistance deviation; The dynamic execution module pre-sets the deviation threshold based on the mean-standard deviation method and calculates the absolute value of the deviation threshold. If the thrust-resistance deviation is greater than 0 and exceeds the deviation threshold, the motor corresponding to the stirring paddle is driven to perform short-time pulse acceleration to form centrifugal flow and sauce redistribution operations; otherwise, the motor corresponding to the stirring paddle is not driven to perform short-time pulse acceleration, and the current steady-state speed stir-frying operation is continued. At the same time, the detection cycle of the next sampling point is entered, and the change of the internal resistance of the pot cavity is continuously monitored until the thrust-resistance deviation is detected to exceed the deviation threshold, and then the short-time pulse acceleration is triggered; among them, the change of the internal resistance of the pot cavity here refers to the change of the difference between the thrust-resistance rate in the pot cavity and the steady-state reference thrust-resistance rate (that is, the thrust-resistance deviation).

[0051] The timing for identifying centrifugal flow of food in the pot cavity in the dynamic detection subsystem refers to the moment when the thrust resistance deviation is greater than 0 and exceeds the deviation threshold. By determining the timing for centrifugal flow, the stirring paddle can be driven in time for rapid stirring.

[0052] The formation of centrifugal flow and sauce redistribution is a manifestation of timely driving the stirring paddle for rapid stirring; When the thrust-drag deviation is greater than 0 and exceeds the deviation threshold, it may be due to local accumulation (for example, meat slices or starch sauce suddenly gathered near the paddle), which will make the motor rotation instantaneously heavier and the torque increased, or the sauce suddenly concentrated at the bottom of the pot, increasing the viscosity resistance and causing the thrust-drag rate to increase instantly.

[0053] The dynamic execution module is also used to drive the motor corresponding to the stirring paddle to perform short-term pulse acceleration when the thrust-resistance deviation is greater than 0 and exceeds the deviation threshold, so as to form a centrifugal flow and sauce redistribution operation. The specific steps include: According to the currently monitored thrust-resistance deviation, the amplitude of the pulse acceleration is calculated as follows: ,in, is the amplitude of pulse acceleration, is the proportional gain coefficient, is the deviation threshold; When executing short-time pulse acceleration on the motor corresponding to the stirring paddle, the limiting operation is performed, specifically: ,in, is the amplitude of the short-time pulse acceleration of the motor corresponding to the stirring paddle. is the upper limit of the pulse acceleration amplitude, that is, the maximum allowable increment of the pulse amplitude; Based on the amplitude of the short-duration pulse acceleration applied to the motor corresponding to the stirring paddle, a centrifugal flow is generated within the pot cavity, redistributing the sauce. Under real-time thrust-resistance deviation detection based on the steady-state reference thrust-resistance ratio, short-duration pulse acceleration is applied to drive centrifugal distribution of the food and sauce within the pot cavity, thereby forming a dynamic coating.

[0054] The purpose of the dynamic execution module is to achieve even adhesion of sauces to dishes (such as sweet and sour dishes, Kung Pao dishes, and stir-fried dishes) through high-speed tossing. Centrifugal flow refers to when the stirring paddle rotates at a certain speed in the pot cavity, or when the stirring paddle is accelerated to a higher speed in a short pulse, the sauce and oil-water mixture in the pot cavity is thrown from the center to the pot wall under the action of centrifugal force, forming a radial flow from the center of the pot to the pot wall, and then falls back to the surface of the food due to gravity, achieving uniform coating of the sauce.

[0055] In this embodiment, the calculation module monitors the deviation between the thrust-resistance ratio within the pot cavity and a steady-state baseline thrust-resistance ratio in real time. Based on this deviation, it determines whether to trigger short-duration pulse acceleration to ensure even mixing and distribution of ingredients and sauce. The dynamic execution module automatically applies short-duration pulse acceleration based on this thrust-resistance deviation, significantly improving stir-frying uniformity and reducing the risk of burning or undercooking ingredients due to uneven heating of the pot wall.

[0056] The dynamic execution module uses a mean-standard deviation method to preset and calculate deviation thresholds, ensuring that pulse acceleration is triggered only when the thrust-resistance deviation exceeds the set threshold. This avoids unnecessary over-adjustments and ensures that each pulse acceleration is triggered precisely, achieving precise control. During the dynamic detection phase, the system generates a centrifugal surge through short-duration pulse acceleration, helping to evenly distribute the sauce across each ingredient, enhancing the taste and color of the dish. This dynamic surge action and precise control ensure standardization and consistency across each dish, resolving the issue of uneven sauce distribution common in traditional cooking. The system continuously monitors changes in thrust-resistance ratio throughout each cooking cycle and dynamically adjusts the amplitude of the short-duration pulse acceleration to ensure a relatively optimal distribution of ingredients and sauce within the pot cavity. Through this adaptive feedback control, the cooking robot automatically adjusts its cooking strategy based on the condition of the ingredients and temperature changes within the pot cavity, further enhancing the integration of the sauce and ingredients.

