A mechanical claw intelligent control method and system in multiple scenes

Abstract

The application relates to a multi-scene mechanical claw intelligent control method and system, and relates to the field of mechanical claw control, which comprises the following steps: collecting a cabin image, selecting a luggage case in the cabin image by a preset luggage case feature frame to obtain a luggage case image; selecting a handle in the luggage case image by a preset handle feature frame to obtain a handle image; determining a handle type according to the handle image, wherein the handle type comprises a pull handle and a rotating handle; determining a handle position in the cabin image according to the handle feature and a preset reference position, and determining a grabbing position based on the handle type and the handle position; based on the pull handle, controlling a preset mechanical hand to grab at the grabbing position and move the luggage case along a preset track, and then stacking the luggage case by a stacking method; based on the rotating handle, grabbing the rotating handle at the grabbing position by a buckling method, moving the luggage case along a track, and then stacking the luggage case by a stacking method. The application has the effect of improving the carrying efficiency.

Description

Technical Field

[0001] This invention relates to the field of robotic gripper control, and in particular to an intelligent control method and system for robotic grippers in multiple scenarios. Background Technology

[0002] With the rapid development of the air transport industry, travel baggage and cargo transportation involve multiple links such as security checks, conveying, transfer and loading, and the demand for automation and intelligence of airport baggage handling systems is increasing day by day.

[0003] After the plane arrives at its destination, in order to retrieve the luggage from the cabin, manual labor is required to move the luggage out of the cabin. Then, other personnel will carry the luggage to a transport vehicle, which will then travel on a conveyor belt to complete the luggage handling process.

[0004] The process of manually retrieving luggage, sending it to the aircraft door via conveyor belt, and then transferring it from the aircraft door to the transport vehicle involves manual labor. The large volume of luggage leads to high handling intensity, and the repetitive, monotonous movements can easily cause fatigue, resulting in low luggage handling efficiency and room for improvement. Summary of the Invention

[0005] To improve handling efficiency, this invention provides a method and system for intelligent control of robotic grippers in multiple scenarios.

[0006] Firstly, the present invention provides an intelligent control method for a robotic gripper in multiple scenarios, employing the following technical solution:

[0007] A method for intelligent control of a robotic gripper in multiple scenarios includes:

[0008] S10: Acquire cabin images, and select the suitcases in the cabin images using a preset suitcase feature box to obtain suitcase images;

[0009] S2: Select the handle in the suitcase image using a preset handle feature box to obtain a handle image;

[0010] S3: Determine the handle type based on the handle image, wherein the handle type includes pull handles and rotary handles;

[0011] S4: Determine the handle position in the cabin image based on the handle features and the preset reference position, and determine the gripping position based on the handle type and the handle position;

[0012] S40: Based on the pull handle, control the preset robotic arm to grasp the suitcase at the grasping position and move it along the preset trajectory, and then stack it according to the preset stacking method;

[0013] S41: Based on the rotating handle, the rotating handle is gripped at the gripping position using a preset gripping method, and the suitcase is moved along a preset trajectory, and then stacked using a preset stacking method.

[0014] By adopting the above technical solution, cabin images are collected, and the corresponding grasping and stacking methods are determined by identifying the type of suitcase and its handle. This effectively improves the automation level and efficiency of baggage handling, while reducing the intensity and error rate of manual operation and improving handling efficiency.

[0015] Optionally, when the cabin image does not frame the view of the luggage compartment, the method further includes:

[0016] S11: The baggage is selected by a preset baggage feature frame in the cabin image, and the selected image is defined as a baggage image;

[0017] S110: Determine the shape and location of the luggage based on the luggage image and a preset reference position;

[0018] S111: Determine the grab point based on the shape and position of the luggage;

[0019] S112: Control the preset robotic arm to move to the gripping point, grip with a preset gripping force, and rotate at a preset angle to continuously collect images of luggage resistance and gripping point.

[0020] S113: Obtain the variation curve based on the luggage resistance fitting;

[0021] S114: Determine the luggage status based on the change curve, the luggage status including an air-filled state and a tight-fitting state;

[0022] S1140: If the luggage is in the tight-fitting state, determine the degree of gripping deformation based on the luggage resistance and the gripping point image;

[0023] S11400: Control the preset robotic arm to reset and collect the recovery time of the gripping point;

[0024] S11401: Determine the luggage material based on the degree of grasping deformation and the recovery time;

[0025] S115: Determine the clamping gap based on the luggage material;

[0026] S116: Control the preset robotic arm to grab the luggage at the gripping point with the clamping gap and move it along a preset trajectory, and then stack it using a preset stacking method.

[0027] By adopting the above technical solution, when no suitcase is detected, it indicates that it is a backpack or similar type of luggage. The luggage image obtained by selecting the luggage features determines the gripping point, and the luggage state is determined based on the degree of gripping deformation, thereby judging the luggage material. Then, the clamping gap is adjusted to achieve adaptive gripping of luggage of different materials and improve gripping efficiency.

[0028] Optionally, after S114 and before S115, the following may also be included:

[0029] S1141: If the luggage is in the empty state, determine the lifting point based on the grasping point image;

[0030] S11410: Determine the lifting force based on the change curve and the preset gripping force matching;

[0031] S11411: Control the preset robotic arm to move to the lifting point, lift with the lifting force, and collect the elasticity value during the lifting process to obtain the elasticity curve;

[0032] S11412: Determine the luggage material based on the elasticity curve.

[0033] By adopting the above technical solution, for luggage in a hollow state, the lifting point is determined based on the grasping point image, and then the lifting force is matched to control the robotic arm to lift, so as to avoid damaging the luggage. The elasticity value is collected to obtain the elasticity curve to identify the luggage material, thereby improving the accuracy of luggage handling.

[0034] Optionally, after controlling the preset robotic arm to grasp the luggage at the gripping point with the clamping gap and move along a preset trajectory, the system further includes:

[0035] S117: Collect baggage weight value;

[0036] S118: If the weight of the luggage is greater than the preset benchmark weight, the luggage handle is selected in the luggage image by using the preset luggage handle feature, and the selected image is defined as the luggage handle image;

[0037] S119: Determine the clamping position based on the luggage handle image and the luggage handle features;

[0038] S120: Match the clamping force based on the luggage weight value;

[0039] S121: Control another preset robotic arm to move to the gripping position to grip the luggage handle with gripping force, and then move along a preset trajectory after gripping and lifting the luggage.

