Control method of robot arm, desktop robot, electronic device

Abstract

The embodiment of the application provides a control method of a mechanical arm, a desktop robot and an electronic device, and the method comprises the following steps: acquiring a panoramic image of an operable area of the desktop robot based on a visual sensor located at a head component of the desktop robot, and determining a to-be-operated position of the desktop robot according to the panoramic image, wherein the to-be-operated position is used for indicating a position of the desktop robot to-be-executed game action; determining a target rotation angle of a mechanical arm of the desktop robot moving to the to-be-operated position; and controlling the mechanical arm to move to the to-be-operated position according to the target rotation angle.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automatic control, in particular, to a control method of a mechanical arm, a desktop robot and an electronic device. BACKGROUND

[0002] With the continuous progress of technology, especially the maturity of artificial intelligence technology, household desktop robots have gone from science fiction to reality and gradually become helpers and partners of modern family life. In this field, desktop robots, as a combination of mechanical arm operation and intelligent decision-making, provide a new way of entertainment for game enthusiasts.

[0003] However, the design of existing desktop robots is usually based on a one-on-one tabletop game mode, which can only play the same game with one user at a time, limiting the diversity of use scenarios and the sharing of entertainment. For example, when there are multiple members in a family who want to play different tabletop games at the same time, traditional desktop robots cannot meet this demand and must play with each user one by one, reducing the continuity and interactivity of entertainment.

[0004] Therefore, in the prior art, the desktop robot only supports one-on-one tabletop games and cannot simultaneously meet the needs of multiple users playing different tabletop games, resulting in obvious limitations in function expansion and user experience.

[0005] In view of the problem that the traditional desktop robot in the prior art can only play with one person, there is no effective solution to this problem at present.

[0006] Therefore, it is necessary to improve the related technology to overcome the defects in the related technology. SUMMARY

[0007] The embodiments of the present application provide a control method of a mechanical arm, a desktop robot and an electronic device to at least solve the problem that the traditional desktop robot in the prior art can only play with one person and has certain limitations.

[0008] According to an embodiment of the present application, a control method of a mechanical arm is provided, comprising: acquiring a panoramic image of an operable area of a desktop robot based on a visual sensor located at a head assembly of the desktop robot, and determining a to-be-operated position of the desktop robot according to the panoramic image, wherein the to-be-operated position is used to indicate a position at which the desktop robot is to perform a game action; determining a target rotation angle of a mechanical arm of the desktop robot moving to the to-be-operated position; and controlling the mechanical arm to move to the to-be-operated position according to the target rotation angle.

[0009] In an example embodiment, acquiring a panoramic image of an operable area of a tabletop robot based on a vision sensor located in a head assembly of the tabletop robot comprises: acquiring partial images respectively captured by the vision sensor at different rotation angles, wherein there is a common field of view between two adjacent partial images; performing a stitching step: extracting feature points in each partial image, and matching feature points in an Nth partial image with feature points in an (N+1)th partial image; calculating a homography transformation matrix of the Nth partial image and the (N+1)th partial image based on the matched feature points; performing homography transformation on the (N+1)th partial image based on the homography transformation matrix of the Nth partial image and the (N+1)th partial image, and stitching the homography-transformed (N+1)th partial image with the Nth partial image, wherein N is a positive integer; and performing the stitching step cyclically until the homography-transformed Mth partial image is stitched with an (M-1)th partial image to obtain the panoramic image, wherein the Mth partial image is the last partial image acquired by the vision sensor in a target period, and M is a positive integer greater than 1.

[0010] In an example embodiment, determining a position to be operated on the tabletop robot based on the panoramic image comprises: determining whether a game state in each region of the panoramic image changes based on a first feature vector of each region in the panoramic image and a second feature vector of each region in a historical panoramic image, wherein the historical panoramic image is a previous image of the panoramic image, and each region corresponds to a tabletop game in a different process; determining whether the tabletop robot needs to perform a game operation based on a game rule of a tabletop game in a target region in which a game state is determined to change; and determining a position corresponding to the target region as the position to be operated on the tabletop robot in a case where it is determined that the tabletop robot needs to perform the game operation.

[0011] In an example embodiment, determining a target rotation angle at which a mechanical arm of the tabletop robot moves to the position to be operated on comprises: determining a first rotation angle of a head assembly corresponding to a first absolute value encoder when a target partial image is captured by the vision sensor, wherein the first absolute value encoder is used to acquire a rotation angle of the head assembly, and the target region is present in the target partial image; and determining the first rotation angle as the target rotation angle.

[0012] In an example embodiment, the controlling the robot arm to move to the to-be-operated position according to the target rotation angle comprises: determining a second rotation angle of a second absolute value encoder corresponding to the robot arm; determining whether the second rotation angle is consistent with the target rotation angle; in a case where the second rotation angle is consistent with the target rotation angle, determining that the robot arm moves to the to-be-operated position; and in a case where the second rotation angle is not consistent with the target rotation angle, controlling the robot arm to move to the to-be-operated position according to the target rotation angle.

