Robot control method, robot, and computer-readable storage medium

Abstract

The application provides a robot control method, a robot and a computer readable storage medium, wherein the robot transports at least one preset object by using a bearing structure, the robot control method comprises the following steps: obtaining a feature model, the feature model comprises feature information of one or more preset objects; determining a target area, the target area is a field of view range of an image acquisition device in the robot or a part of the field of view range of the image acquisition device in the robot, and the bearing structure is located in the target area; obtaining a current image; performing model detection on the current image according to the feature model, identifying feature data of the preset object in the target area, and obtaining a screening result; and controlling the robot to perform a preset action according to the screening result. The embodiment of the application can obtain the features of the preset object in the bearing structure, cooperate with the current task of the robot, has the advantages of high accuracy and strong universality, and can be applied to a complex application scene environment.

Description

Technical Field

[0001] This invention relates to the field of intelligent devices and control technology, and in particular to a robot control method, a robot, and a computer-readable storage medium. Background Technology

[0002] Currently, there are robots used for transfer and delivery, widely applied in warehouses, restaurants, hotels, canteens, hospitals, and other scenarios. These robots can transport pre-set objects to target locations, reducing manual operations and enabling unified scheduling and management of the work process, thus improving the efficiency of object transfer. Taking a restaurant food delivery robot as an example, the operator places pre-set objects, such as processed food, tableware, napkins, and receipts, onto the robot's support structure. The robot can then locate the target position corresponding to the pre-set object on the support structure and move towards that position.

[0003] Existing robots used for transfer and delivery lack the ability to identify pre-defined objects within the supporting structure. In real-world operating scenarios, the situation is even more complex, limiting the robots' intelligence. During specific tasks, frequent operator intervention is often required, with manual input of commands to execute tasks targeting the pre-defined objects within the structure. Simply setting up a camera to photograph the structure suffers from poor versatility in practice. Different applications, pre-defined objects, or task objectives require tailored designs, making the process cumbersome and the algorithms bloated.

[0004] The content of the background section is merely the technology known to the inventor and does not necessarily represent the prior art in this field. Summary of the Invention

[0005] To address one or more deficiencies in existing technologies, this invention provides a robot control method, wherein the robot utilizes a support structure to transport at least one preset object, and the robot control method includes:

[0006] Obtain a feature model, wherein the feature model includes feature information of one or more preset objects;

[0007] A target area is defined, which is the field of view of the image acquisition device in the robot, or a part of the field of view of the robot's image acquisition device, and the supporting structure is located within the target area;

[0008] Get the current image;

[0009] Based on the feature model, model detection is performed on the current image to identify the feature data of preset objects in the target area and obtain the filtering results.

[0010] The robot is controlled to perform preset actions based on the screening results.

[0011] According to one aspect of the present invention, the step of obtaining the feature model includes:

[0012] Collect feature data of one or more preset objects located on the supporting structure and establish a dataset;

[0013] The dataset is divided into a training set, a validation set, and a test set;

[0014] Model training is performed on the data of the preset objects.

[0015] According to one aspect of the invention, the step of determining the target region includes:

[0016] Acquire field-of-view images within the field of view of the image acquisition device;

[0017] Determine the boundaries of the load-bearing structure based on the field-of-view image;

[0018] The center position of the load-bearing structure is determined based on the preset contour features of the load-bearing structure and the boundary of the load-bearing structure.

[0019] The position of the load-bearing structure is determined based on its center position and outline, and the target area is determined based on the position of the load-bearing structure.

[0020] According to one aspect of the invention, the step of obtaining the screening result includes:

[0021] Output a detection box for the object region based on the feature model;

[0022] The center of the detection frame is compared with the center of the target region;

[0023] Determine the filtering results, wherein the filtering results include the detection boxes within the target area.

[0024] According to one aspect of the invention, the step of obtaining the screening results further includes:

[0025] The center distance between the center of the target area and the center of the detection frame is used as the center distance, and the center distance corresponding to the detection frame is compared with the radius of the preset contour feature of the bearing structure.