[0040] By adopting the above technical solution, the baggage is grabbed to collect the baggage weight value. When the baggage is overweight, the baggage handle image is identified and the clamping position is determined. The handle is positioned to match the clamping force, and another robotic arm is controlled to cooperate in clamping and lifting. This adapts to baggage of different weights, avoids the possibility of the baggage falling, and improves the stability and efficiency of handling.

[0041] Optionally, the stacking method includes:

[0042] S5: Acquire placement images at preset placement positions;

[0043] S51: Determine the stacking height and stacking space based on the placement image;

[0044] S52: Obtain the current handling volume based on the cabin image, and determine the stacking position based on the current handling volume and the stacking space;

[0045] S53: Calculate the number of stacking layers based on the stacking height and the current handling volume;

[0046] S54: Match the preset adjustment height required by the robotic arm according to the number of stacked layers;

[0047] S55: Control the preset robotic arm to stack sequentially according to the stacking position and the adjusted height.

[0048] By adopting the above technical solution, the stacking height and stacking space are determined by acquiring placement images, and the handling volume is obtained by combining them with cabin images. This allows for the determination of the stacking position and subsequent stacking, enabling reasonable planning of the stacking layout, optimization of space utilization, ensuring neat stacking and structural stability, and improving stacking efficiency.

[0049] Optionally, after determining the stacking position, the following may also be included:

[0050] S6: Determine the luggage density based on the luggage weight value and the current handling volume;

[0051] S60: When the luggage density is less than the preset leveling density, control the preset robotic arm to level the luggage at the stacking position with a preset leveling force;

[0052] S61: When the baggage density is not less than the preset leveling density, determine the baggage shape based on the cabin image;

[0053] S62: Determine the maximum and minimum values ​​of the three dimensions of the luggage based on its shape;

[0054] S63: Match the placement volume according to the three-dimensional extreme values;

[0055] S64: Determine a correction position based on the placement volume, the stacking space, and the luggage material, and replace the stacking position with the correction position;

[0056] S65: Control the preset robotic arms to stack sequentially at the corrected position and the adjusted height.

[0057] By adopting the above technical solution, the luggage density is obtained by the luggage weight value and the current handling volume. When the luggage density is less than the leveling density, it means that there is still remaining space inside the luggage that can be leveled, reducing the space occupied by the luggage and optimizing the space layout. When the luggage density is not less than the leveling density, the three-dimensional maximum and minimum values ​​are determined according to the shape of the luggage, and then the corrected position is determined for stacking, thereby improving the space utilization rate of luggage stacking.

[0058] Optionally, when the pull handle is grasped, the grasping method includes:

[0059] S400: Determine the shape and volume of the pull handle based on the handle image;

[0060] S401: Determine the handle width based on the handle volume;

[0061] S402: Determine the opening size during gripping based on the handle width;

[0062] S403: Determine the gripping angle based on the shape of the pull-out handle;

[0063] S404: Control the preset robotic arm to move to the grasping position and open it to the specified opening size, and grasp the pull handle at the specified grasping angle to complete the grasping operation.

[0064] By adopting the above technical solution, the shape and volume of the handle are determined by the handle image, and then the handle width, gripping opening size and gripping angle are determined. The robotic arm is controlled to grip the pull handle with the opening size and gripping angle, ensuring a firm grip, reducing gripping errors, and improving the gripping accuracy and reliability of the robotic claw.

[0065] Optional, preset extraction methods include:

[0066] S410: Determine the deflection position of the rotating handle based on the handle image;

[0067] S411: Determine the flipping direction based on the aforementioned bias position;

[0068] S412: Determine the clamping angle based on the gripping position and the handle width;

[0069] S413: Determine the shape of the rotating handle based on the handle image;

[0070] S414: Determine the lifting distance based on the shape of the rotating handle;

[0071] S415: At the gripping position, the rotating handle is pulled out at the gripping angle and lifted by the lifting distance to complete the gripping.

[0072] By employing the above technical solution, the bias position and flipping direction of the rotating handle are determined based on the handle image to facilitate flipping and gripping. The gripping angle is determined by combining the handle position and handle width, and the lifting distance is determined according to the shape of the rotating handle, thereby completing the gripping of the rotating handle. This operation enhances the mechanical gripper's ability to grasp handles with complex shapes and improves the flexibility and adaptability of the operation.

[0073] Optionally, when the suitcase moves along a preset trajectory, it also includes:

[0074] S7: Collect the weight value of the suitcase when it is being moved;

[0075] S70: If the weight value is greater than the preset reference weight value, then the side handle is selected based on the luggage image using a preset side handle feature box, and the position of the selected side handle is defined as the side handle position.

[0076] S71: Update the suitcase image based on the side handle position and define it as a side handle image;

[0077] S72: Determine the curvature of the side handle based on the side handle image;

[0078] S73: Match the gripping shape based on the curvature of the side handle;

[0079] S74: Control another preset robotic arm to move to the side handle position and complete the gripping of the side handle in the gripping form to complete the adjustment.

[0080] By adopting the above technical solution, the weight value is collected when the suitcase is moved. When it is overweight, the side handle is selected by framing the suitcase image, the curvature of the side handle is determined and the gripping shape is matched, and the robotic arm is controlled to complete the gripping of the side handle, so as to ensure the safety and stability of luggage handling.

[0081] Secondly, this application provides an intelligent control system for a robotic gripper in multiple scenarios, employing the following technical solution:

[0082] A multi-scenario intelligent control system for robotic grippers includes:

[0083] The acquisition module is used to acquire cabin images;

[0084] The memory is used to store the program of a smart control method for a robotic gripper in multiple scenarios;

[0085] The processor loads and executes programs from memory.

[0086] In summary, this application includes at least one of the following beneficial technical effects:

[0087] 1. By collecting cabin images and identifying the type of suitcase and its handle, the corresponding grasping and stacking methods are determined, which effectively improves the automation level and efficiency of baggage handling, while reducing the intensity and error rate of manual operation and improving handling efficiency.