[0013] In an example embodiment, after the controlling the robot arm to move to the to-be-operated position according to the target rotation angle, the method further comprises: determining a tabletop game corresponding to the to-be-operated position; determining a target position of a picking component of the robot arm according to rules of the tabletop game and a current state of the tabletop game; and controlling the picking component to grasp a target object at the target position and / or moving the target object to the target position, wherein the picking component is configured to grasp or release the target object.

[0014] In an example embodiment, the acquiring the panoramic image of the operable region of the tabletop robot based on the visual sensor of the head component of the tabletop robot comprises: determining whether multiple rounds of tabletop games exist in the operable region simultaneously; in a case where multiple rounds of tabletop games do not exist in the operable region simultaneously, determining a target operable region in which a tabletop game is played, and acquiring an image of the target operable region of the tabletop robot based on the visual sensor of the head component of the tabletop robot; and in a case where multiple rounds of tabletop games exist in the operable region simultaneously, acquiring the panoramic image of the operable region of the tabletop robot based on the visual sensor of the head component of the tabletop robot.

[0015] According to another embodiment of the present application, a tabletop robot is provided, comprising: a visual sensor configured to acquire a panoramic image of an operable region of the tabletop robot; a driving module configured to determine a to-be-operated position of the tabletop robot according to the panoramic image, determine a target rotation angle of a robot arm of the tabletop robot to move to the to-be-operated position, and control the robot arm to move to the to-be-operated position according to the target rotation angle, wherein the to-be-operated position is configured to indicate a position at which the tabletop robot is to perform a game action.

[0016] In an example embodiment, the tabletop robot comprises: a head component and a body spring piece, wherein a rotating contact spring piece of the head component is in contact with a body spring piece seat through elastic force.

[0017] According to a further embodiment of the present application, a computer readable storage medium is also provided, in which a computer program is stored, wherein the computer program is configured to perform the steps of any of the method embodiments described above when executed.

[0018] According to a further embodiment of the present application, an electronic device is also provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to perform the steps of any of the method embodiments described above.

[0019] According to a further embodiment of the present application, a computer program product is also provided, comprising a computer program, wherein the computer program is executed by a processor to implement the steps of any of the method embodiments described above.

[0020] According to the present application, a panoramic image of an operable area of a desktop robot is acquired based on a visual sensor of a head assembly of the desktop robot, and a to-be-operated position of the desktop robot is determined according to the panoramic image, wherein the to-be-operated position is used to indicate a position at which the desktop robot is to perform a game action; a target rotation angle of a mechanical arm of the desktop robot moving to the to-be-operated position is determined; and the mechanical arm is controlled to move to the to-be-operated position according to the target rotation angle. In the embodiments of the present application, the desktop robot captures images from multiple angles by continuously rotating the head assembly using the visual sensor, and then fuses the images into a panoramic image (i.e., a 360° image) through image stitching technology; based on the panoramic image, the desktop robot can determine the to-be-operated position through image analysis and pattern recognition technology, and then control the desktop robot to play a game with multiple people within a 360° range. Therefore, the problem that a traditional desktop robot can only play a game with one person and has certain limitations can be solved. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application together with the specification.

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced here. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without any creative effort.

[0023] Figure 1 is a hardware structure block diagram of a desktop robot according to a control method of a mechanical arm in the embodiments of the present application;

[0024] Figure 2is a flowchart of a control method of a mechanical arm according to an embodiment of the present application;

[0025] Figure 3 is a schematic diagram of a desktop robot according to an embodiment of the present application (one);

[0026] Figure 4 is a schematic diagram of a desktop robot according to an embodiment of the present application (two);

[0027] Figure 5 is a schematic diagram of a desktop robot according to an embodiment of the present application (three);

[0028] Figure 6 is a structural block diagram of a desktop robot according to an embodiment of the present application. DETAILED DESCRIPTION

[0029] Hereinafter, the embodiments of the present application will be described in detail with reference to the accompanying drawings and in conjunction with embodiments.

[0030] It should be noted that the terms "first", "second", and the like in the description and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence.

[0031] The method embodiments provided in the embodiments of the present application can be executed in a desktop robot or similar computing device. Taking the case of running on a desktop robot, Figure 1 is a hardware structural block diagram of a desktop robot of a control method of a mechanical arm according to an embodiment of the present application. As shown in Figure 1 , the desktop robot can include one or more (only one is shown in Figure 1 ) processor 102 (the processor 102 can include but not limited to a processing device such as a microprocessor MPU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned desktop robot can also include a transmission device 106 for communication function and an input and output device 108. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above-mentioned desktop robot. For example, the desktop robot can also include more or less components than Figure 1 shown, or have a different configuration from Figure 1 shown.