[0057] The calculation module monitors the thrust resistance rate in the pot cavity in real time, and compares it with the steady-state reference thrust resistance rate to calculate the thrust resistance deviation. For example, if the system detects that the thrust resistance rate in the pot cavity is =1.2 and steady-state reference thrust resistance =1.0, then the thrust resistance deviation =1.2-1.0=0.2. The thrust-resistance deviation reflects the change in friction between the food in the pot cavity and the stirring paddle, indicating that the food may be concentrated in a certain location or there is a high amount of sauce attached to the pot wall. The system needs to adjust the stirring strategy.

[0058] When the thrust-resistance deviation exceeds the set threshold, the dynamic execution module calculates the pulse acceleration amplitude according to the deviation and drives the stirring paddle to perform short-time pulse acceleration. For example, when the thrust-resistance deviation exceeds the set threshold, the dynamic execution module calculates the pulse acceleration amplitude according to the deviation and drives the stirring paddle to perform short-time pulse acceleration. If the value exceeds the set threshold of 0.1, a short-time pulse acceleration is triggered. The module calculates the pulse amplitude that needs to be increased, assuming the proportional gain coefficient =0.8, then the pulse acceleration amplitude is: Pulse acceleration amplitude = 0.8×0.1=0.8, which means that the system will adjust the speed or duty cycle of the stirring paddle and perform short-time pulse acceleration to evenly distribute the ingredients and sauce.

[0059] When the sauce in the pot cavity is concentrated in the center of the pot and cannot evenly wrap the ingredients, the system generates centrifugal force through pulse acceleration to redistribute the sauce to the surface of the ingredients, so that each piece of food can be evenly coated with the sauce, thereby improving the taste and quality of the dish.

[0060] The pulse acceleration amplitude is an incremental adjustment to the stirring paddle motor speed, which is used to quickly break up ingredients and sauces to promote even stir-frying in the pot.

[0061] The proportional gain coefficient determines how much the corresponding pulse acceleration will increase for each unit increase in the deviation. It is usually obtained through experimental debugging or control parameter adjustment. The method is: use different Perform response curve test, for example: It will cause slow response and failure to quickly break up the accumulation. It will cause overshoot and easy splashing, until the pot cavity is adjusted to quickly form centrifugal throwing but not excessively disturbed, which is a relatively optimal state.

[0062] Example 4

[0063] Please refer to Figure 1 ,Specifically: the change analysis subsystem includes a pot temperature change module and a pot temperature ,comparison module; The pot temperature change module collects pot wall temperature data in real time from multiple temperature sensors arranged around the pot cavity during the process of performing short-time pulse acceleration to form centrifugal flow for sauce redistribution. This module calculates the maximum pot cavity temperature gradient after execution. Specifically: ,in, is the maximum value of the pot cavity temperature gradient after execution, To obtain the maximum value, is the temperature value at the i-th temperature sampling point, is the average temperature of each temperature sampling point, i is the number of the temperature sampling point; The pot temperature comparison module extracts the temperature sampling point corresponding to the maximum value of the pot cavity temperature gradient after execution to obtain the difference between the temperature at the temperature sampling point before the short-time pulse acceleration is executed and the temperature average value at the corresponding moment, marked as the temperature difference value. By comparing the temperature difference value with the maximum value of the pot cavity temperature gradient after execution at the corresponding temperature sampling point, the execution change temperature difference is calculated.

[0064] The pot temperature comparison module is also used to compare the executed temperature difference with the preset temperature threshold. If the executed temperature difference exceeds the temperature threshold, it is determined that a local overheating area has occurred inside the pot cavity and the adjustment mechanism is triggered. Otherwise, the mechanism is not triggered. The adjustment mechanism is used to adjust the amplitude of the short-time pulse acceleration of the motor corresponding to the stirring paddle. The adjustment mechanism is triggered to perform dynamic temperature balancing adjustment, specifically including: reducing the amplitude of the short-time pulse acceleration of the motor corresponding to the stirring paddle, and reducing it to the rotation state when the short-time pulse acceleration is not performed, so as to promote the local heat conduction and diffusion of the pot wall.