[0088] 2. When no suitcase is detected, it indicates that it is a backpack or similar type of luggage. The luggage image obtained by selecting the luggage feature box determines the gripping point. The luggage status is determined based on the degree of gripping deformation, thereby determining the luggage material. Then, the clamping gap is adjusted to achieve adaptive gripping of luggage of different materials and improve gripping efficiency.

[0089] 3. Obtain the luggage density by combining the luggage weight and current handling volume. For luggage made of elastic material with a density less than the leveling density, it indicates that there is still remaining space inside the luggage that can be leveled, reducing the space occupied by the luggage and optimizing the space layout. When the luggage density is not less than the leveling density, determine the three-dimensional maximum and minimum values ​​based on the shape of the luggage to determine the correction position for stacking, thereby improving the space utilization rate of luggage stacking. Attached Figure Description

[0090] Figure 1 This is a flowchart of a method for intelligent control of a robotic gripper in multiple scenarios according to an embodiment of the present invention;

[0091] Figure 2 This is a flowchart of the luggage stacking method according to an embodiment of the present invention. Figure 1 ;

[0092] Figure 3 This is a flowchart of the luggage stacking method according to an embodiment of the present invention. Figure 2 ;

[0093] Figure 4 This is a flowchart of the method for determining the correction position according to an embodiment of the present invention. Detailed Implementation

[0094] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0095] This application discloses a multi-scenario intelligent control method for robotic grippers. By classifying and processing luggage and suitcases in the cabin, the method distinguishes the handle type based on the suitcase for easy handling, distinguishes the material of the luggage for targeted handling, and obtains different stacking positions based on different materials, thereby improving stacking efficiency.

[0096] Reference Figure 1 A method for intelligent control of a robotic gripper in multiple scenarios includes the following steps:

[0097] S10: Acquire cabin images, and select the suitcases in the cabin images using preset suitcase feature boxes to obtain suitcase images.

[0098] Cabin images refer to image data at the aircraft cabin door. The images show luggage or suitcases moving to the cabin door via a conveyor belt, captured by cameras pre-installed on the equipment. These cameras are pre-set on the robot by technicians based on the actual situation, and will not be elaborated on here.

[0099] Luggage features refer to data on the appearance characteristics of the luggage, such as its color and shape. These are pre-set by technicians based on the actual situation and will not be elaborated here.

[0100] The robot is a device equipped with robotic arms on both the left and right sides, and the bottom is a movable part, which will not be described in detail here.

[0101] Luggage images refer to partial images with luggage features identified and extracted from cabin images.

[0102] By annotating a large number of images with luggage features, the labeled images are input into the YOLO large model. The luggage features are extracted by repeatedly stacking the images using a PyTorch network architecture. The error between the results and the data is calculated. When the error is less than 1%, the model is put into use.

[0103] The cabin image is input into the large model. When the luggage features are identified, the luggage features are marked and selected from the cabin image. The luggage features are then identified in the selected image, which is the luggage image.

[0104] S2: Select the handle in the suitcase image using a preset handle feature box to obtain the handle image.

[0105] Handle features refer to the images and related data of the appearance features of the suitcase handle, such as the shape, color, and position of pull-out handles and rotary handles. These are preset by technicians according to the actual situation and will not be elaborated here.

[0106] Handle images refer to images from which handle features are identified and extracted.

[0107] By annotating a large number of images with handle features, and then inputting these annotated images into the YOLO large model, the handle features are extracted by repeatedly stacking the images using a PyTorch network architecture. The error between the results and the data is calculated, and the model is put into use when the error is less than 1%.

[0108] The suitcase image is input into the large model. When the handle feature is detected, the handle feature is marked and selected from the suitcase image. The handle feature is then identified in the selected image, which is the handle image.

[0109] S3: Determine the handle type based on the handle image. Handle types include pull handles and rotary handles.

[0110] Handle type refers to the type of luggage handle, the most common being pull-out handles and rotary handles.

[0111] The pull handle, which is lifted by pulling, is located on the suitcase.

[0112] The handle on the suitcase is located on a rotating surface.

[0113] The handle features include the appearance shape of pull handles and rotary handles. When the handle image is obtained by selecting the handle features, the handle type corresponding to the handle features is known. Then, the handle type can be determined based on the handle image.

[0114] S4: Determine the handle position in the cabin image based on the handle features and the preset reference position, and determine the gripping position based on the handle type and handle position.

[0115] A reference position is a standard reference location used to determine the specific location of an image in a real-world scene. The exact location in the image can be obtained by using the reference position and a preset camera position. This is pre-set by technicians according to the actual situation and will not be elaborated upon here.

[0116] The handle position refers to the specific coordinates of the handle on the luggage compartment in the cabin image.

[0117] The gripping position refers to the specific location where the robotic arm grips the suitcase.

[0118] The handle is located by selecting a bounding box within the cabin image based on its features.

[0119] Different handle types correspond to different gripping positions. For pull handles, the handle position is the gripping position; for rotary handles, the gripping position and the handle position are separated by a preset offset distance. Moving the handle position in the preset offset direction by the preset offset distance is the gripping position. The offset distance is preset by the technician according to the actual situation and will not be elaborated here.

[0120] S40: Based on the pull handle, control the preset robotic arm to grasp the suitcase at the grasping position and move it along the preset trajectory, and then stack it according to the preset stacking method.

[0121] The mechanical fingers are devices used to grasp and move suitcases and luggage. They are two mechanical arms pre-installed on the robot. By simulating the movements of human arms, they can achieve automated grasping, handling, and assembly functions. They are pre-programmed by technicians according to the actual situation and will not be described in detail here.

[0122] The trajectory refers to the path the robotic arm takes when moving the suitcase and placing it on the transport vehicle. It is pre-set by technicians based on the actual situation and will not be elaborated here.

[0123] Stacking methods refer to the methods of stacking suitcases and luggage according to size and volume. For specific methods, refer to S5 to S55.

[0124] Based on the pull handle, the robotic arm is controlled to grasp the suitcases at the gripping position, and then the suitcases are moved along a trajectory and stacked.

[0125] S41: Based on the rotating handle, the rotating handle is gripped at the gripping position using a preset gripping method, and the suitcase is moved along a preset trajectory, and then stacked using a preset stacking method.

[0126] The gripping method refers to the method by which the robotic arm grasps the rotating handle. For specific methods, refer to S410 to S415.

