Adative cooking system

Abstract

The present disclosure relates to a system for automated cooking based on image driven ingredient identification, said system comprising: an image processing engine configured to receive image data corresponding to one or more food items; process the image data using a trained recognition model to detect and classify the one or more food items; and determine, based on the detected and classified food items, a corresponding cooking instruction dataset by matching the detected food items with a plurality of stored machine-readable cooking instruction datasets. The present disclosure further relates to a cooking control engine operatively coupled to a cooking device, the cooking control engine being configured to: parse the determined cooking instruction dataset; and generate control signals for one or more cooking actuators of the cooking device to execute a cooking process in accordance with the cooking instruction dataset.

Description

ADATIVE COOKING SYSTEMTECHNICAL FIELD

[0001] The present disclosure relates to adaptive control systems for cooking devices and, more particularly, to a control architecture in which image data and device-sensor data are used to update a machine-readable cooking instruction state during an ongoing cooking cycle and to generate revised control commands for physical actuators of the cooking device.BACKGROUND

[0002] Conventional guided-cooking systems generally execute a predefined recipe or time-temperature program. When actual ingredient quantity, ingredient identity, substitution, moisture condition, user timing or other execution conditions depart from the original assumption, such systems typically rely on manual intervention or continue to follow instructions that are no longer technically appropriate.

[0003] Pure recommendation systems and static recipe-selection tools do not solve the problem of real-time technical control of a cooking device. There remains a need for an adaptive cooking control system that detects a difference between expected and observed cooking state, updates a machine -readable instruction state accordingly, and uses the updated state to alter operation of physical cooking actuators during the same cooking cycle.OBJECTS OF THE PRESENT DISCLOSURE

[0004] An object of the present disclosure is to provide an adaptive cooking control system that updates a machine-readable instruction state during an ongoing cooking cycle.

[0005] Another object of the present disclosure is to generate revised control commands for physical actuators of a cooking device on the basis of image-derived and sensor-derived cooking-state information; and

[0006] Yet another object of the present disclosure is to accommodate ingredient substitution, quantity deviation, and user-driven change while preserving technical control of cooking parameters.SUMMARY

[0007] This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

[0008] According to one aspect, the present disclosure provides an adaptive cooking control system comprising an image processing engine configured to receive image data representative of ingredients associated with a cooking device, a monitoring module configured to receive sensor data representative of a current cooking state, an instruction management module configured to store a machine-readable cooking instruction state defining control parameters for physical actuators of the cooking device, a control module configured to detect a deviation between an expected cooking state and the current cooking state, and an execution engine configured to update the machine-readable cooking instruction state and transmit revised control commands to the cooking device during an ongoing cooking cycle.

[0009] The physical actuators may include, for example, one or more heating actuators, a liquid-addition actuator and / or a stirring actuator. The machine -readable instruction state may be stored in a structured format such as JSON. The system may further log updates and observed outcomes for later refinement of baseline instruction states, but the present invention resides in the real-time technical control chain linking sensed deviation, instruction-state update and changed physical device actuation.

[0010] The present disclosure relates to a system for automated cooking based on image driven ingredient identification, said system comprising: an image processing engine configured to receive image data corresponding to one or more food items; process the image data using a trained recognition model to detect and classify the one or more food items; and determine, based on the detected and classified food items, a corresponding cooking instruction dataset by matching the detected food items with a plurality of stored machine-readable cooking instruction datasets. The present disclosure further relates to a cooking control engine operatively coupled to a cooking device, the cooking control engine being configured to: parse the determined cooking instruction dataset; and generate control signals for one or more cooking actuators of the cooking device to execute a cooking process in accordance with the cooking instruction dataset.

[0011] In some embodiments of the present disclosure, the image processing engine may receive said image from an image capturing unit that is configured to capture images of ingredients that are present in a cooking portion of the cooking device and / or present in vicinity of the cooking device.

[0012] In some embodiments of the present disclosure, the image capturing unit may be located on a lid of the cooking device such that when the lid is in closed position, the image capturing unit may be able to capture images of what is inside the cooking portion of thecooking device, and when the lid is in open position, the image capturing unit may be able to capture images of what is present in the vicinity of the cooking device.

[0013] In some embodiments of the present disclosure, the cooking instruction file may be a JavaScript Object Notation (JSON) file that indicates how the detected one or more food items should be cooked in terms of any or a combination of time period of cooking, extent of water to be added, extent of stirring required, and temperature at which cooking is to be done.

[0014] In some embodiments of the present disclosure, when a user, during cooking, changes proportion of the detected one or more food items and / or adds a new food item, the cooking instruction file may be updated accordingly.

[0015] In some embodiments of the present disclosure, when the user, during cooking, changes proportion of the detected one or more food items and / or adds a new food item, the cooking device may prompt the user that the new proportion deviates from the cooking instruction file and / or direct the user to add / reduce quantity of one or more food items to maintain a desired ratio.