[0065] The rotation state here includes the speed, drive duty cycle, instantaneous output torque, etc., which refers to the steady-state speed of the motor corresponding to the stirring paddle; In this embodiment, the system effectively achieves temperature balance within the pot cavity through the real-time feedback mechanism and temperature control of the variation analysis subsystem, thereby improving the automation and intelligence level of the cooking robot. Specifically, this system provides the following beneficial effects: Through the pot temperature variation module, the system can collect temperature data from multiple points on the pot wall in real time, calculate and extract the maximum temperature gradient within the pot cavity after execution. This process accurately tracks thermal changes on the pot wall and promptly identifies localized overheating areas within the pot cavity. For example, when stir-frying Kung Pao Chicken, if the system detects a high temperature in a certain area of ​​the pot wall, it triggers the corresponding thermal regulation mechanism. This prevents localized charring or overcooking of the ingredients while further preparing the sauce for even coating of the ingredients.

[0066] The maximum pot cavity temperature gradient refers to the maximum temperature difference between sampling points on the pot wall within the pot cavity. It measures the degree of temperature nonuniformity within the pot wall area. The pot temperature comparison module analyzes the temperature difference after the pot wall is executed to accurately determine whether there are localized overheating areas. If the temperature difference exceeds the set threshold, the system triggers a regulation mechanism to automatically reduce the amplitude of the short-term pulse acceleration. This prevents the ingredients from being overcooked due to excessive pot wall temperature. It also allows time for the sauce to return to the center of the pot, facilitating the subsequent rapid blending of the sauce and ingredients.

[0067] The temperature difference is determined by comparing the temperature difference before and after the short pulse acceleration to determine whether temperature adjustment is needed. After the adjustment mechanism is triggered, the system will adjust the amplitude of the pulse acceleration and limit it according to the temperature change of the pot wall. For example, when the temperature of the pot wall rises to the overheating zone during frying, the system will reduce the amplitude of the pulse acceleration and return to the original steady-state speed, thereby promoting the conduction and diffusion of heat in the local area of ​​the pot wall to avoid overheating. The rotation state refers to the stable speed state of the stirring paddle, that is, the speed when the motor runs at a preset duty cycle, ensuring smooth stirring of the ingredients in the pot without pulse acceleration.

[0068] Through dynamic temperature balancing and continuous feedback, the system effectively avoids localized overheating during short-duration pulse acceleration, promoting heat conduction and diffusion through the pot wall, ensuring temperature uniformity within the pot cavity. Once the pot wall temperature difference returns to normal, the system automatically enters stable operation mode and resumes push-resistance deviation monitoring. This process not only optimizes the stir-frying effect but also ensures consistent taste and color of the dish.

[0069] Example 5

[0070] Please refer to Figure 1 Specifically: the feedback subsystem, in the stage of promoting local heat conduction and diffusion of the pot wall, re-collects the pot wall temperature data in real time to re-obtain the execution change temperature difference, and re-judges whether to trigger the mechanism. If the mechanism is not triggered, it indicates that the thermal gradient at the pot cavity phase temperature sampling point has fallen back to within the threshold, indicating that the local overheating area in the pot wall has diffused, and the controller immediately re-enters the push-resistance deviation monitoring stage to form centrifugal flow and sauce redistribution operations again, so that the pot cavity thermal state adjustment and resistance adjustment form an interactive coupling closed loop, until the user participates in the adjustment function to stop the cooking robot operation. The adjustment function includes voice mode and press the pause button mode.

[0071] After re-entering the push-resistance deviation monitoring stage, if it is subsequently detected that the push-resistance deviation exceeds the deviation threshold, the pulse acceleration amplitude is calculated again and short-time pulse acceleration is performed. At the same time, the pot cavity temperature gradient is collected again to determine whether it is necessary to enter the short-time deceleration thermal diffusion stage, thereby realizing multi-physical field cyclic closed-loop control of dynamic adjustment of mechanical resistance inside the pot cavity and dynamic diffusion of local temperature, and solving the problem of deterioration in color and taste of food caused by local overcooking.