[0026] In the filtering results, retain the detection boxes whose center distance is not greater than the preset contour feature radius.

[0027] According to one aspect of the present invention, the step of controlling the robot to perform a preset action includes:

[0028] Continuously, or at a preset frequency, acquire images of multiple target areas, and determine the changes in the number and type of preset objects in the load-bearing structure based on the temporal sequence of the images of multiple target areas;

[0029] Based on the changes in the number and types of pre-set objects within the supporting structure, the robot is controlled to change its current motion state or maintain its current motion state.

[0030] According to one aspect of the invention, the robot's motion state includes waiting at the current position and moving towards a target position; the step of controlling the robot to change or maintain the current motion state includes:

[0031] If the change in the number of pre-set objects within the supporting structure matches the expected change, control the robot to change its current motion state; if the change in the number of pre-set objects within the supporting structure does not match the expected change, control the robot to maintain its current motion state.

[0032] According to one aspect of the invention, the robot's motion state further includes the robot's motion speed and acceleration; the step of controlling the robot to change its current motion state or maintain its current motion state includes:

[0033] When a preset object of a preset type is detected on the supporting structure, the robot's movement speed and / or acceleration are increased or decreased.

[0034] According to one aspect of the invention, the invention also includes a robot, the robot comprising:

[0035] main body;

[0036] A load-bearing structure is disposed on the main body;

[0037] A sensor, disposed on the main body, and configured to acquire an image of the supporting structure;

[0038] A control system that communicates with the sensor and is configured to execute the robot control method as described above.

[0039] According to one aspect of the invention, the invention also includes a computer-readable storage medium including computer-executable commands stored thereon, which, when executed by a processor, implement the robot control method as described above.

[0040] Compared with existing technologies, embodiments of the present invention provide a robot control method, including a process of acquiring a target area using an image processing device, acquiring feature data of preset objects within the target area, obtaining a filtering result, and controlling the robot to perform preset actions based on the filtering result. The target area includes a supporting structure, thereby enabling the acquisition of features of preset objects within the supporting structure, such as quantity, type, position, and color information. This method can be coordinated with the robot's current task, offering advantages such as high accuracy and versatility, and is applicable to complex application scenarios. The present invention also includes an embodiment of a robot and a computer-readable storage medium capable of executing the aforementioned robot control method. Attached Figure Description

[0041] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0042] Figure 1 This is a flowchart illustrating a robot control method in one embodiment of the present invention;

[0043] Figure 2 This is a flowchart illustrating the process of training a model using data of a preset object in one embodiment of the present invention.

[0044] Figure 3 This is a schematic diagram of preset object model training in one embodiment of the present invention;

[0045] Figure 4 This is a flowchart illustrating a robot control method including a process for determining a target area in one embodiment of the present invention;

[0046] Figure 5 This is a flowchart illustrating a robot control method including a process for obtaining screening results in one embodiment of the present invention;

[0047] Figure 6 This is a flowchart illustrating a robot control method for controlling a robot to perform a preset action process, as shown in one embodiment of the present invention.

[0048] Figure 7 This is a structural block diagram of a robot in one embodiment of the present invention;

[0049] Figure 8 This is a schematic diagram of the robot structure in one embodiment of the present invention. Detailed Implementation

[0050] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0051] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0052] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, an electrical connection, or a connection that allows for communication; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0053] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0054] The following disclosure provides many different embodiments or examples for implementing various structures of the invention. To simplify the disclosure, specific examples of components and arrangements are described below. These are merely examples and are not intended to limit the invention. Furthermore, reference numerals and / or letters may be repeated in different examples; such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. In addition, examples of various specific processes and materials are provided in this invention, but those skilled in the art will recognize the application of other processes and / or the use of other materials.

[0055] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0056] Figure 1 The specific flow of the robot control method 100 according to an embodiment of the present invention is shown below, in conjunction with... Figure 1 Please elaborate.