[0016] In some embodiments of the present disclosure, wherein when the user, during cooking, changes proportion of the detected one or more food items and / or adds a new food item, the system may dynamically adjust based on the new proportion to optimize the cooking.

[0017] In an aspect, the system is configured, on the basis of the updated machine-readable cooking instruction state, to change one or more of heating level, heating duration, liquidaddition amount, stirring speed, stirring timing and temperature cut-off. In another aspect, the control signals are generated based on sensor data generated by one or more sensors, said sensor data comprises one or more of temperature data, weight data, moisture data, power data and stirrer-load data. In yet another aspect, the system further comprises a database that is configured to store the updated machine-readable cooking instruction state together with observed cooking outcomes for later refinement of a baseline instruction state. In yet another aspect, the cooking device comprises a base, a cooking portion, and the one or more actuators controlled by an execution engine, and wherein the adaptive cooking control system is configured to communicate with the cooking device through one or more interfaces.

[0018] The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, features, and techniques of the disclosure will become more apparent from the following description taken in conjunction with the drawings.

[0019] Various objects, features, aspects and advantages of the inventive subject matter will become more apparent from the following detailed description of preferred embodiments,along with the accompanying drawing figures in which like numerals represent like components.BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0021] FIG. 1 illustrates a block diagram of a system for Al-powered cooking, according to an aspect of the present disclosure.

[0022] FIG. 2 illustrates an example flowchart of a method for Al-powered cooking, according to an aspect of the present disclosure.

[0023] FIG. 3 illustrates an exemplary and non-limiting representation showing the implementation of the proposed system in an aspect of the present disclosure.DETAILED DESCRIPTION

[0024] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such detail as to clearly communicate the disclosure. However, the amount of detail offered is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present disclosure as defined by the appended claims.

[0025] Each of the appended claims defines a separate invention, which for infringement purposes is recognized as including equivalents to the various elements or limitations specified in the claims. Depending on the context, all references below to the "invention" may in some cases refer to certain specific embodiments only. In other cases, it will be recognized that references to the "invention" will refer to subject matter recited in one or more, but not necessarily all, of the claims.

[0026] Various terms are used herein. To the extent a term used in a claim is not defined, it should be given the broadest definition persons in the pertinent art have given that term as reflected in printed publications and issued patents at the time of filing.

[0027] The present disclosure generally relates cooking systems. More particularly, the present disclosure relates to an Artificial Intelligence (Al)-powered cooking systemsincorporating real-time feedback, adaptive recipe modification, and live JavaScript Object Notation (JSON) updates.

[0028] The present disclosure relates to the Al-powered cooking system (referred to as system hereinafter) configured for real-time adaptation of cooking operations based on user actions and sensor feedback. The system may be operatively coupled with a multi-mode cooking device or apparatus capable of supporting traditional heating techniques, including open flame, induction, convection, and conduction, along with microwave -based heating. The apparatus may enable selective or simultaneous operation of these heating modes to provide flexible cooking configurations.

[0029] In one implementation, the cooking device may include a base, a support structure configured to hold a cooking container in a suspended manner, and a lid movably coupled to the base between open and closed positions. A microwave-generating module may be integrated with the lid such that, upon closure, microwave energy is directed into the container to enable combined or standalone microwave heating.

[0030] The cooking device may ensure secure engagement between the lid and the container, maintaining alignment during operation and preventing gaps. Further, the suspended positioning of the container may reduce heat transfer back to the base, thereby improving thermal efficiency and enabling sustained high-temperature operation without overheating the base structure.

[0031] The Al-powered cooking system may enable adaptive control of cooking parameters in response to real-time events. For instance, when a delay in ingredient addition is detected, the system may automatically reduce heating power to prevent overheating or burning.

[0032] When ingredients are added during an ongoing cooking process, the system may detect changes in weight or volume using sensors and / or image-based recognition. Based on such detection, the system may dynamically update structured cooking instructions represented in JavaScript Object Notation (JSON), and correspondingly adjusts temperature, cooking duration, and stirring parameters.

[0033] In one example, if the actual quantity of the ingredient deviates from an expected value, the system may determine the variation and proportionally adjusts associated parameters, such as liquid quantity, heating intensity, and cooking time. Continuous monitoring enables real-time correction of cooking conditions to maintain desired outcomes.

[0034] The system may further implement a closed-loop feedback mechanism, where sensor inputs are continuously processed to regulate heating levels, thereby preventingovercooking or undercooking. Heat levels may be temporarily reduced or maintained until the next cooking step is confirmed.

[0035] A real-time instruction update module of the system may continuously modify the JSON-based cooking profile as the cooking process evolves. Updated instructions may be transmitted to an execution engine associated with the cooking device, enabling immediate adjustment of stirrer speed, heating power, and cooking duration.

[0036] The system may also provide interactive assistance to the user by generating realtime recommendations. In response to user interventions or detected anomalies, such as incorrect ingredient proportions, the system may suggest corrective measures to maintain recipe integrity.