[0127] Based on the rotating handle, the handle is engaged at the gripping position to grab the suitcase, and the suitcase is moved along a trajectory to the placement position for stacking.

[0128] When the cabin image does not include the luggage compartment image, it also includes:

[0129] S11: Select baggage from the cabin image using preset baggage features, and define the selected image as a baggage image.

[0130] Baggage characteristics refer to data on the appearance features of baggage, such as the shape and color of different baggage. These are preset by technicians according to the actual situation and will not be elaborated here.

[0131] Baggage images refer to images from which baggage features are identified and extracted.

[0132] By annotating a large number of images with baggage features, the baggage features are extracted by repeatedly stacking images using a PyTorch network architecture. The error between the results and the data is calculated. When the error is less than 1%, the model is put into use.

[0133] The cabin image is input into the large model. When baggage features are identified, the baggage features are marked and selected from the cabin image. The baggage features are then identified in the selected image, which is the baggage image.

[0134] S110: Determine the shape and location of luggage based on luggage images and preset reference locations.

[0135] Luggage shape refers to the external outline of luggage.

[0136] Baggage location refers to the specific location of the baggage in the cabin image.

[0137] The luggage features include the luggage's shape. Once the luggage image is selected, the corresponding luggage shape can be determined.

[0138] Since the reference position is a known fixed reference, the position of the luggage in the image is the position of the luggage image obtained by selecting the reference position. With the help of the reference position, the specific position of the image in the actual scene can be determined, and the position of the luggage in the image can also be calculated to determine the specific position of the luggage in the actual scene, that is, the luggage position.

[0139] S111: Determine the grab point based on the shape and location of the luggage.

[0140] The gripping point refers to the specific location on the luggage when the robotic arm grips it.

[0141] Different luggage shapes at different luggage locations correspond to different grab points. Grab points are obtained by inputting the luggage shape and luggage location into a preset grab point database. The grab point database is a database that is preset by technicians according to the actual situation. The grab point database contains the relationship between luggage shape and luggage location and grab points. The actual parameters are preset by technicians according to the actual situation, which will not be elaborated here.

[0142] S112: Control the preset robotic arm to move to the gripping point, grip with the preset gripping force, and rotate at the preset angle to continuously collect images of luggage resistance and gripping point.

[0143] The gripping force refers to the force applied by the robotic arm to keep the luggage stable when gripping it. It is preset by technicians according to the actual situation and will not be elaborated here.

[0144] The angle refers to the angle at which the luggage can achieve the ideal deformation when rotated. It is preset by technicians according to the actual situation and will not be elaborated here.

[0145] Baggage resistance refers to the force exerted by the baggage during the rotation of the robotic arm to prevent it from rotating and to maintain its original shape. Baggage resistance is collected by mechanical sensors pre-set on the robotic arm. These mechanical sensors are pre-set by technicians and will not be described in detail here.

[0146] The capture point image refers to the image data at the capture point.

[0147] The robotic arm is controlled to move to the gripping point, grip with gripping force, and rotate, while continuously collecting images of luggage resistance and the gripping point.

[0148] S113: The variation curve is obtained by fitting the baggage resistance.

[0149] The variation curve refers to the curve of how baggage resistance changes over time. The variation curve is obtained by constructing a coordinate system using the magnitude of baggage resistance and time collected by the mechanical sensor as the x-axis and y-axis, respectively.

[0150] S114: Determine the baggage status based on the change curve. The baggage status includes the state of being airy and the state of being tightly packed.

[0151] Luggage condition refers to the state of the items inside the luggage, including whether they are empty or tightly packed.

[0152] A "hollow" state refers to a state where there is empty space inside the luggage or it is not tightly packed.

[0153] "Closely packed" refers to a state where the interior of the luggage is tightly packed and relatively full.

[0154] Different change curves correspond to different baggage states. The change curves are input into the preset state database to obtain the corresponding baggage state. The state database is a database that is preset by technicians according to the actual situation. The capture point database contains the relationship between the change curves and the baggage shape. The actual parameters are preset by technicians according to the actual situation, which will not be elaborated here.

[0155] S1140: If the luggage is in a tightly attached state, determine the degree of gripping deformation based on the luggage resistance and the gripping point image.

[0156] Grab deformation refers to the degree of deformation of the surface material of the luggage during rotation.

[0157] The degree of gripping deformation is obtained by inputting the gripping point image and baggage resistance into the deformation database. The deformation database contains the relationship between the gripping point image, baggage resistance and the degree of gripping deformation. The actual parameters are set in the deformation database in advance by technicians, who set the gripping point image and baggage resistance corresponding to different degrees of gripping deformation.

[0158] If the luggage is in a tightly packed state, it indicates that the luggage is relatively full. The degree of gripping deformation can be determined by the luggage resistance and the gripping point image.

[0159] S11400: Controls the preset robot arm reset and collects the recovery time of the grasping point.

[0160] Recovery time refers to the time required for the surface of luggage to return to its initial state after being grabbed and rotated.

[0161] The recovery time is calculated by taking the time from the start of the grabbing rotation until the grabbing point image matches the luggage image. This recovery time is obtained by comparing the grabbing point image acquired after the grabbing is completed using a preset device.

[0162] S11401: Determine the material of luggage based on the degree of grasping deformation and recovery time.

[0163] Luggage material refers to the type of material used in luggage, including elastic and non-elastic materials.

[0164] Elastic materials refer to materials with a certain degree of elasticity, such as fabrics used for luggage.

[0165] Non-elastic materials refer to materials that do not have elasticity, such as plastic, metal, and other luggage fabrics.

[0166] For the same degree of grasping deformation, a shorter recovery time indicates an elastic material, while a longer recovery time indicates a non-elastic material. For the same recovery time, a smaller degree of grasping deformation indicates an elastic material, while a longer recovery time indicates a non-elastic material.

[0167] The system pre-stores the correspondence between the degree of grasping deformation and recovery time and the luggage material, so the corresponding luggage material can be obtained based on the degree of grasping deformation and recovery time.

[0168] S115: Determine the clamping gap based on the material of the luggage.

[0169] Grip gap refers to the distance between the two grippers when a robotic arm grips luggage.

[0170] Elastic materials require a larger clamping interval, while non-elastic materials require a smaller clamping interval. The system has a pre-stored correspondence between clamping intervals and luggage materials, so the corresponding clamping interval can be obtained based on the luggage material.