[0057] The robot in this embodiment is equipped with a support structure for placing and fixing at least one preset object, such as a material bin, tray, or bracket. The robot can move in the work environment to transport the preset object within the support structure to a target location. The types of preset objects vary in different application scenarios. For example, in restaurants and hotels, robots used for food delivery or item distribution may use pre-processed meals, menus, bills, tableware, napkins, room keys, or goods purchased online by consumers. In other application scenarios, the preset objects can be other types, which are not limited in this embodiment.

[0058] like Figure 1 As shown, in the robot control method 100, in step S101, a feature model is acquired. The feature model includes feature information of one or more preset objects. Specifically, according to a preferred embodiment of the invention, the preset object is a plate. The feature information of the preset object includes the shape of the plate and whether food is present on it. For different preset objects, the feature information may include shape, quantity, position, etc., such as a set of tableware. The feature model of the preset object can be obtained through an image acquisition device on the robot, for example, by taking a picture of the preset object using a camera or other visual sensor. After acquiring the image of the preset object, its feature information can be obtained through analysis.

[0059] In step S102, a target area is determined. In this embodiment, the target area is the field of view of the image acquisition device in the robot, or a part of the field of view of the robot's image acquisition device, and the robot's supporting structure is located within the target area. Preferably, the image acquisition device in the robot faces the supporting structure, and the field of view covers the entire position of the supporting structure. The relative positional relationship between the image acquisition device and the supporting structure is fixed, and the supporting structure is in a fixed position within the field of view of the image acquisition device. It is preferable to set the target area within the field of view of the image acquisition device to eliminate interference and influence from other parts within the field of view of the image acquisition device on the target area, thereby improving the accuracy of the image of the target area.

[0060] In step S103, the current image is acquired by using the image acquisition device in the robot to capture an image containing the supporting structure. Typically, the field of view of the image acquisition device and the area of ​​the supporting structure do not completely overlap; the current image includes areas other than the supporting structure. Preferably, the supporting structure is positioned at the center of the field of view of the image acquisition device to avoid displacement of the supporting structure or the image acquisition device due to collisions, vibrations, etc., resulting in the inability to fully acquire the image of the supporting structure. For robots with multiple supporting structures, multiple corresponding image acquisition devices can be set up, or a single image acquisition device can be set up and multiple corresponding target areas can be defined, or a target area including all supporting structures can be defined. In some embodiments of the present invention, the robot includes multiple supporting structures stacked vertically, and correspondingly, multiple image acquisition devices are set up corresponding to multiple supporting structures.

[0061] In step S104, model detection is performed on the current image based on the feature model to identify the feature data of preset objects in the target area, thereby obtaining the filtering results. The current image acquired by the image acquisition device includes the supporting structure. By performing model detection on the current image using the feature model of the preset objects, the feature data of the preset objects in the target area can be identified. Specifically, this includes, for example, the type, quantity, location, and relative positional relationship of the preset objects. After obtaining the feature data, the filtering results can be obtained according to preset rules. The specific process of obtaining the filtering results will be described in detail in subsequent embodiments.

[0062] In step S105, the robot is controlled to perform preset actions according to the screening results. According to a preferred embodiment of the present invention, the preset actions of the robot may include movement, stopping, acceleration, deceleration, turning, alarm, etc.

[0063] In this embodiment, the robot control method 100 improves the accuracy of recognizing the preset object within the bearing structure by establishing a feature model of the preset object and determining the target area including the bearing structure. Furthermore, for complex specific application situations, it simplifies the control of the robot to execute preset actions based on the screening results, thereby increasing the applicability of the robot control method. It only requires setting an appropriate correspondence between the corresponding application scenario and the preset object to meet the application needs of most robots.

[0064] Figure 2 The specific process for obtaining the feature model according to a preferred embodiment of the present invention is shown. Figure 3 The training memory logic of the preset object model is shown below, combined with Figure 2 and Figure 3 Detailed explanation.