[0037] In some embodiments, the Al-powered cooking system may include a monitoring module configured to track user actions, ingredient additions, and timing deviations, and an adaptive control module configured to dynamically regulate heating parameters based on detected variations in cooking conditions. A stirring control module may adjust stirring speed in accordance with cooking phase and ingredient properties, including viscosity. The system may further modify cooking instructions to accommodate variations, including extending cooking duration, adjusting heating intensity, or altering stirring sequences.

[0038] Ingredient recognition and quantification may be achieved through image processing and sensor-based estimation. The system may compare detected values with expected parameters and initiate adjustments when deviations exceed predefined thresholds. Based on such deviations, the system may modify liquid content, adjust heating power across different modes, and recalibrate cooking duration to ensure consistent results. Ingredient substitutions may also be detected and accommodated through corresponding updates in cooking parameters.

[0039] The system may include an image processing engine(s) or a sensor framework comprising temperature, viscosity, and ambient condition sensors to provide continuous input to the system. The Al-based control model may process the data to dynamically regulate heating intensity, cooking time, and stirring operations. A safety mechanism may be incorporated to prevent burning by reducing heat when delays or anomalies are detected. The system may also adapt to variations in cooking progression by extending or modulating heat application as required.

[0040] The system may include a dynamic instruction management module configured to update cooking steps in real time based on sensor data, user actions, and environmental conditions. The execution engine ensures immediate implementation of updated instructions.User interaction may be supported through a live assistance module or a user interface that provides contextual prompts, alerts, and recommendations based on cooking status and detected issues. The system may also suggest alternative steps or ingredient substitutions where necessary.

[0041] The cooking device may be equipped with multiple sensors, including an ambient sensor for monitoring internal conditions, a food temperature sensor for tracking cooking progress, and a surface temperature sensor for precise heat regulation. Sensor data may be continuously transmitted to a processing unit, which updates the cooking profde and controls the cooking device in real time, thereby maintaining optimal cooking conditions through a feedback-driven control loop.

[0042] The system may further include a recipe generation and management engine configured to extract, structure, and store recipes from external sources, enabling continuous learning and generation of optimized cooking instructions. Ingredient recognition may be enhanced through computer vision models that identify ingredient types and estimate corresponding quantities from captured images, enabling automatic adjustment of cooking parameters.

[0043] Cooking instructions may be compiled into a structured JSON format that is dynamically updated during operation based on real-time inputs. The control module ensures synchronized execution of these instructions across different heating modes.

[0044] The system may further integrate with external commerce platforms to detect missing ingredients and facilitate procurement or suggest alternatives, with corresponding updates to the cooking workflow. Overall, the disclosed system enables real-time adaptive cooking control by continuously adjusting heating parameters, stirring operations, and cooking duration in response to user behavior and sensor inputs, thereby ensuring consistent and optimized cooking outcomes.

[0045] Various embodiments of the present disclosure will be explained in detail with reference to FIGs. 1 to 3.

[0046] FIG. 1 illustrates a block diagram of a system 100 for Al-powered cooking, according to an aspect of the present disclosure.

[0047] With reference to FIG. 1, the system 100 for Al-powered cooking is illustrated. The system 100 may be operatively coupled with the cooking device 120 and a database 140, and may be configured to enable automated, adaptive cooking based on image-driven intelligence and dynamic instruction management.

[0048] In an embodiment, the system 100 may include one or more processor(s) 102, a memory 104, and an interface(s) 106, which collectively facilitate execution of software modules and communication with the cooking device 120 and the database 140. The memory 104 may store executable instructions and cooking instruction files, while the interface(s) 106 enable data exchange between system components.

[0049] The processor 102 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and / or any devices that manipulate data based on operational instructions. Among other capabilities, the processor 102 may be configured to fetch and execute computer-readable instructions stored in the memory 104 of the system 100. The memory 104 may store one or more computer-readable instructions or routines or cooking instruction files, which may be fetched and executed to create or share data packets over a network service. The memory 104 may include any non-transitory storage device including, for example, a volatile memory such as a Random-Access Memory (RAM), or a non-volatile memory such as an Erasable Programmable Read-Only Memory (EPROM), a flash memory, and the like.

[0050] In an embodiment, the interface(s) 106 may comprise a variety of interfaces, for example, interfaces for data input and output devices, referred to as I / O devices, storage devices, and the like. The interface(s) 106 may facilitate communication of the system 100 with various devices or modules coupled to the system 100. The interface(s) 106 may also provide a communication pathway for components of the system 100. Examples of such components include, but are not limited to, a processing engine(s) 108, the cooking device 120, and the database 140. The interface(s) 106 may be interchangeably referred to as a user interface.

[0051] The system 100 may include an image processing engine 108, an instruction management module 110, a monitoring module 112, and one or more control module(s) 114, each operatively coupled with the processor(s) 102.

[0052] In certain embodiments, the image processing engine 108 may be configured to receive an image of one or more food items to be cooked. The image processing engine 108 may be configured to detect the food items from the received image using image recognition techniques, and process the image to identify a closest matching cooking instruction file stored in the memory 104 or the database 140.