[0171] S116: Control the preset robotic arm to grab the luggage at the gripping point with a clamping gap and move it along a preset trajectory, and then stack it using a preset stacking method.

[0172] The robotic arm is controlled to grip luggage at gripping intervals at the gripping point to ensure the stability of the gripping, and then the luggage is moved and stacked along a trajectory.

[0173] The following are included after S114 and before S115:

[0174] S1141: If the luggage is in an empty state, determine the lifting point based on the grab point image.

[0175] The lifting point refers to the center position where the robotic arm lifts the luggage.

[0176] The center point of the luggage is identified in the grasp point image, which is the lifting point. By capturing a large number of grasp point images with lifting points and labeling the lifting points, the labeled grasp point images are input into a large model. The lifting points are extracted by repeatedly stacking images using a PyTorch network architecture, and the error between the result and the data is calculated. When the error is less than 1%, it is put into use.

[0177] The image of the grasping point is input into the large model. When the lifting point is identified, the lifting point is marked, thus obtaining the lifting point.

[0178] When the baggage is in a hollow state, it is grabbed at the grab point. Since the baggage is empty inside, the material of the baggage cannot be obtained. Therefore, the material of the baggage is determined by lifting it.

[0179] S11410: Determines the lifting force based on the change curve and the preset gripping force matching.

[0180] Lifting force refers to the force applied by a robotic arm when lifting luggage.

[0181] Different elasticity curves correspond to different elasticity of luggage materials, and the initial gripping force is the baseline value; the elasticity curves and the correspondence between gripping force and lifting force are pre-stored, so the corresponding lifting force can be obtained based on the elasticity curves and gripping force.

[0182] S11411: Control the preset robotic arm to move to the lifting point, lift with lifting force, and collect the elasticity value during the lifting process to obtain the elasticity curve.

[0183] Elasticity value refers to the real-time elasticity value detected by the mechanical sensor during the lifting process.

[0184] The elasticity curve refers to the curve of elasticity value changing with time. The elasticity curve is obtained by constructing a coordinate system by using the magnitude of the elasticity value collected by the mechanical sensor and the time as the x-axis and y-axis, respectively.

[0185] The robotic arm is controlled to move to the lifting point, and the lifting force is applied to lift the object, while the elasticity value is collected to obtain the elasticity curve.

[0186] S11412: Determine the material of luggage based on the elasticity curve.

[0187] Different elasticity curves correspond to different luggage materials. The smoother the elasticity curve, the less elastic the luggage material. The correspondence between elasticity curves and luggage materials is stored in advance, so the corresponding luggage material can be obtained based on the elasticity curve.

[0188] After controlling the pre-set robotic arm to grasp the luggage at the gripping point with a clamping gap and move it along a pre-set trajectory, the process also includes:

[0189] S117: Collect baggage weight values.

[0190] Baggage weight refers to the weight data of the baggage, which is obtained by collecting the weight of the baggage during movement through mechanical sensors.

[0191] S118: If the weight of the luggage is greater than the preset baseline weight, select the luggage handle in the luggage image using the preset luggage handle feature, and define the selected image as the luggage handle image.

[0192] The baseline weight value refers to the preset standard weight value used to determine whether luggage is overweight. It is preset by technicians based on the actual situation and will not be elaborated here.

[0193] Luggage handle features refer to data on the appearance characteristics of luggage handles, such as the shape, color, and position of the luggage handles. These are preset by technicians based on the actual situation and will not be elaborated here.

[0194] Luggage handle images refer to images from which the features of luggage handles are identified and extracted.

[0195] By annotating a large number of images with luggage handle features, the system inputs these labeled images into the YOLO large model. The system then uses a PyTorch network architecture to repeatedly stack the images to extract the luggage handle features. The system calculates the error between the results and the data, and when the error is less than 1%, the system is put into use.

[0196] The luggage image is input into the large model. When the luggage features are recognized, the luggage features are marked and selected from the luggage image. The luggage handle features are then identified in the selected image, which is the luggage handle image.

[0197] S119: Determine the clamping position based on the baggage handle image and baggage handle features.

[0198] The gripping position refers to the specific position where the robotic arm grips the luggage handle.

[0199] The luggage handle features are used to select a bounding box in the luggage image. The position of the bounding box in the luggage handle image is the clamping position.

[0200] S120: Matches clamping force based on luggage weight.

[0201] Clamping force refers to the force applied by a robotic arm when gripping luggage.

[0202] The greater the weight of the luggage, the greater the corresponding clamping force. The system has a pre-stored correspondence between luggage weight and clamping force, so the corresponding clamping force can be obtained based on the luggage weight.

[0203] S121: Control another preset robotic arm to move to the gripping position to grip the luggage handle with gripping force, and then move along a preset trajectory after gripping and lifting the luggage.

[0204] Control another robotic arm to move to the gripping position and clamp the luggage handle with clamping force to make the luggage movement process more stable. After clamping and lifting the luggage, it moves along a trajectory.

[0205] Stacking methods include:

[0206] S5: Capture the placement image at the preset placement location.

[0207] The placement location refers to where the luggage is placed, i.e., the location of the transport vehicle. This is pre-set by technicians based on the actual situation and will not be elaborated on here.

[0208] The placement image refers to the image data captured by the camera at the placement location.

[0209] S51: Determine the stack height and stack space based on the placement image.

[0210] Stacking height refers to the maximum height to which luggage can be stacked.

[0211] Stacking space refers to the area where luggage can be stacked.

[0212] Determining the stack height and stack space by placing images is common knowledge known to those skilled in the art, and will not be elaborated upon here.

[0213] S52: Obtain the current handling volume based on the cabin image, and determine the stacking position based on the current handling volume and stacking space.

[0214] Current handling volume refers to the volume of luggage that needs to be handled at this time.

[0215] Stacking location refers to the specific position where luggage is stacked on the transport vehicle.

[0216] The current transported volumes are stacked sequentially in the stacking space from low to high and from front to back. After each stacking is completed, the cabin image is updated, and the stacking position is redefined for further stacking.

[0217] Determining the current handling volume using cabin images and determining the stacking position based on the current handling volume and stacking space are common knowledge known to those skilled in the art, and will not be elaborated here.