[0065] like Figure 2 and Figure 3 As shown, in step S201, feature data of one or more preset objects located on the supporting structure are collected, and a dataset is established. In this step, preset objects that may exist in actual application can be placed in advance within the supporting structure, and the types, quantities, and locations of the preset objects are known. For example, the supporting structure with the preset objects placed can be photographed using an image acquisition device, and other information about the preset objects can be manually input. Specifically, the types, quantities, locations, and other information of various preset objects can be classified and sorted. Directly acquiring the feature data of the preset objects on the supporting structure can further improve the accuracy of the feature data, ensuring accurate correspondence with the current image acquired during actual use. Furthermore, the process of acquiring the feature data of the preset objects can also be performed while the robot is performing a delivery task, thereby increasing the amount of data in the dataset and improving the accuracy of identifying the preset objects. Figure 3 As shown, in a preferred embodiment of the present invention, the acquired feature data can be preprocessed, such as by filtering, classifying, and sorting.

[0066] In step S202, the dataset is divided into a training set, a validation set, and a test set for use in step S203, where the data of the preset objects is used to train the model. According to a preferred embodiment of the present invention, the data of the preset objects is fed into the training set for iterative analysis to obtain the optimal model weights, which are then used as the feature model of the preset objects and applied in the robot control method to further improve the accuracy of recognizing the preset objects. For different types of preset objects and combinations of multiple preset objects, corresponding datasets can be established separately and trained to obtain corresponding feature models.

[0067] Figure 4 The following diagram illustrates a specific flow of a robot control method 300 according to a preferred embodiment of the present invention, including a process for determining a target area. For example... Figure 4 As shown, steps S301, S307, S308, and S309 in robot control method 300 are basically the same as steps S101, S103, S104, and S105 in robot control method 100, and will not be described again here. Robot control method 300 can also use the aforementioned method of obtaining a feature model of a preset object to obtain an accurate feature model, which can then be used in the subsequent recognition process.

[0068] In step S302, a field-of-view image within the field of view of the image acquisition device is acquired. In this embodiment, the robot's image acquisition device can acquire an image of the supporting structure, meaning the supporting structure is located within the field of view of the image acquisition device. This step utilizes the image acquisition device to capture an image of the field of view containing the supporting structure.

[0069] In step S303, the boundary of the supporting structure is determined based on the field of view image. Specifically, the boundary of the supporting structure can be determined based on the preset positional constraint relationship between the image acquisition device and the supporting structure. However, in practical applications, the robot's image acquisition device and the supporting structure may deviate due to limited installation accuracy or collisions and vibrations, leading to a decrease in the accuracy of determining the boundary of the supporting structure. According to a preferred embodiment of the present invention, the actual boundary position of the supporting structure can be obtained from the field of view image captured in real time by the image acquisition device. For example, the Sobel operator can be used to perform boundary processing on the field of view image of the image acquisition device to extract the boundary of the supporting structure.

[0070] In step S304, the center position of the support structure is determined based on the preset contour features of the support structure and the boundary of the support structure obtained in step S303. The preset contour features of the support structure can be the shape of the support structure at the time of manufacture. For example, when the support structure is circular, the preset contour features can be determined by the radius. Further, according to a preferred embodiment of the present invention, the preset contour features of the support structure are substituted into the boundary of the support structure to determine the deviation value and perform correction. For example, by changing the position and shape features of the preset contour of the support structure until the preset contour features of the support structure are close to the boundary of the support structure in the field of view image, preferably, the boundary of the support structure is replaced by all or part of the preset contour of the support structure, and the boundary contour and center position of the support structure are finally determined. Specifically, for example, the boundary and preset contour of the support structure are pixelated, and the preset contour of the support structure is determined to coincide as much as possible with the boundary obtained by actual shooting based on the percentage of pixel overlap. The position of the support structure is corrected, and finally, the center position of the preset contour of the support structure is taken as the center of the support structure.