[0053] The present disclosure relates to a system for automated cooking based on image driven ingredient identification, said system comprising: an image processing engine configured to receive image data corresponding to one or more food items; process the imagedata using a trained recognition model to detect and classify the one or more food items; and determine, based on the detected and classified food items, a corresponding cooking instruction dataset by matching the detected food items with a plurality of stored machine-readable cooking instruction datasets. The present disclosure further relates to a cooking control engine operatively coupled to a cooking device, the cooking control engine being configured to: parse the determined cooking instruction dataset; and generate control signals for one or more cooking actuators of the cooking device to execute a cooking process in accordance with the cooking instruction dataset.

[0054] The present disclosure relates to a system for automated cooking based on image driven ingredient identification, said system comprising: an image processing engine configured to receive image data corresponding to one or more food items; process the image data using a trained recognition model to detect and classify the one or more food items; and determine, based on the detected and classified food items, a corresponding cooking instruction dataset by matching the detected food items with a plurality of stored machine-readable cooking instruction datasets. The present disclosure further relates to a cooking control engine operatively coupled to a cooking device, the cooking control engine being configured to: parse the determined cooking instruction dataset; and generate control signals for one or more cooking actuators of the cooking device to execute a cooking process in accordance with the cooking instruction dataset.

[0055] In an exemplary and non-limiting aspect, image data acquired from an image capturing unit can first subjected to a pre-processing pipeline to standardize the input for downstream inference. The pre-processing includes operations such as spatial resizing to a predefined resolution, pixel value normalization, color space conversion, and noise reduction using filtering techniques. In certain implementations, illumination correction and contrast enhancement may also be performed to mitigate variability arising from environmental lighting conditions.

[0056] The pre-processed image can then be converted into a multi-dimensional tensor and provided as input to a trained recognition model. The trained recognition model may comprise a deep learning architecture, such as a convolutional neural network (CNN), a region-based detection network (e.g., Faster R-CNN), a single-shot detector (e.g., YOLO or SSD), or a transformer-based vision model. The model is pre-trained on a large-scale dataset and optionally fine-tuned on a domain-specific dataset comprising labeled images of food items and ingredients.

[0057] During inference, the model performs hierarchical feature extraction through multiple layers, wherein lower layers can capture spatial and texture-based features and deeper layers can capture semantic representations. For object detection, the model can generate a set of candidate regions of interest using either anchor-based or anchor-free mechanisms. Each candidate region is processed to output bounding box coordinates along with corresponding class probabilities representing identified food item categories.

[0058] In an aspect, a classification head can assign one or more class labels to each detected region based on the computed probability distribution, and a confidence score is associated with each detection. In parallel, bounding box regression refines the spatial localization of detected objects within the image. Post-processing operations, such as nonmaximum suppression (NMS), are applied to eliminate redundant or overlapping detections based on intersection-over-union (loU) thresholds, thereby retaining only the most probable detections.

[0059] In certain embodiments, the model may further estimate additional attributes, such as portion size, volume approximation, or state of the ingredient (e.g., chopped, whole, cooked), using auxiliary prediction heads or regression models. The outputs of the recognition model are structured as a set of detected food items, each associated with metadata including class label, confidence score, spatial coordinates, and optionally derived attributes.

[0060] The detected and classified food items can then be mapped to a predefined ontology or ingredient taxonomy to ensure consistency with downstream processing modules. In some implementations, a confidence threshold is applied to filter out low-probability detections, and ensemble or multi-frame aggregation techniques may be employed to improve robustness across sequential image captures. The resulting structured representation of detected food items serves as an input to subsequent modules for selection or generation of machine -readable cooking instruction datasets, thereby enabling automated and context-aware cooking operations.

[0061] The identified cooking instruction file may define a sequence of cooking steps for preparing the detected food items. The instruction management module 110 may retrieve and structure the instructions in the cooking instruction file, while the control module(s) 114 transmit corresponding control signals to the cooking device 120. An execution engine 130 of the cooking device 120 may execute these instructions to perform cooking operations.

[0062] In some embodiments, the cooking device 120 may include an image capturing unit 122, which captures images of ingredients. The captured images may be transmitted via the interface(s) 106 to the image processing engine 108. The image capturing unit 122 may beconfigured to capture images within the cooking portion of the cooking device 120, and / or in the surrounding vicinity of the cooking device 120.

[0063] In an implementation, the image capturing unit 122 may be associated with a lid 128 of the cooking device 120. In another implementation, the image capturing unit 122 may be integrated with the lid 128 of the cooking device 120. Accordingly, when the lid 128 is in a closed position, the image capturing unit 122 may capture images of an interior cooking portion, and images of what is inside the cooking portion of the cooking device 120. When the lid is in open position, the image capturing unit 122 may be able to capture images of what is present in the vicinity of the cooking device 120 or captures images of ingredients present in the vicinity of the cooking device 120.

[0064] In an embodiment, the cooking instruction file, as managed by the instruction management module 110 and stored in the database 140, may be structured in a JavaScript Object Notation (JSON) format. The JSON file may encode cooking parameters including, without limitation, cooking duration, quantity of liquid (e.g., water) to be added, stirring requirements, and temperature or power levels. The execution engine 130 of the cooking device 120, under control of the control module(s) 114, may execute the encoded cooking parameters during cooking.

[0065] In an exemplary and non-limiting implementation, upon detecting and classifying the identified food items, they can be mapped against a cooking instruction dataset / file (say in a JSON format) that is represented in a structured format comprising class labels, confidence scores, and optionally associated attributes such as quantity, state, and spatial context. In an aspect, structured representation of the detected food items can be provided as input to a recipe selection and matching module (for instance) configured to determine a corresponding cooking instruction dataset.

[0066] The system can be configured to maintain a plurality of stored machine -readable cooking instruction datasets, each associated with metadata defining required ingredients, ingredient ratios, preparation states, and cooking parameters. The metadata may be represented using a standardized schema or ontology to enable consistent comparison with the detected food item representation.

[0067] In another aspect, in order to determine the corresponding cooking instruction dataset, the system can perform a matching operation between the detected food items and the ingredient metadata associated with each stored dataset. In one implementation, the matching comprises computing a similarity score based on overlap between detected ingredients and required ingredients, optionally weighted by confidence scores, ingredient importance, andquantity estimates. The similarity computation may further account for partial matches, substitutable ingredients, and hierarchical relationships defined in the ingredient ontology.

[0068] In certain exemplary and non-limiting embodiments, feature vectors representing the detected food items and the stored cooking instruction datasets can be generated and compared using distance metrics or learned embedding spaces. Alternatively, rule-based or hybrid approaches may be employed, wherein deterministic rules filter candidate datasets followed by machine learning -based ranking.

[0069] The system may generate a ranked list of candidate cooking instruction datasets based on computed similarity scores and select a highest-ranking dataset that satisfies predefined selection criteria, such as minimum ingredient coverage or threshold confidence levels. In cases where multiple candidate datasets satisfy the criteria, additional contextual factors, including user preferences, historical usage, or available cooking resources, may be incorporated to refine selection.

[0070] In some implementations, the system dynamically adapts the selected cooking instruction dataset to align with the detected ingredient set by scaling ingredient quantities, omitting unavailable ingredients, or incorporating substitutions. Constraint validation may be performed to ensure that the adapted dataset remains executable within operational limits of the cooking device.

[0071] The selected and optionally adapted cooking instruction dataset can then be output in a structured format suitable for parsing by a cooking control engine, thereby enabling automated execution of cooking operations based on the detected and classified food items.

[0072] Further, in an embodiment, the monitoring module 112, in conjunction with the image processing engine 108, may continuously track changes during cooking. When a user, modifies the proportion of existing ingredients, and / or adds new ingredients, the image processing engine 108 may detect such changes via updated images from the image capturing unit 122. Based on this detection, the instruction management module 110 may update the JSON-based cooking instruction file in real-time. The updated instructions may then be transmitted to the cooking device 120 for execution.

[0073] Upon detecting deviations in ingredient proportions, the system 100, via the monitoring module 112 and the control module(s) 114, may generate prompts or notifications to the user through the interface(s) 106. Additionally, the system 100 may alert the user that the current ingredient proportions deviate from the defined cooking instruction file, and / or recommend corrective actions, such as adding or reducing specific ingredients to maintain a desired ratio.

[0074] The system 100 may further be configured to autonomously adapt cooking parameters based on detected changes. Specifically, the image processing engine 108 may detect updated ingredient composition, the instruction management module 110 may recalculate the cooking parameters, and the control module(s) 114 may adjust operational parameters such as heating intensity, cooking time, and stirring via the execution engine 130 based on the updated cooking parameters. This ensures optimized cooking outcomes despite variations in ingredient proportions.

[0075] In operation, the image capturing unit 122 may acquire images of ingredients, which are analysed by the image processing engine 108 to identify food items and corresponding cooking instructions. These instructions, maintained as JSON files in the database 140, may be dynamically managed and updated by the instruction management module 110. The control module(s) 114 may transmit execution signals to the cooking device 120, where the execution engine 130 performs cooking actions. The monitoring module 112 may ensure continuous feedback and adaptation throughout the cooking process.

[0076] An adaptive cooking control system includes an image processing engine for receiving image data representative of ingredients associated with a cooking device, a monitoring module for receiving sensor data representative of a current cooking state, an instruction management module for storing a machine-readable cooking instruction state defining control parameters for physical actuators of the cooking device, a control module for detecting a deviation between an expected cooking state and the current cooking state, and an execution engine configured to update the machine-readable cooking instruction state and transmit revised control commands to the cooking device during an ongoing cooking cycle. The machine-readable instruction state may be stored in a structured format, including JSON.