[0218] S53: Calculate the number of stacking layers based on the stacking height and the current handling volume.

[0219] The stacking layer number refers to the maximum number of layers that luggage can be stacked. It is obtained by dividing the stacking height by the current transport volume to get the current height of the luggage.

[0220] S54: Matches the preset adjustment height required by the robot arm based on the number of stacked layers.

[0221] Adjusting the height refers to the height that a single layer needs to be adjusted when the robotic arm is stacking.

[0222] When the number of layers changes, the height of the corresponding robotic arm also needs to be adjusted. The number of stacked layers and the required adjustment height are in one-to-one correspondence. The system has a pre-stored correspondence between the number of stacked layers and the adjustment height, so the corresponding adjustment height can be obtained based on the number of stacked layers.

[0223] S55: Controls the preset robotic arm to stack items sequentially according to the stacking position and adjusted height.

[0224] The robotic arm is controlled to stack items sequentially according to their stacking position and height. After stacking is completed, the cabin image is updated, and steps S5 to S55 are repeated to stack the next piece of luggage or suitcase.

[0225] For example, if there are a total of 3 stacked layers, the robotic arm will lift and adjust the height after each layer is stacked.

[0226] Following the determination of the stacking position, the following also applies:

[0227] S6: Determine baggage density based on baggage weight and current handling volume.

[0228] Baggage density refers to the average density of baggage as a whole.

[0229] Baggage density is calculated by the ratio of baggage weight to the current handling volume.

[0230] S60: When the luggage density is less than the preset leveling density, control the preset robotic arm to level the luggage in the stacked position with the preset leveling force.

[0231] The leveling density refers to the preset standard density, which is used to determine whether the luggage needs to be leveled. It is preset by technicians according to the actual situation and will not be elaborated here.

[0232] The leveling force refers to the force applied by the robotic arm when leveling luggage. It is preset by technicians according to the actual situation and will not be elaborated here.

[0233] When the luggage density is less than the leveling density, it means that the items inside the luggage are made of lighter materials, such as clothes. At the same time, the luggage is made of elastic material and can be leveled. Then, the robotic arm is controlled to level the elastic material luggage in the stacked position with leveling force, thereby improving space utilization.

[0234] S61: When the baggage density is not less than the preset leveling density, determine the baggage shape based on the cabin image.

[0235] Baggage shape refers to the outer outline of the baggage. Baggage features include the shape of the baggage. When the baggage image is selected, the corresponding baggage shape can be determined. The baggage image is extracted from the cabin image.

[0236] When the luggage density is not less than the preset leveling density, it means that the items in the luggage are made of hard materials or there are too many items, making it impossible to level them. The current handling volume is not suitable and may easily cause damage to the luggage. The space volume needs to be re-divided.

[0237] S62: Determine the maximum and minimum values ​​of the three dimensions of the luggage based on its shape.

[0238] The three-dimensional extrema refer to the maximum values ​​of luggage in the three dimensions of length, width, and height.

[0239] The method of obtaining the three-dimensional maximum and minimum values ​​of the corresponding luggage by combining the luggage shape with the representation and geometric calculation logic is common knowledge known to those skilled in the art, and will not be elaborated here.

[0240] S63: Match placement volume based on 3D extrema.

[0241] Placement volume refers to the volume allocated to each piece of luggage based on the maximum and minimum values ​​of its three dimensions.

[0242] The maximum and minimum values ​​of the three dimensions are used as the length, width, and height of the volume to divide it, and the placement volume is calculated by multiplying the length, width, and height.

[0243] S64: Determine the correction position based on the placement volume, stacking space, and luggage material, and replace the stacking position with the correction position.

[0244] The corrected position refers to the adjusted stacking position.

[0245] Based on waterproof baggage, the current handling volume is obtained from the cabin image. The waterproof position is determined based on the current handling volume and stacking space. Based on water-resistant baggage, the placement image of the placement position is updated to determine the placement volume. The shielding position is determined based on the updated placement image and placement volume. The water-resistant position of the water-resistant baggage is determined based on the current handling volume and shielding position. When waterproof baggage is picked up, stacking begins, and the corrected position is changed to the waterproof position. After stacking is completed, the corrected position is updated to the water-resistant position. The corrected positions are changed and updated sequentially until stacking is completed.

[0246] S65: Control the preset robotic arm to correct the position and adjust the height to stack sequentially.

[0247] The robotic arms are controlled to adjust the position and height of the luggage and stack it in sequence, thereby rationally planning the space according to the different types of luggage and ensuring the safe transportation of luggage.

[0248] When gripping a pull-out handle, the gripping methods include:

[0249] S400: Determine the handle shape and handle volume based on the handle image.

[0250] The shape of a pull-out handle refers to its geometric shape.

[0251] The handle volume refers to the three-dimensional volume of the handle.

[0252] The method of determining the shape and volume of a pull handle based on a handle image is common knowledge known to those skilled in the art and will not be elaborated here.

[0253] S401: Determine the handle width based on the handle volume.

[0254] Handle width refers to the width of the handle.

[0255] If the volume of the handle is known, then the corresponding width dimension in the width direction is also known; the width dimension is the handle width.

[0256] S402: Determine the opening size when gripping based on the handle width.

[0257] The opening size refers to the distance that the robotic arm needs to open when grasping, in order to accommodate the width of the handle.

[0258] The handle width is known. The opening size of the robotic arm is obtained by adding the handle width to the preset opening distance. The opening distance is preset by the technicians according to the actual situation and will not be elaborated here.

[0259] S403: Determine the gripping angle based on the shape of the pull-out handle.

[0260] The gripping angle refers to the angle that the robotic arm needs to adjust during gripping to fit the shape of the pull handle.

[0261] Different pull handle shapes correspond to different gripping angles. The system has a pre-stored correspondence between pull handle shapes and gripping angles, so the corresponding gripping angle can be matched based on the pull handle shape.

[0262] S404: Control the preset robotic arm to move to the gripping position and open it to the desired size, then grip the pull handle at the desired angle to complete the gripping operation.

[0263] The robotic arm is controlled to move to the gripping position and open to the desired size. Then, it extends into the pull handle to grip at the appropriate angle, completing the gripping operation and facilitating subsequent handling.

[0264] The preset extraction methods include:

[0265] S410: Determine the deflection position of the rotating handle based on the handle image.

[0266] The offset position refers to the position of the handle on the suitcase. The handle can be offset to the upper or lower side. The direction of the offset can be determined based on the handle image, and thus the offset position can be determined. The specific operation method is common knowledge known to those skilled in the art, and will not be described in detail here.

[0267] S411: Determine the flip direction based on the bias position.

[0268] The flip direction refers to the direction that the robotic arm needs to flip when grasping.

[0269] Once the position of the bias is determined, the direction of the bias is also determined, and the direction opposite to the position of the bias is the direction of reversal.

[0270] S412: Determine the clamping angle based on the gripping position and handle width.

[0271] The gripping angle refers to the angle at which the robotic arm grips and rotates the handle.

[0272] Different gripping positions correspond to different clamping angles. The wider the handle, the larger the clamping angle. The clamping angle is obtained by inputting the gripping position and handle width into the clamping database. The clamping database contains images of gripping points and the relationship between luggage resistance and clamping angle. The actual parameters are set in advance by technicians in the clamping database, with the gripping position and handle width corresponding to different clamping angles.

[0273] S413: Determine the handle shape based on the handle image.

[0274] The shape of the rotating handle refers to the geometry of the handle.

[0275] The method of determining the shape of the rotating handle based on the handle image is common knowledge known to those skilled in the art, and will not be elaborated here.

[0276] S414: Determine the lifting distance based on the shape of the rotating handle.

[0277] Lifting distance refers to the distance that the robotic arm needs to lift after grasping the suitcase to ensure that it can be moved smoothly.

[0278] Different handle shapes correspond to different lifting distances. The system has a pre-stored correspondence between handle shapes and lifting distances, so the corresponding lifting distance can be obtained based on the handle shape.

[0279] S415: Pull out the rotating handle at the gripping position at the gripping angle and lift it by the lifting distance to complete the gripping.

[0280] The robotic arm is positioned at the gripping position, with the rotating handle pulled out at the gripping angle, and then lifted by a lifting distance to complete the gripping, thus ensuring the adaptability of the robotic arm.

[0281] When the suitcase moves along a preset trajectory, it also includes:

[0282] S7: Collect the weight value of the suitcase when it is being moved.

[0283] The weight value refers to the weight and size of the suitcase, which is obtained in real time through preset mechanical sensors.

[0284] S70: If the weight value is greater than the preset baseline weight value, the side handle is selected based on the luggage image using the preset side handle feature box, and the position of the selected side handle is defined as the side handle position.

[0285] The baseline weight value refers to the preset standard weight value, which is used to determine whether additional gripping operations are needed to ensure the stability of the handling. It is preset by technicians according to the actual situation and will not be elaborated here.

[0286] Side handle features refer to the data on the appearance of the side handle of the suitcase, such as the shape, color, and position of the side handle. These are pre-set by technicians based on the actual situation and will not be elaborated here.

[0287] The side handle position refers to the specific location of the side handle in the image of the suitcase.

[0288] By annotating a large number of images with side handle features, the YOLO large model is input into the model. The side handle features are extracted by repeatedly stacking images using a PyTorch network architecture. The error between the results and the data is calculated. When the error is less than 1%, the model is put into use.

[0289] The suitcase image is input into the large model. When the side handle feature is detected, the side handle feature is marked and selected from the suitcase image. The side handle feature is then identified in the selected image, which is the side handle image.

[0290] If the weight value is greater than the preset baseline weight value, it means that the suitcase is too heavy and needs to be moved with the assistance of another robotic arm.

[0291] S71: Update the luggage image based on the side handle position and define it as the side handle image.

[0292] The side handle image refers to the portion of the updated suitcase image that features the side handle.

[0293] S72: Determine the curvature of the side handle based on the side handle image.

[0294] The curvature of the side handle refers to the degree of bending of the side handle.

[0295] The method for determining the curvature of the side handle based on the side handle image involves first performing preprocessing such as grayscale conversion and noise reduction on the side handle image to enhance edge clarity; then, using an edge detection algorithm to extract the side handle contour and obtain a precise set of edge pixels. The contour point set is then fitted into a smooth curve, which can be achieved using the least squares method for polynomial curve fitting or spline curve fitting to obtain the mathematical expression of the curve; the curvature of the side handle is determined by calculating the curvature of each point on the curve and taking the average curvature, or by determining the curvature of key parts; alternatively, multiple feature points can be selected on the fitted curve, and the angle between the tangents of adjacent feature points can be calculated. These angle changes characterize the side handle curvature, which is common knowledge known to those skilled in the art and will not be elaborated upon here.

[0296] S73: Grasping pattern based on side handle curvature matching.

[0297] The gripping form refers to the shape of the robotic arm when gripping the side handle, which is adjusted according to the curvature of the side handle.

[0298] Different side handle curvatures correspond to different gripping patterns. The gripping angle is obtained by inputting the side handle curvature into the gripping database. The gripping database contains the relationship between the side handle curvature and the gripping pattern. The actual parameters are set in the gripping database by technicians in advance, with the side handle curvature corresponding to different gripping patterns set in advance.

[0299] S74: Control another preset robotic arm to move to the side handle position to grasp the side handle in a gripping mode to complete the adjustment.

[0300] Then, another robotic arm is controlled to move to the side handle position to grasp the side handle, thereby completing the adjustment and moving the suitcase, thus improving the stability during handling.

[0301] Based on the same inventive concept, embodiments of the present invention provide a multi-scenario intelligent control system for robotic grippers, including:

[0302] The acquisition module is used to acquire cabin images, baggage resistance, grab point images, recovery time, elasticity value, baggage weight value, placement images, and weight values.

[0303] The memory is used to store the program for intelligent control methods of robotic grippers in any multi-scenario application.

[0304] The processor loads and executes programs from memory.

[0305] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0306] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent control of a robotic gripper in multiple scenarios, characterized in that, include: S10: Acquire cabin images, and select the suitcases in the cabin images using a preset suitcase feature box to obtain suitcase images; S2: Select the handle in the suitcase image using a preset handle feature box to obtain a handle image; S3: Determine the handle type based on the handle image, wherein the handle type includes pull handles and rotary handles; S4: Determine the handle position in the cabin image based on the handle features and the preset reference position, and determine the gripping position based on the handle type and the handle position; S40: Based on the pull handle, control the preset robotic arm to grasp the suitcase at the grasping position and move it along the preset trajectory, and then stack it according to the preset stacking method; S41: Based on the rotating handle, the rotating handle is gripped at the gripping position using a preset gripping method, and the suitcase is moved along a preset trajectory, and then stacked using a preset stacking method; When the cabin image does not include the luggage image, the following is also included: S11: The baggage is selected by a preset baggage feature frame in the cabin image, and the selected image is defined as a baggage image; S110: Determine the shape and location of the luggage based on the luggage image and a preset reference position; S111: Determine the grab point based on the shape and position of the luggage; S112: Control the preset robotic arm to move to the grasping point, grasp with a preset grasping force, and rotate at a preset angle, and continuously collect images of luggage resistance and grasping point. Luggage resistance refers to the force that the luggage resists rotation in order to maintain its original shape during the rotation of the robotic arm. S113: Obtain the variation curve based on the luggage resistance fitting; S114: Determine the luggage status based on the change curve, the luggage status including an air-filled state and a tight-fitting state; S1140: If the luggage is in the tight-fitting state, determine the degree of gripping deformation based on the luggage resistance and the gripping point image; S11400: Control the preset robotic arm to reset, and collect the recovery time of the gripping point. The recovery time refers to the time required for the surface of the luggage to recover from the deformed state after being gripped and rotated back to the initial state. S11401: Determine the luggage material based on the degree of grasping deformation and the recovery time; S115: Determine the clamping gap based on the luggage material; S116: Control the preset robotic arm to grab the luggage at the gripping point with the gripping gap and move it along the preset trajectory, and then stack it in the preset stacking method; The following are included after S114 and before S115: S1141: If the luggage is in the empty state, determine the lifting point based on the grasping point image; S11410: Determine the lifting force based on the change curve and the preset gripping force matching; S11411: Control the preset robotic arm to move to the lifting point, lift with the lifting force, and collect the elasticity value during the lifting process to obtain the elasticity curve; S11412: Determine the luggage material based on the elasticity curve.

2. The intelligent control method for a robotic gripper in multiple scenarios according to claim 1, characterized in that, After the pre-set robotic arm grasps the luggage at the gripping point with the clamping gap and moves along the pre-set trajectory, the following steps are also included: S117: Collect baggage weight value; S118: If the weight of the luggage is greater than the preset benchmark weight, the luggage handle is selected in the luggage image by using the preset luggage handle feature, and the selected image is defined as the luggage handle image; S119: Determine the clamping position based on the luggage handle image and the luggage handle features; S120: Match the clamping force based on the luggage weight value; S121: Control another preset robotic arm to move to the gripping position to grip the luggage handle with gripping force, and then move along a preset trajectory after gripping and lifting the luggage.

3. The intelligent control method for a robotic gripper in multiple scenarios according to claim 1, characterized in that, The stacking method includes: S5: Acquire placement images at preset placement positions; S51: Determine the stacking height and stacking space based on the placement image; S52: Obtain the current handling volume based on the cabin image, and determine the stacking position based on the current handling volume and the stacking space; S53: Calculate the number of stacking layers based on the stacking height and the current handling volume; S54: Match the preset adjustment height required by the robotic arm according to the number of stacked layers; S55: Control the preset robotic arm to stack sequentially according to the stacking position and the adjusted height.

4. The intelligent control method for a robotic gripper in multiple scenarios according to claim 3, characterized in that, Following the determination of the stacking position, the following also applies: S6: Determine the luggage density based on the luggage weight value and the current handling volume; S60: When the luggage density is less than the preset leveling density, control the preset robotic arm to level the luggage at the stacking position with a preset leveling force; S61: When the baggage density is not less than the preset leveling density, determine the baggage shape based on the cabin image; S62: Determine the maximum and minimum values ​​of the luggage in three dimensions based on the shape of the luggage. The maximum and minimum values ​​in three dimensions refer to the maximum values ​​of the luggage in the length, width and height dimensions. S63: Match the placement volume according to the three-dimensional extreme values; S64: Determine a correction position based on the placement volume, the stacking space, and the luggage material, and replace the stacking position with the correction position; S65: Control the preset robotic arms to stack sequentially at the corrected position and the adjusted height.

5. The intelligent control method for a robotic gripper in multiple scenarios according to claim 1, characterized in that, When the pull handle is grasped, the grasping method includes: S400: Determine the shape and volume of the pull handle based on the handle image; S401: Determine the handle width based on the handle volume; S402: Determine the opening size during gripping based on the handle width; S403: Determine the gripping angle based on the shape of the pull-out handle; S404: Control the preset robotic arm to move to the grasping position and open it to the specified opening size, and grasp the pull handle at the specified grasping angle to complete the grasping operation.

6. The intelligent control method for a robotic gripper in multiple scenarios according to claim 5, characterized in that, The preset extraction methods include: S410: Determine the deflection position of the rotating handle based on the handle image; S411: Determine the flipping direction based on the aforementioned bias position; S412: Determine the clamping angle based on the gripping position and handle width; S413: Determine the shape of the rotating handle based on the handle image; S414: Determine the lifting distance based on the shape of the rotating handle; S415: At the gripping position, the rotating handle is pulled out at the gripping angle and lifted by the lifting distance to complete the gripping.

7. The intelligent control method for a robotic gripper in multiple scenarios according to claim 1, characterized in that, When the suitcase moves along a preset trajectory, it also includes: S7: Collect the weight value of the suitcase when it is being moved; S70: If the weight value is greater than the preset reference weight value, then the side handle is selected based on the luggage image using a preset side handle feature box, and the position of the selected side handle is defined as the side handle position. S71: Update the suitcase image based on the side handle position and define it as a side handle image; S72: Determine the curvature of the side handle based on the side handle image; S73: Match the gripping shape based on the curvature of the side handle; S74: Control another preset robotic arm to move to the side handle position and complete the gripping of the side handle in the gripping form to complete the adjustment.

8. A multi-scenario intelligent control system for a robotic gripper, characterized in that, include: The acquisition module is used to acquire cabin images; A memory for storing a program of a multi-scenario intelligent control method for a robotic gripper as described in any one of claims 1 to 7; The processor loads and executes programs from memory.

Images