[0071] In step S305, the position of the bearing structure in the field of view image of the image acquisition device is determined according to the center position of the bearing structure and the preset contour features of the bearing structure. This step can offset the errors caused by deformation or missing parts of the bearing structure boundary in some cases, or the incomplete boundary of the bearing structure acquired by the image acquisition device due to the influence of light, or the difficulty in recognition, so as to improve the accuracy of identifying the position of the bearing structure.

[0072] In step S306, the target area is determined based on the location of the supporting structure. The target area includes the entire range of the supporting structure. Preferably, the target area should minimize the area of ​​the non-supporting structure to reduce the influence of other locations in the field of view image acquired by the image acquisition device on the recognition of the preset object.

[0073] Figure 5 The following is a detailed flowchart of a robot control method 400 according to a preferred embodiment of the present invention, including the process of obtaining screening results. Steps S401, S402, S403 and S409 in the robot control method 400 are substantially the same as steps S101, S102, S103 and S105 in the robot control method 100, and will not be described again.

[0074] In step S404, a detection box is output for the target area based on the feature model, and the actual position of the factors and objects in the bearing structure is matched. The feature model of the preset object includes the feature information of the preset object. In this embodiment, the aforementioned method for obtaining the feature model can be used to obtain the accurate feature model of the preset object.

[0075] In step S405, the center of the detection frame is compared with the center of the target area, where the center of the detection frame is the center of the detection frame's geometry. Further, according to a preferred embodiment of the present invention, in step S406, the distance between the center of the target area and the center of the detection frame is used as the center distance, and the center distance corresponding to the detection frame is compared with the radius of a preset contour feature of the supporting structure. In this embodiment, the feature model includes feature information of a preset object. The preset object is placed on the supporting structure, and the detection frame corresponds to the actual position of the preset object. The center distance in this step reflects the position of the preset object from the center of the supporting structure.

[0076] In step S407, detection boxes with a center distance not greater than the radius of the preset contour feature of the supporting structure are retained in the filtering results. Each detection box has a corresponding center distance. If the center distance of a detection box is greater than the radius of the preset contour feature of the supporting structure, it indicates that the position of the detection box exceeds the preset contour of the supporting structure, and the position of the detection box is not within the range of the supporting structure. The filtering results include detection boxes within the target area. In this embodiment, the detection boxes are filtered by center distance. In practical applications, the field of view image obtained by the image acquisition device is represented in the form of an image. In areas other than the supporting structure, there may also be preset objects, or other objects or images that the feature model of the preset objects cannot distinguish, such as changes in light and shadow. In this embodiment, the process of judging the position of the detection boxes is added to filter out the filtering results that exceed the range of the supporting structure, further improving the accuracy of preset object recognition. In step S408, the filtering results are determined for subsequent control of the robot.

[0077] Figure 6 The following diagram illustrates the specific flow of a robot control method 500 according to a preferred embodiment of the present invention, including the process of controlling the robot to perform preset actions. Specifically, according to the preferred embodiment of the present invention, the preset actions of the robot may include waiting at the current position, moving to a target position, etc. Steps S501, S502, S503, and S504 in the robot control method 500 are basically the same as steps S101, S102, S103, and S104 in the robot control method 100, and will not be described again.

[0078] In step S505, images of multiple target areas are continuously acquired or acquired at a preset frequency, and the quantity and type of preset objects in the support structure are determined based on the temporal sequence of the images of the multiple target areas. This step can reflect the changes in the type and quantity of preset objects in the support structure within a certain time period. Taking a food delivery robot used in a restaurant as an example, the support structure contains dishes, and the food delivery robot receives and executes the food delivery task. The preset objects include the dishes. Specifically, the food delivery robot is located in the initial position, and the operator or other cooperating equipment places the dishes corresponding to the food delivery task on the support structure of the food delivery robot. During this process, the quantity of preset objects in the support structure changes from 0 to 1. After the dishes are delivered to the target location, the dishes are taken out, and the quantity of preset objects in the support structure changes again.