[0072] In this embodiment, during the local heat conduction and diffusion phase of the pot wall, the feedback subsystem recollects pot wall temperature data in real time, enabling accurate calculation of the changing temperature difference. The core function of this process is to promptly detect the temperature uniformity within the pot cavity. If the temperature difference exceeds a set threshold, the system triggers a further adjustment mechanism to prevent the local temperature of the pot wall from being too high, thereby preventing the ingredients from being overcooked. For example, assuming a user commands stir-frying Kung Pao Chicken, the system will monitor the pot wall temperature in real time using multiple temperature sensors. If the system detects a large temperature difference on the pot wall (for example, if the temperature in a certain area is too high), it triggers an adjustment mechanism through the feedback subsystem to lower the temperature in that area to ensure even heating of the ingredients. Based on the even heat diffusion, the feedback subsystem re-enters the push-resistance deviation monitoring phase, ensuring optimal stirring and stir-frying of the ingredients within the pot by real-time monitoring of the push-resistance ratio within the pot cavity.

[0073] For example, when stir-frying Kung Pao Chicken, the thrust-resistance deviation within the pot cavity may fluctuate. In this case, the system immediately activates a short-duration pulse acceleration function to speed up the stir-frying of the ingredients and ensure even coating of the sauce. However, since this process may cause uneven stirring or sauce shedding, after executing the pulse acceleration, the feedback subsystem monitors the pot temperature to determine whether to enter the short-duration deceleration heat diffusion phase. If the temperature is detected to be locally too high, the system automatically enters the short-duration deceleration phase, reducing the stirring intensity and allowing the heat from the pot wall to gradually diffuse throughout the pot cavity, preventing local overheating that may degrade the taste and color of the ingredients.

[0074] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A cooking robot control system based on speech processing, characterized by: include, The steady-state locking subsystem recognizes the user's voice commands and issues speed control instructions to the motor corresponding to the stirring paddle. The system also monitors the operating data of the stirring paddle motor in the pot cavity, obtains several sets of thrust-drag ratios, and determines the steady-state control mode based on the difference between the thrust-drag ratios. The dynamic detection subsystem, based on the steady-state control mode and real-time monitoring of the thrust-drag ratio, identifies when to centrifuge the food in the pot cavity. It then drives the motor corresponding to the stirring paddle to perform short-term pulse acceleration to create centrifugal flow and sauce redistribution. The change analysis subsystem collects pot wall temperature data in real time and analyzes the temperature difference before and after the short-time pulse acceleration is executed to obtain the execution change temperature difference. Based on the execution change temperature difference, the adjustment mechanism is triggered. The execution change temperature difference is used to reflect the temperature difference before and after the short-time pulse acceleration is executed. After executing the adjustment mechanism, the feedback subsystem continuously feeds back the temperature difference changes, so that the pot cavity thermal state adjustment and resistance adjustment form an interactive coupling closed loop until the cooking operation is completed.

2. The cooking robot control system based on voice processing according to claim 1 is characterized in that: The steady-state locking subsystem includes a speech recognition module, a motor operation monitoring module, and a pattern building module; The voice recognition module: When the cooking robot starts cooking, the voice recognition unit deployed in the controller recognizes the user's voice commands in real time, extracts keywords, and indexes the control parameter configuration file of the corresponding dish according to the voice command, including the target speed range and related threshold settings corresponding to the corresponding dish; The motor operation monitoring module indexes the control parameter configuration file corresponding to the dish based on the voice command, and sends a speed control command to the motor corresponding to the stirring paddle through the controller, so that the motor is driven by a PWM signal with a preset duty cycle, directly entering the target speed range, and monitoring the operation data of the stirring paddle motor in the pot cavity in real time. The operation data includes real-time sampling of the motor phase current and angular velocity; In the mode construction module, the controller calculates the output torque of the agitator in real time based on the known torque constant of the motor, and obtains a set of thrust-drag rate sequences by calculating the ratio of the output torque to the angular velocity within the sampling period. The thrust-drag rate sequence includes the thrust-drag rate corresponding to each sampling point, and monitors the size difference of several groups of thrust-drag rates in the thrust-drag rate sequence to determine the steady-state control mode.

3. The cooking robot control system based on voice processing according to claim 2 is characterized in that: Monitoring the difference between several groups of thrust-drag rates within the thrust-drag rate sequence and determining the steady-state control mode includes the following steps: Real-time monitoring of the difference between adjacent thrust-drag rates in the thrust-drag rate sequence; If the difference does not exceed the preset difference threshold, it indicates that the current pot is in steady-state control mode. At this time, the push-resistance rate will be continuously monitored and the push-resistance rate sequence will be updated; If the difference exceeds the preset difference threshold, it indicates that the current pot is not in steady-state control mode. At this time, the push-resistance rate will be continuously monitored and the update of the push-resistance rate sequence will be stopped to form the final updated push-resistance rate sequence; When the steady-state control mode is maintained, the thrust-resistance rate is integrated and averaged to form a steady-state reference thrust-resistance rate, which is used to form a standardized resistance reference value in the initial state of the pot cavity.