[0077] FIG. 2 illustrates an example flowchart of a method 200 for Al-powered cooking, according to an aspect of the present disclosure.

[0078] With reference to FIG. 2, the method 200 begins at step 202, where the user provides ingredients either through manual input or by enabling image-based capture. This input forms a foundation for the system’s understanding of what needs to be cooked. The captured data may then be passed to an Al control module (similar to control module 114 as illustrated in FIG. 1), at step 204. The Al control module analyses the input to identify the types of ingredients and estimate their quantities using image recognition and / or sensor-based techniques. This step ensures that the system 100 has an accurate representation of the cooking inputs before proceeding further.

[0079] Once the ingredients are identified, the system 100 may perform recipe selection, at step 206, by retrieving a suitable recipe from the database 140 or generating one using Al models. The selected recipe may be then structured into a machine-readable format, typically a JSON file. This structured format may enable precise definition of cooking parameters and allow seamless communication between different modules of the system 100.

[0080] At step 208, the system 100 may generate a dynamic JSON profile that includes initial cooking instructions such as cooking duration, heating levels, stirring requirements, and liquid quantities. These instructions may be executed by the execution engine, at step 210, which controls the cooking device 120 by regulating induction heating, microwave energy, stirring mechanisms, and liquid dispensing. This marks the initiation of the actual cooking process based on the predefined parameters.

[0081] During cooking, the system 100 may continuously gather real-time sensor feedback, at step 212, including parameters such as temperature, moisture content, and weight changes. This data reflects the actual cooking conditions and is used to evaluate whether the process is progressing as expected. Any deviation from the intended conditions may be detected at this stage, enabling the system 100 to respond proactively.

[0082] Based on the sensor feedback, the system 100 may perform live JSON modification, at step 214, dynamically updating cooking parameters such as heat intensity, stirring speed, and liquid levels. These updated instructions may be immediately fed back to the execution engine 130, forming a closed-loop control system that ensures continuous optimization. Finally, at step 216, the system 100 may store the observed variations and applied adjustments, enabling Al-driven learning and future recipe optimization, thereby improving performance and accuracy in subsequent cooking cycles.

[0083] FIG. 3 illustrates an exemplary and non-limiting representation 300 showing the implementation of the proposed system 100 in an aspect of the present disclosure.

[0084] FIG. 3 illustrates an integrated Al-powered cooking ecosystem that combines ingredient sourcing, intelligent processing, adaptive cooking execution, and continuous learning, all centered around the cooking device 120.

[0085] The process begins with ingredient acquisition and assimilation. Ingredients may be sourced via quick commerce 302 or prepared manually through weighing, chopping, and plating 304, and may be consolidated through an ingredient assimilation module 306. Additionally, the ingredients may be suggested or supplemented based on inputs from a smart fridge 308, online sources (scraped data), or Al-generated recipes. These inputs collectivelyform the ingredients dataset 310, which includes vegetables, meats, spices, and other components.

[0086] The identified ingredients may be processed by an Al processing module 312, which determines the final ingredient set along with their corresponding weights. This processing may also incorporate contextual parameters such as location, altitude, ambient temperature, pressure, and moisture conditions to improve accuracy. Based on this refined input, a recipe module 314 may perform recipe selection, creation, addition, or substitution, ensuring that the cooking process aligns with available ingredients and desired outcomes.

[0087] The selected recipe may then be converted into a dynamic JSON profile 316, which defines structured cooking instructions such as heating levels, cooking duration, stirring parameters, and liquid requirements. These instructions may be executed by the execution engine 130, which controls cooking operations including induction heating, microwave heating, stirring, and liquid addition (e.g., oil, water, milk). The cooking device 120 may simultaneously provide sensor data, power data, and load data, enabling precise monitoring of cooking conditions.

[0088] During execution, the system 100 may continuously gather real-time sensing and feedback 318, including temperature, weight, moisture, and other cooking parameters. This feedback may be used to perform live JSON modification 322, dynamically adjusting cooking instructions such as heat levels, stirring speed, liquid addition, and temperature cut-offs. These updates may be immediately fed back to the execution engine 130, forming a closed-loop adaptive control system.

[0089] Finally, the system 100 may incorporate an Al learning and optimization module 320 / 324, which leverages historical JSON recipes, experimental data, and observed cooking variations to improve future performance. Additional inputs, such as stirrer load and consistency analysis 326, further enhance learning. The system 100 may store these insights and continuously refine recipe selection, parameter tuning, and cooking strategies, enabling progressive optimization of quality and flavour profiles across subsequent cooking cycles.

[0090] Therefore, the present disclosure discloses the Al -assisted cooking system 100 configured for real-time adaptation based on user actions. The system 100 may be operatively coupled with a multi-mode cooking device 120 that supports conventional cooking methods such as open flame, induction, convection, and conduction, while also enabling microwavebased cooking. The cooking device 120 allows the user to selectively use or combine these modes, including transitioning from traditional heating to microwave cooking during the process or operating them simultaneously. In one implementation, the cooking device 120 mayinclude the base, the support structure configured to support a container in a suspended position, and a pivotable lid. A microwave-generating device may be integrated with the lid such that, in the closed position, microwave heating can be applied to the contents of the container.