[0079] In step S506, it is determined whether the change in the number of preset objects within the supporting structure matches the expected change. The expected change in this step corresponds to the robot's current task. Taking a food delivery robot in a restaurant as an example, when the robot is at its initial position, the expected change in the number of preset objects within the supporting structure is acquiring the same number of dishes as the task; when the robot reaches its target position, the expected change in the number of preset objects within the supporting structure is unloading the same number of dishes as the task. This step determines whether the change in the number of preset objects within the supporting structure in the actual obtained filtering results matches the expected change.

[0080] When the change in the number of preset objects within the support structure does not conform to the expected change, in step S508, the robot is controlled to maintain its current motion state. For example, if the food delivery robot is at the initial position and the support structure has not acquired the same number of dishes as the task, it continues to wait at the initial position; if the food delivery robot moves to the target position and the support structure has not unloaded the same number of dishes as the task, it continues to wait at the target position and maintains its current motion state.

[0081] When the change in the number of preset objects within the support structure matches the expected change, in step S507, the robot is controlled to change its current motion state. For example, when the food delivery robot is at the initial position, and the support structure has acquired the same number of dishes as the task, the current motion state is changed (stopped at the initial position), and the robot moves towards the target position. When the food delivery robot moves to the target position, and the support structure has unloaded the same number of dishes as the task, the current motion state is changed (stopped at the target position), and the robot continues to move towards the target position of the next task.

[0082] According to a preferred embodiment of the present invention, the robot's motion state further includes the robot's motion speed and acceleration. In the robot control method of this embodiment, the step of controlling the robot to change or maintain its current motion state further includes increasing or decreasing the robot's motion speed and / or acceleration when a preset object of a preset type is detected on the supporting structure. For different types of preset objects, the robot is controlled to select appropriate motion speed and acceleration. For example, when transporting liquids, to prevent spillage, the robot can be controlled to move at a lower motion speed, and an upper limit can be set on the acceleration. For preset objects that need to be quickly transported to a target location and where there is no risk of damage or spillage, the robot's motion speed and acceleration can be increased.

[0083] The present invention also includes an embodiment of a robot 1, wherein the robot 1 includes a main body 10, a supporting structure 20, sensors 30, and a control system 40. The main body 10 is the main structural frame of the robot 1, and all components of the robot 1 are fixed to the main body, for example... Figure 8As shown in the figure. The main body 10 can be made of alloy or organic materials to form a fixed shape. The various components of the robot 1 are installed in corresponding positions in the main body 10 to achieve specific functions.

[0084] like Figure 8 As shown, the support structure 20 is disposed on the main body 10. The support structure 20 is used to support and place objects. For example, the support structure 20 is configured in the shape of a tray, and the specific object to be transferred is placed on the tray and moves with the robot. Of course, in different embodiments of the present invention, the support structure 20 can also be configured as a box with a certain depth.

[0085] Sensor 30 is mounted on the main body 10, and the supporting structure 20 is located within the field of view of sensor 30. Sensor 30, as an image acquisition device, is capable of acquiring images of the supporting structure 20. Figure 8 As shown, according to the specific design of robot 1 and the installation position of the support structure 20, sensor 30 can be set directly above or diagonally above the support structure 20. Preferably, sensor 30 is set directly above the support structure 20, the support structure 20 is located at the center of the field of view of sensor 30, and the plane in the support structure 20 used to place objects is perpendicular to the optical axis of sensor 30, so as to minimize image distortion caused by tilt angle.

[0086] like Figure 7 As shown, the control system 40 communicates with the sensor 30 and can execute the load-bearing structure contour extraction method described in the previous embodiment to obtain the accurate boundary contour of the load-bearing structure 20. This provides a basis for subsequent processes such as determining the current state of the load-bearing structure. Furthermore, the robot 1 also includes a motion device 50, such as a driveable roller. The motion device 50 is signal-connected to the control system 40, and the control system 40 can control the motion device 50 to change the motion state of the robot 1.

[0087] According to other embodiments of the present invention, the robot further includes: a mobile chassis, a functional controller for providing user operation, a low-level controller for map generation and path planning, and a component controller for controlling the mobile unit and the environmental detection unit.

[0088] The mobile unit is equipped with at least two sets of drive wheels, each set located on one side of the mobile chassis, which drive the robot to move. A component controller controls the rotational speed of the drive wheels, thereby driving the robot to move. Preferably, the drive wheels on different sides of the mobile chassis are controlled separately, and the robot's steering is achieved by controlling the drive wheels to rotate at different speeds.

[0089] Specifically, the mobile unit has two sets of drive wheels, one set serving as the left drive wheel and the other as the right drive wheel. The left and right drive wheels are located on opposite sides of the chassis. When the robot needs to turn, the left and right drive wheels are controlled to rotate at different speeds to achieve robot turning. Optionally, the mobile unit may also include at least two sets of driven wheels, one set of drive wheels corresponding to one set of driven wheels. At least one set of driven wheels serves as the left driven wheel, and at least one set serves as the right driven wheel. The left and right driven wheels assist the left and right drive wheels in driving the robot's shell and mobile chassis, reducing the load on the drive wheels and improving the robot's stability. The driven wheels can also be positioned along the centerline of the left and right drive wheels to distribute the robot's weight as evenly as possible.

[0090] The bottom of the mobile chassis is equipped with at least one turn signal unit, and each turn signal unit includes at least one turn signal. The component controller can also control the turn signals in the turn signal units to illuminate according to a preset pattern when the robot turns. The robot provided in this embodiment of the invention improves the robot's movement safety by controlling the turn signals to illuminate when the robot turns, thereby alerting pedestrians.

[0091] In conjunction with the foregoing embodiments, optionally, when the speed difference between the drive wheels on both sides of the mobile chassis exceeds a preset value, the component controller controls the turn signals in the turn signal unit to illuminate according to a preset method. Furthermore, the robot also includes a voice module, which is electrically connected to the component controller. When the robot turns, the voice module is controlled to issue voice prompts.

[0092] The lidar system can be mounted on the main body and can rotate along a set plane, so that the photoelectric receiving array of the lidar system forms a scanning cylinder.

[0093] Specifically, the lidar system can be installed at the opening in the robot's shell to emit laser signals to detect surrounding objects. This lidar system includes a photoelectric receiving array and a laser emitting unit array. When the lidar system rotates along a set plane, the photoelectric receiving array can form a scanning cylinder, obtaining a larger scanning range. This facilitates the acquisition of detailed object shapes and avoids collisions with obstacles. Optionally, the set plane can be a horizontal plane to facilitate object detection during robot movement. Furthermore, other set planes, such as a vertical plane, can be selected according to specific user needs; this embodiment does not limit this selection.

[0094] According to embodiments of the present invention, a robot path planning method is also provided, which can be applied to any robot. It should be noted that the steps can be performed sequentially or, depending on the actual situation, multiple steps can be performed simultaneously; no limitation is made here. The robot path planning method provided in this application includes the following steps:

[0095] First, a map of the robot's operational range is constructed to determine the current working area's location map. This location map is formed by the robot mapping its environment. Specifically, the robot is equipped with data acquisition sensors and a modeling processor. The modeling processor uses environmental data collected by the sensors to create an environmental map. In this embodiment, the data acquisition sensors include LiDAR, ultrasonic sensors, and infrared sensors. These sensors collect data about the robot's working area. The modeling processor uses this data to create a map, generating different map layers using different sensors, such as static layers, dynamic obstacle layers, ultrasonic layers, and visual layers. These layers are then fused to obtain a location map for the robot's positioning and navigation.

[0096] Next, a route is planned based on the location map.

[0097] Furthermore, the robot's current position and target position are determined based on the positioning map, the positions of obstacles are determined based on the positioning map, or a path is planned based on the current position, the target position, and the positions of obstacles.

[0098] Specifically, the target location is either the location set by the user or the location determined by the robot's processing system to be moved to. This target location can be the next location determined during the movement, or the final location the robot needs to reach. The current location is the robot's real-time position information determined by its position sensors.

[0099] The location map is used to determine the position of obstacles on the map. This implementation method allows the robot to determine the location of obstacles and plan a route without compromising navigation accuracy.

[0100] Finally, the robot is controlled to move according to the planned path.

[0101] According to a preferred embodiment of the present invention, a computer-readable storage medium is further included, comprising computer-executable commands stored thereon, which, when executed by a processor, implement the aforementioned method for extracting the profile of the load-bearing structure.

[0102] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A robot control method, wherein the robot uses a support structure to transport at least one preset object, the robot control method comprising: Obtain a feature model, wherein the feature model includes feature information of one or more preset objects; Determining a target area, where the target area is the field of view of the image acquisition device in the robot, or a part of the field of view of the robot's image acquisition device, and the supporting structure is located within the target area, includes: acquiring a field of view image within the field of view of the image acquisition device; determining the boundary of the supporting structure based on the field of view image; determining the center position of the supporting structure based on the preset contour features of the supporting structure and the boundary of the supporting structure; determining the position of the supporting structure based on the center position and the contour of the supporting structure; and determining the target area based on the position of the supporting structure; wherein, by pixelating the boundary of the supporting structure and the preset contour of the supporting structure, and determining the overlap between the preset contour features of the supporting structure and the boundary of the supporting structure based on the percentage of pixel overlap, the position of the supporting structure is corrected, and the center position of the preset contour of the supporting structure is taken as the center position of the supporting structure; Get the current image; The process involves performing model detection on the current image based on a feature model, identifying feature data of preset objects in the target region, and obtaining filtering results. This includes: outputting detection boxes for the target region based on the feature model; comparing the center of the detection boxes with the center of the target region; determining the filtering results, wherein the filtering results include detection boxes within the target region; using the distance between the center of the target region and the center of the detection boxes as the center distance, comparing the center distance corresponding to the detection boxes with the radius of a preset contour feature of the supporting structure; and retaining detection boxes in the filtering results whose corresponding center distance is not greater than the radius of the preset contour feature. The robot is controlled to perform preset actions based on the screening results.

2. The robot control method according to claim 1, wherein the step of acquiring the feature model includes: Collect feature data of one or more preset objects located on the supporting structure and establish a dataset; The dataset is divided into a training set, a validation set, and a test set; Model training is performed on the data of the preset objects.

3. The robot control method according to claim 1 or 2, wherein the step of controlling the robot to perform a preset action includes: Continuously, or at a preset frequency, acquire images of multiple target areas, and determine the changes in the number and type of preset objects in the load-bearing structure based on the temporal sequence of the images of multiple target areas; Based on the changes in the number and types of pre-set objects within the supporting structure, the robot is controlled to change its current motion state or maintain its current motion state.

4. The robot control method according to claim 3, wherein the robot's motion states include waiting at the current position and moving towards the target position; The steps for controlling the robot to change its current motion state or maintain its current motion state include: If the change in the number of pre-set objects within the supporting structure matches the expected change, control the robot to change its current motion state. If the number of pre-set objects within the supporting structure does not change as expected, the robot is controlled to maintain its current motion state.

5. The robot control method according to claim 3, wherein the robot's motion state further includes the robot's motion speed and acceleration; the step of controlling the robot to change its current motion state or maintain its current motion state includes: When a preset object of a preset type is detected on the supporting structure, the robot's movement speed and / or acceleration are increased or decreased.

6. A robot, comprising: main body; A load-bearing structure is disposed on the main body; A sensor, disposed on the main body, and configured to acquire an image of the supporting structure; A control system that communicates with the sensor and is configured to perform the robot control method as described in any one of claims 1-5.

7. A computer-readable storage medium comprising computer-executable commands stored thereon, the executable commands, when executed by a processor, implementing the robot control method as described in any one of claims 1-5.

Images