[0032] The memory 104 can be used to store computer programs, such as software programs of application software and modules, such as a computer program corresponding to the control method of the mechanical arm in the embodiments of the present application. The processor 102 can execute various functional applications and data processing, i.e., implement the above method, by running the computer program stored in the memory 104. The memory 104 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include a memory remotely arranged with respect to the processor 102, which can be connected to the desktop robot through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0033] The transmission device 106 is used to receive or send data via a network. The specific examples of the above network can include a wireless network provided by a communication provider of the desktop robot. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet in a wireless manner.

[0034] In the embodiments, a control method of a mechanical arm is provided, which is applied to the above desktop robot, Figure 2 is a flowchart of the control method of the mechanical arm according to the embodiments of the present application, as Figure 2 shown, the flow includes the following steps:

[0035] In step S202, a panoramic image of the operable area of the desktop robot is acquired based on a visual sensor located on a head assembly of the desktop robot, and a to-be-operated position of the desktop robot is determined according to the panoramic image, wherein the to-be-operated position is used to indicate a position at which a game action is to be performed by the desktop robot.

[0036] In step S202, the visual sensor on the head assembly of the desktop robot is used to acquire the panoramic image of the operable area. This is completely different from the traditional desktop robot which can only acquire images in front or at a fixed angle. Acquiring the panoramic image means that the desktop robot can observe the entire desktop without dead angles, including the game state in each direction, such as the distribution of the chessboard and the chess pieces. Based on the panoramic image, the desktop robot can quickly identify different game states and determine the to-be-operated position, i.e., the precise position at which the game action needs to be performed.

[0037] It should be noted that in order to ensure stable signal transmission between the visual system and the body during the infinite cycle rotation of the desktop robot head assembly, a spring contact design is adopted. The rotating contact spring and the body spring seat are in contact through the elastic force. Even in 360° rotation, the electrical connection can be maintained to ensure continuous data transmission. This solves the problem of signal interruption or wire winding that traditional desktop robots may encounter when the head is rotating.

[0038] Step S204, determining the target rotation angle of the mechanical arm of the desktop robot moving to the to-be-operated position;

[0039] It should be noted that the desktop robot monitors the rotation angle of the head through the first absolute value encoder on the head assembly. When the to-be-operated position is determined, the desktop robot calculates the angle (target rotation angle) that the mechanical arm needs to rotate to accurately align the position. Based on the control of the target rotation angle, the mechanical arm can flexibly move to any desired position to perform game actions without being limited to fixed direction operations like traditional desktop robots. This time-sharing control and synchronous scheduling mechanism ensures that the desktop robot can flexibly perform tasks within a 360° range to serve more users and perform more types of desktop games.

[0040] Step S206, controlling the mechanical arm to move to the to-be-operated position according to the target rotation angle.

[0041] Through the above steps, based on the visual sensor located in the head assembly of the desktop robot, a panoramic image of the operable area of the desktop robot is obtained, and the to-be-operated position of the desktop robot is determined according to the panoramic image, wherein the to-be-operated position is used to indicate the position of the desktop robot to be executed game action; the target rotation angle of the mechanical arm of the desktop robot moving to the to-be-operated position is determined; and the mechanical arm is controlled to move to the to-be-operated position according to the target rotation angle. In the embodiment of the present application, the desktop robot captures images from multiple angles by continuously rotating the head assembly using the visual sensor, and then fuses these images into a panoramic image (i.e. 360° image) through image stitching technology; based on the panoramic image, the desktop robot can determine the to-be-operated position through image analysis and pattern recognition technology, and then control the desktop robot to play games with multiple people within a 360° range. Therefore, the problem that traditional desktop robots can only play chess with one person and have certain limitations can be solved.

[0042] Optionally, the step S202 can be implemented by the following method: acquiring partial images respectively captured by the visual sensor at different rotation angles, wherein there is a common field of view between two adjacent partial images; performing a stitching step: extracting feature points in each partial image, and matching feature points in the Nth partial image with feature points in the (N+1)th partial image; calculating a homography transformation matrix of the Nth partial image and the (N+1)th partial image according to the matched feature points; performing homography transformation on the (N+1)th partial image according to the homography transformation matrix of the Nth partial image and the (N+1)th partial image, and stitching the homography-transformed (N+1)th partial image with the Nth partial image, wherein N is a positive integer; and performing the stitching step cyclically until the (Mth) partial image after homography transformation is stitched with the (M-1)th partial image to obtain the panoramic image, wherein the (Mth) partial image is the last partial image acquired by the visual sensor in a target period, and M is a positive integer greater than 1.

[0043] First, the head assembly of the desktop robot captures multiple partial images at different rotation angles using its visual sensor, and these partial images collectively cover the entire operating area. For example, assuming that the field of view of the desktop robot is 90°, in order to cover a 360° area, it needs to capture images at 0°, 90°, 180°, and 270° respectively. