Deep learning artificial intelligence doner kebab robot

Abstract

The invention relates to the doner kebab robot (A), which performs the mass production of doner kebab, which is frequently consumed in the ready-made food industry, performs automatic doner kebab cooking and cutting functions by means of the use of artificial intelligence with the help of deep learning technique, and comprises an image processing computer (2), where the volume, temperature, colour and shape values of the image data received from the camera (1 ) are processed, at least one gas valve (4) where the amount of gas is automatically adjusted for heat adjustment, and at least one camera (1 ) capable of taking snapshots at least 24 frames per second.

Description

[0001] Deep Learning Artificial Intelligence Doner Kebab Robot

[0002] Technical Field

[0003] The invention is about obtaining error-free products by means of the use of artificial intelligence with the help of deep learning technique in the doner kebab cooking and cutting robot, where the mass production of doner kebab, which is frequently consumed in the ready-made food industry, is made.

[0004] The invention is particularly related to a deep learning artificial intelligence doner kebab robot in which processes such as cooking time, amount of flammable gas or resistance power, rotation speed, depth of cut and determination of the surface to be cut are robotised and automated by analysing the data obtained from the cameras and sensors it comprises.

[0005] State of the Art

[0006] Ready-made food production lines consist of many units. Doner kebab is a type of ready meal that is frequently consumed in daily life and has become a global favourite. Since doner kebabs are cooked and cut by hand in places such as doner kebab halls and restaurants, obtaining a sufficiently cooked product requires mastery skills.

[0007] Today, as doner kebab consumption increases all over the world, the need for packaged and frozen products increases, and therefore the need for production also increases. The method frequently used in the production of packaged ready-made doner kebabs is to line up multiple doner kebabs side by side in series and continuously cut the cooked product at a fixed thickness, at predefined time intervals and speed, by a mechanical or automatic knife. Parameters such as the rotation speed of the cooked product and the amount of heat given are constantly monitored by the operator at the machine, and the operator intervenes when a setting change is required.

[0008] The biggest disadvantage of this cooking method is that it requires mastery skills and requires one or more experienced operators depending on the number of times the machine cooks. Configurations have been studied in which a laser sensor determines the position of the knife by measuring the distance to the meat cluster while the machine is cutting. Since the area around the doner kebab block attached to the machine is not always cooked evenly, some parts may remain undercooked or may overcook and burn. Since the cutting blade works depending on the distance parameter, it cuts in the same way regardless of the cooking amount, so the undercooked or burnt parts that are not suitable for food safety are mixed with the intact product, go to the packaging line and are packaged and delivered to the end user. It is not possible to detect undercooked product on the production line. The process of manually separating the overly burnt parts is not preferred as it requires an operator and would violate food safety regarding the production of products without human contact.

[0009] It seems that some studies have been being conducted to solve the current problems. In the literature review, a doner kebab production machine was included in the patent application titled "WOOD-FIRED DONER MACHINE WITH INTELLIGENT COOKING MANAGEMENT SYSTEM" in the document numbered TR2019 / 21183. The summary of said invention reads: “The invention belongs to the technical fields of food and machine technology. With the invention, a new doner kebab cooking system is proposed that provides the taste of wood-fire cooking, which is a traditional doner kebab cooking system, has the ease of use of electric doner kebab cooking, allows healthier doner kebab production with its developed features, and has a smart cooking management system that guides the user with high energy efficiency. In this cooking system, it is possible to produce and cook different foods (vegetable doner, vegan products, cheese, etc.) as well as doner kebab (meat doner, chicken doner, leaf doner, mixed doner) by developing new formulations. In this application, it is aimed to obtain an aromatic taste by spraying the hot air obtained from the wood fire onto the product while cooking the doner kebab using electric resistance instead of gas radiant heaters. By using elements such as diameter sensor, temperature sensor, distance sensor and camera on the machine, machine motion control is provided with the help of data received from each sensor. Since the distance of the doner kebab block to the heat source is adjusted by a moving mechanism according to the measured heat, it requires an additional movement mechanism. Since the camera on the machine cannot perform any analysis other than warning the user based on colour values, an additional diameter sensor and distance sensor are needed to measure the volume and distance of the doner kebab block.

[0010] As a result, due to the negativities explained above and the inadequacy of existing solutions on the subject, it became necessary to make a development regarding the doner kebab robot.

[0011] The Aim of the Invention

[0012] The invention is inspired by current situations and aims to solve the above- mentioned drawbacks.

[0013] The main aim of the invention is to ensure fully automatic doner kebab production in ready-made doner kebab production lines, without human contact, in accordance with food safety, by using deep learning technology.

[0014] Another aim of the invention is to provide a doner kebab robot that comprises an image processing computer that processes the volume, temperature, colour and shape values of at least 24 frames per second of image data with the help of at least one camera, and controls the elements of the doner kebab robot to automatically identify different types of products and cook them appropriately by means of the artificial intelligence software. Another aim of the invention is to ensure that said image processing computer can analyse and process data such as total cutting time, total cutting length, energy consumption of heat generator sections and transmit it to a data centre in industry 4.0 standards with the help of the wired and / or wireless data transfer module it comprises and enable the instantaneous information flow to be controlled.

[0015] Another aim of the invention is that said doner kebab robot comprises an automatic gas valve and / or potentiometer connected to more than one combustion section, where homogeneous cooking is achieved by adjusting the cooking speed, and where the gas output and flame amount are automatically adjusted with the command from the image processing computer. The gas valve also includes manual controls for emergencies.

[0016] Another aim of the invention is to include at least one camera, preferably with a resolution of 1920x1080 pixels, that can take an image of at least 640x480 pixels, preferably taking 60 frames (fps) images, taking the baked product images previously introduced into the system as reference, and enabling the volume measurement of the product and the position of the automatic cutting knife to be determined.

[0017] Another aim of the invention is to automatically detect the fault according to the difference between the cooking amounts of the other parts in case there is no cooking due to a malfunction that may occur in the combustion parts of said camera.

[0018] Another aim of the invention is to include a cylindrical or linear cutting blade that can automatically cut from any angle on the surface of the doner kebab block by means of its ability to move two axes on a slide with horizontal (y-axis) or vertical (x-axis) movement according to the position data received from the image processing computer. Another aim of the invention is that said cutting knife can perform regional cutting to separate uncooked or overcooked parts in case of heat imbalance that may occur in a section in heated systems with more than one section.

[0019] Another aim of the invention is to include an automatic rotation mechanism in which the rotation speed is adjusted according to the cooking sample data coming from the image processing computer in order to automatically adjust the production speed depending on the desired weight.

[0020] Another aim of the invention is to include a digital control panel comprising at least one screen that displays production parameters and allows manual intervention.

[0021] Another aim of the invention is to provide automatic control of the amount of heat, rotation speed and cutting depth depending on reasons such as customer density and production capacity.

[0022] Another aim of the invention is to include a chamber with load cells that will weigh the amount of cooked product and accelerate or slow down the system according to the amount of accumulated product.

[0023] Another aim of the invention is to eliminate the need for qualified personnel by performing functions such as heat amount, cooking amount, cooking time, rotation speed, and cutting speed by artificial intelligence software.

[0024] Another aim of the invention is to automatically identify products in subsequent cooking by recording the cooking qualities of products of different qualities and geometries that have been previously defined or cooked for the first time.

[0025] Another aim of the invention is to provide energy efficiency by preventing high energy consumption by means of the automatic instantaneous adjustment of the heat amount compared to the heat amount adjusted by eye. Another aim of the invention is to ensure food safety by preventing the uncooked product in sufficient quantities from reaching the end user.

[0026] Another aim of the invention is to increase brand reliability by delivering the right amount of cooked product to the consumer.

[0027] Another aim of the invention is to provide change warning information by calculating the lifetime comparison of cutting blades and heat generators with the obtained malfunction data and operating time calculations by artificial intelligence.

[0028] Another aim of the invention is to automate the knife sharpness detection and knife sharpening process, which is done by hand and eye control in all current systems, by using axis movements, to prevent the deterioration of knives due to incorrect sharpening, and to eliminate the operator risk in accordance with occupational safety.

[0029] The structural and characteristic features and all the advantages of the invention will be more clearly understood by means of the figures given below and the detailed description written with references to these figures, and therefore the evaluation should be made by taking these figures and the detailed explanation into consideration. Since the product that is the subject of the invention may have different geometric shapes, technical material changes and material changes, certain material names may not be mentioned.

[0030] Figures to Help Understand the Invention

[0031] Figure 1 shows a detailed perspective view of the doner kebab robot that is the subject of the invention.

[0032] Figure 2 shows the assembled front view of the doner kebab robot that is the subject of the invention. Description of Part References

[0033] A. Doner kebab robot

[0034] 1. Camera

[0035] 2. image processing computer

[0036] 3. Heater Panel

[0037] 4. Automatic gas valve

[0038] 5. Manual gas valve

[0039] 6. Rotation mechanism

[0040] 7. Movable arm

[0041] 8. Cutting blade

[0042] 9. Oil and product tank

[0043] 10. Adjustable arm

[0044] 1 1 . Control panel

[0045] 12. Scanning section

[0046] 13. Doner kebab block

[0047] The drawing are not necessarily scaled and the detailed that are not essential to understand the present invention may have been omitted. Apart from this, elements that are at least substantially identical or have at least substantially identical functions are shown with the same reference numbers.

[0048] Detailed Description of the Invention

[0049] The present invention relates to the doner kebab robot (A) that meets the above-mentioned requirements, eliminates all disadvantages and brings some additional advantages.

[0050] In this detailed explanation, the preferred embodiments of the doner kebab robot (A), which is the subject of the invention, are explained only for a better understanding of the subject. Figure 1 shows a detailed perspective view of the doner kebab robot (A) that is the subject of the invention. In the figure are the doner kebab block (13), which is placed vertically to the rotation mechanism (6) on the doner kebab robot (A), the rotation mechanism (6) that enables said doner kebab block (13) to rotate at a speed depending on the command received from the image processing computer (2), image processing computer (2), which makes sense of the instant images obtained from the camera (1 ) and matches them with predefined images, automating all cooking processes by means of the deep learning artificial intelligence software inside, the heating panel (3), which may consist of one or more sections and preferably provides heat energy by burning flammable gas, the automatic gas valve (4), which can be adjusted automatically to adjust the gas flow according to the required temperature, the manual gas valve (5), which allows instantaneous manual adjustment of gas output, especially in emergency situations, the cutting blade (8), which cuts the determined cutting section at the desired wall thickness according to the rotation speed and position data received from the image processing computer (2), the movable arm (7), which ensures efficient cutting of the circular doner kebab block (13) by providing rapid movement of said cutting blade in the horizontal (Y axis) and vertical axis (Z axis), and oil and product tank (9) where the cut product and the oils flowing during cooking are collected.

[0051] Figure 2 shows the assembled front view of the doner kebab robot (A) that is the subject of the invention. In the figure are the adjustable arm (10), which houses the image processing computer (13) and allows movement to adjust the angle of the camera (1 ), the control panel (10) containing at least one screen where production parameters are displayed and which allows manual intervention, the doner kebab block (13) obtained from solid materials such as meat and vegetables to obtain the doner kebab product, the camera (1 ) that instantly scans the surface of said doner kebab block and takes images, and the scanning section (12), which serves to separately process the corresponding scanning area for the control of each heating panel (3), depending on the number of multiple heating panels (3).

[0052] Working Principle:

[0053] Heating panels (3) are used for the cooking function of the doner kebab robot (A). In our invention, the heating panels (3) are designed to work with gas and can also work with other alternatives that produce heat energy such as electricity and natural heat sources.

[0054] Poultry meat and red meat are preferred for doner kebab production. One or more doner kebab robots can work synchronously on the production line, and machine communication can be done wired and wirelessly in accordance with industry 4.0 standards. The rotation mechanism (6) located at the midpoint of the doner kebab block (13) and the doner kebab robot (A) is placed vertically on a metal rod. When the doner kebab robot (A) starts, automatic gas valves (4) open the gas flow and start combustion. Simultaneously, the rotation mechanism (6) starts to move, ensuring that the product is cooked. During cooking, the high-resolution camera (1 ) starts to take instant images. The number of images taken per second is preferably 60 frames (fps) for efficient analysis of 24 or more moving objects. The resulting images are evaluated using artificial intelligence software powered by a deep learning algorithm. Each image obtained is analysed simultaneously and matched with the baked product data previously introduced to the system, and the cooking process continues until it is brought to a similar state. When sufficiently cooked product is detected in the image taken from the camera (1 ), the image processing computer (2) transmits the cutting command and the time it takes to start cutting depending on the rotation speed to the cutting knife (8) and the moving arm (7), which allows the knife to move on the horizontal / vertical axis. With the movement of the moving arm (7) in two axes, the doner kebab block (13) in circular form is cut efficiently with the cutting knife (8) in a proportionally defined thickness. The camera (1 ) determines the position of the cutting blade (8) by calculating the position of the doner kebab block (13) in the image it receives. When a doner kebab block (13) is heated using more than one heating panel (3), there is an automatic gas valve (4) that provides gas control for each heating panel (3). Artificial intelligence automatically divides the surface of the doner kebab block (13) corresponding to the positions of the heater panel (3) into virtual scanning sections (12) according to the number of gas valves (4). The main aim of this is to increase the heat locally where the product is undercooked, instead of heating the entire surface of the product, and to reduce the heat at the overcooked point to prevent burning and turn it off completely when necessary. In this way, non-burning or defective heating panels (3) can be detected. The cut products are delivered to the doner kebab robot (A) by entering the oil and product tank positioned with an axis tilt and proceeding on the production line to the packaging elements. It identifies the doner kebab robot (A) by analysing the images of different products, so it does not require manual adjustment as elements such as the cooking time, temperature, and rotation speed of the food product are automatically transferred to the system. Functions such as user- made connection settings, changing the production speed of the line, testing the units can be controlled and operating values can be monitored instantly via the control panel (1 1 ).

Claims

CLAIMS1. A doner kebab robot (A), which mass produces doner kebab, which is frequently consumed in the fast-food industry, and performs automatic doner kebab cooking and cutting functions by means of the use of artificial intelligence with the help of deep learning technique, comprising;- Image processing computer (2), where the volume, temperature, colour and shape values of the image data received from the camera (1 ) are processed,- At least one gas valve (4) through which the amount of gas is automatically adjusted for temperature adjustment, and- At least one camera capable of taking snapshots at least 24 frames per second (1 ).

2. The working method of the doner kebab robot (A), which performs automatic doner kebab cooking and cutting functions by means of the use of artificial intelligence, comprising the process steps of:- Sending the image taken from the camera (1 ) to the image processing computer (2),- Determining the cooking amount by comparing the resulting image with defined visuals,- Providing automatic gas valve (4) control with a potentiometer feature to increase the gas flow or resistance power in areas where there is not enough cooking in the doner kebab block (13),- Reducing the gas flow or stopping it completely by calculating the burning time in parts where the cooking amount is higher than in other areas,- Carrying out the cutting process by transmitting position data and movement time to the automatic cutting knife in order to cut the finished surface areas to the desired thickness,- Making automatic product identification in subsequent cooking by recording the cooking qualities of products of different qualities and geometries, and- Transferring the obtained data to central data units and mobile imaging devices in a wired and / or wireless manner in accordance with industry 4.0 standards for remote monitoring, analysis and control.

3. A doner kebab robot (A) according to Claim 1 , comprising automatic rotation mechanism (6) in which the rotation speed is adjusted according to the cooking data coming from the image processing computer (2).

4. A doner kebab robot (A) according to Claim 1 , comprising the virtual scanning section (12) which serves to separately process the corresponding scanning area for the control of each heating panel (3) depending on the number of multiple heating panels (3).

5. A doner kebab robot (A) according to Claim 1 , comprising a disc cutting knife (8) that can automatically cut from any angle on the surface of the doner kebab block (13) by means of its two-axis movement capability on a slide.

6. The invention relates to the doner kebab robot (A) according to claim 1 , characterized in that it comprises a digital control panel (11) including at least one display that presents the production parameters and allows manual intervention.

Images