[0091] The cooking device 120 ensures secure engagement between the lid and the container, maintaining proper alignment without gaps during operation. The container may be positioned at a safe distance from the base or induction element, thereby reducing heat transfer back to the base and improving efficiency. This configuration enables sustained high-temperature operation while maintaining safety and structural stability.

[0092] The system 100 dynamically responds to user actions during cooking. For example, if there is a delay in adding ingredients, the system 100 automatically reduces induction or microwave power to prevent overheating or burning. Conversely, when ingredients are added mid-process, the system 100 detects updated weight or volume through sensors or image-based recognition and adjusts cooking parameters accordingly.

[0093] The system 100 utilizes a dynamic recipe model defined in JSON, which is continuously updated in real time. Based on detected changes, the system 100 modifies parameters such as ingredient proportions, liquid requirements, temperature, stirring speed, and cooking duration. For instance, if the actual quantity of an ingredient exceeds the initially estimated amount, the system 100 compensates by adjusting associated inputs such as water content and heat levels.

[0094] The real-time monitoring module tracks parameters including temperature, weight, moisture, and viscosity using integrated sensors. These inputs may be processed by an Al-based control module that continuously refines cooking conditions through a feedback loop. Heat levels, stirring intensity, and cooking time are dynamically optimized to ensure consistent results and prevent overcooking or undercooking.

[0095] The system 100 further includes a live JSON update / management module that synchronizes modified cooking instructions with the execution engine 130 in real time. This enables immediate adjustment of operational parameters such as induction power, microwave intensity, stirrer speed, and liquid dispensing without manual intervention. The system 100 also accounts for external variables such as ambient conditions and cookware variations.

[0096] In addition, the system 100 provides interactive assistance to the user through real-time recommendations and alerts. The system 100 can suggest corrective actions in response to errors, such as ingredient imbalance, and offer alternative steps or substitutions based on current conditions and user preferences.

[0097] The disclosed system incorporates machine learning capabilities to improve performance over time. Historical cooking data, including prior JSON profdes and sensor feedback, is analysed to optimize future recipes, enhance flavour outcomes, and refine control strategies.

[0098] The cooking device 120 is equipped with multiple sensors, including ambient sensors, food temperature sensors, and surface temperature sensors, which collectively monitor internal and external cooking conditions. These sensors continuously transmit data to the execution engine 130, enabling precise control and real-time adaptation.

[0099] By pre-heating the interior of the food, the duration of immersion in hot oil is significantly reduced. Induction heating maintains oil temperature within a narrow range, ensuring uniform cooking and minimizing oil uptake. As a result, frying time can be reduced by, approximately 30% to 50%, while oil absorption is reduced by at least 60%. The method produces food with a crisp exterior and fully cooked interior, while maintaining lower fat content. Typical oil absorption can be reduced to approximately 6%-8% by weight, compared to significantly higher levels in conventional frying methods.

[0100] For frying applications, food items may be pre-cooked using microwave energy and subsequently finished in oil maintained at controlled temperatures. This approach further reduces cooking time and oil absorption while achieving desired texture and doneness. Therefore, the disclosed system and methods enable efficient, adaptive, and healthier cooking by integrating real-time monitoring, Al-driven control, and multi-mode heating technologies.

[0101] While the foregoing describes various embodiments of the invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof. The scope of the invention is determined by the claims that follow. The invention is not limited to the described embodiments, versions or examples, which are included to enable a person having ordinary skill in the art to make and use the invention when combined with information and knowledge available to the person having ordinary skill in the art.ADVANTAGES OF THE PRESENT DISCLOSURE

[0102] The present disclosure discloses Al-powered cooking system that continuously monitors user actions, ingredient changes, and cooking conditions, allowing automatic adjustment of heat, time, and stirring, thereby reducing reliance on manual intervention and improving consistency in outcomes.

[0103] The present disclosure combines induction, microwave, and other heating modes within a single apparatus, such that the system enables seamless switching or simultaneousoperation, thereby allowing optimized cooking strategies for different recipes and stages of cooking.

[0104] The present disclosure provides Al-based processing and sensor feedback that ensures precise control over parameters such as temperature, ingredient quantity, moisture, and viscosity, leading to better reproducibility and minimizes errors.

[0105] The present disclosure uses real-time JSON-based recipe updates which allows the system to instantly adapt to variations such as delayed ingredient addition, incorrect quantities, or substitutions, ensuring the recipe remains balanced.

[0106] The present disclosure ensures automatic heat regulation in response to delays or unexpected changes prevents overheating, burning, and degradation of food quality.

[0107] The present disclosure provides real-time guidance, suggestions, and corrective actions, making the system suitable for both novice and experienced users while reducing the cognitive load during cooking.

[0108] The present disclosure includes machine learning capabilities that enable the system to learn from past cooking data, improving future performance, taste optimization, and personalization.

[0109] The present disclosure facilitates combination of microwave pre-heating and induction-based finishing, thereby significantly shortening cooking durations compared to conventional methods.

[0110] The present disclosure achieves desirable results such as crisp exteriors and well-cooked interiors by intelligently coordinating internal and external heating.

[0111] The present disclosure ensures shorter cooking times and controlled heat exposure that retains vitamins and nutrients that are typically lost in prolonged conventional cooking.

[0112] The present disclosure discloses the system that can handle a wide range of food items, including vegetables, meats, and fried products, with minimal adjustments.

[0113] The present disclosure ensures continuous sensor-driven feedback that enables real-time corrections, ensuring stable and optimal cooking conditions throughout the cooking process.

Claims

WE CLAIM:

1. A system for automated cooking based on image driven ingredient identification, said system comprising:an image processing engine configured to:receives image data corresponding to one or more food items; process the image data using a trained recognition model to detect and classify the one or more food items;determine, based on the detected and classified food items, a corresponding cooking instruction dataset by matching the detected food items with a plurality of stored machine -readable cooking instruction datasets; and; and a cooking control engine operatively coupled to a cooking device, the cooking control engine being configured to:parse the determined cooking instruction dataset; andgenerate control signals for one or more cooking actuators of the cooking device to execute a cooking process in accordance with the cooking instruction dataset.

2. The system as claimed in claim 1, wherein the image processing engine receives said image data from an image capturing unit that captures images of the one or more food items that are present in a cooking container of the cooking device and / or present in vicinity of the cooking device .

3. The system as claimed in claim 2, wherein the image capturing unit is located on a lid of the cooking device such that when the lid is in closed position, the image capturing unit is able to capture images of inner side of the cooking container of the cooking device, and when the lid is in open position, the image capturing unit is able to capture images of what is present in the vicinity of the cooking device.

4. The system as claimed in claim 1, wherein the cooking instruction file is a structured machine-readable format file that indicates manner in which the detected one or more food items should be cooked in terms of any or a combination of time period of cooking, extent of water to be added, extent of stirring required through a stirrer, and temperature at which cooking is to be done.

5. The system as claimed in claim 1, wherein the one or more cooking actuators are selected from any or a combination of a lower heating actuator, a microwave-generating actuator, a liquid-addition actuator and a stirring actuator.

6. The system as claimed in claim 1, wherein when a user, during cooking, changes proportion of the detected one or more food items and / or adds a new food item, the cooking instruction fde is dynamically updated accordingly.

7. The system as claimed in claim 1, wherein when a user, during cooking, changes proportion of the detected one or more food items and / or adds a new food item, the cooking device generates user prompts responsive to deviation thresholds and / or directs the user to add / reduce quantity of one or more food items to maintain a desired ratio.

8. The system as claimed in claim 1, wherein when a user, during cooking, changes proportion of the detected one or more food items and / or adds a new food item, the system dynamically adjusts based on the new proportion to optimize the cooking.

9. The system as claimed in claim 1, wherein the system is configured, on the basis of the updated machine-readable cooking instruction state, to change one or more of heating level, heating duration, liquid-addition amount, stirring speed, stirring timing and temperature cut-off.

10. The system as claimed in claim 1, wherein the control signals are generated based on sensor data generated by one or more sensors, said sensor data comprises one or more of temperature data, weight data, moisture data, power data and stirrer-load data.

11. The system as claimed in claim 1, said system further comprising a database configured to store the updated machine-readable cooking instruction state together with observed cooking outcomes for later refinement of a baseline instruction state.

12. The system as claimed in claim 1, wherein the cooking device comprises a base, a cooking portion, and the one or more actuators controlled by an execution engine, and wherein the adaptive cooking control system is configured to communicate with the cooking device through one or more interfaces.

13. A method for automated cooking based on image driven ingredient identification, said method comprising the steps of:receiving, through an image processing engine, image data corresponding to one or more food items;processing the image data using a trained recognition model to detect and classify the one or more food items;determining, based on the detected and classified food items, a corresponding cooking instruction dataset by matching the detected food items with a plurality of stored machine-readable cooking instruction datasets; andparsing, through a cooking control engine operatively coupled to a cooking device, the determined cooking instruction dataset; andgenerating, through the cooking control engine, control signals for one or more cooking actuators of the cooking device to execute a cooking process in accordance with the cooking instruction dataset.

14. The method as claimed in claim 8, wherein the image processing engine receives said image data from an image capturing unit that captures images of the one or more food items that are present in a cooking container of the cooking device and / or present in vicinity of the cooking device.

15. The method as claimed in claim 9, wherein the image capturing unit is located on a lid of the cooking device such that when the lid is in closed position, the image capturing unit is able to capture images of inner side of the cooking container of the cooking device, and when the lid is in open position, the image capturing unit is able to capture images of what is present in the vicinity of the cooking device.

Images