4. The cooking robot control system based on voice processing according to claim 3 is characterized in that: The dynamic detection subsystem includes a calculation module and a dynamic execution module; The calculation module, after determining the final updated thrust-drag rate sequence, the cooking robot enters the dynamic detection phase, and obtains the thrust-drag deviation by analyzing and calculating the deviation between the real-time monitored thrust-drag rate and the steady-state reference thrust-drag rate; The dynamic execution module pre-sets the deviation threshold based on the mean-standard deviation method and calculates the absolute value of the deviation threshold. If the thrust resistance deviation is greater than 0 and exceeds the deviation threshold, the motor corresponding to the stirring paddle is driven to perform a short-term pulse acceleration to form a centrifugal flow and sauce redistribution operation; Otherwise, the motor corresponding to the stirring paddle is not driven to perform short-time pulse acceleration, and the current steady-state speed stir-frying operation is continued. At the same time, the detection cycle of the next sampling point is entered, and the resistance change inside the pot cavity is continuously monitored until the push-resistance deviation is detected to exceed the deviation threshold, and then the short-time pulse acceleration is triggered.

5. The cooking robot control system based on voice processing according to claim 4 is characterized in that: The dynamic execution module is also used to drive the motor corresponding to the stirring paddle to perform short-term pulse acceleration when the thrust-resistance deviation is greater than 0 and exceeds the deviation threshold, so as to form a centrifugal flow and sauce redistribution operation. The specific steps include: According to the currently monitored thrust-resistance deviation, the amplitude of the pulse acceleration is calculated as follows: ,in, is the amplitude of pulse acceleration, is the proportional gain coefficient, is the deviation threshold, It is the push resistance deviation; When executing short-time pulse acceleration on the motor corresponding to the stirring paddle, the limiting operation is performed, specifically: ,in, is the amplitude of the short-time pulse acceleration of the motor corresponding to the stirring paddle. is the upper limit of the pulse acceleration amplitude; Based on the amplitude of the short-time pulse acceleration executed on the motor corresponding to the stirring paddle, the food in the pot cavity is centrifugally surged to achieve the sauce redistribution operation.

6. The cooking robot control system based on voice processing according to claim 5, characterized in that: The change analysis subsystem includes a pot temperature change module and a pot temperature comparison module; The pot temperature change module collects pot wall temperature data in real time from multiple temperature sensors arranged around the pot cavity during the process of performing short-time pulse acceleration to form centrifugal flow for sauce redistribution. This module calculates the maximum pot cavity temperature gradient after execution. Specifically: ,in, is the maximum value of the pot cavity temperature gradient after execution, To obtain the maximum value, is the temperature value at the i-th temperature sampling point, is the average temperature of each temperature sampling point, i is the number of the temperature sampling point; The pot temperature comparison module extracts the temperature sampling point corresponding to the maximum value of the pot cavity temperature gradient after execution to obtain the difference between the temperature at the temperature sampling point before the short-time pulse acceleration is executed and the temperature average value at the corresponding moment, marked as the temperature difference value. By comparing the temperature difference value with the maximum value of the pot cavity temperature gradient after execution at the corresponding temperature sampling point, the execution change temperature difference is calculated.

7. The cooking robot control system based on voice processing according to claim 6, characterized in that: The pot temperature comparison module is also used to compare the executed temperature difference with the preset temperature threshold. If the executed temperature difference exceeds the temperature threshold, it is determined that a local overheating area appears inside the current pot cavity and the adjustment mechanism is triggered. Otherwise, the mechanism is not triggered.

8. The cooking robot control system based on voice processing according to claim 7 is characterized in that: The adjustment mechanism is triggered to perform dynamic temperature balancing adjustment, specifically including: reducing the amplitude of the short-time pulse acceleration of the motor corresponding to the stirring paddle, and reducing it to the rotation state when the short-time pulse acceleration is not performed, so as to promote the local heat conduction and diffusion of the pot wall.

9. The cooking robot control system based on voice processing according to claim 8, characterized in that: The feedback subsystem re-collects the pot wall temperature data in real time during the stage of promoting local heat conduction and diffusion of the pot wall to re-obtain the execution change temperature difference and re-judge whether to trigger the mechanism. If the mechanism is not triggered, it indicates that the local overheating area in the pot wall has spread. The controller immediately re-enters the push-resistance deviation monitoring stage to form centrifugal flow and sauce redistribution operations again until the user participates in the adjustment function to stop the cooking robot. The adjustment function includes voice mode and press the pause button mode.

Images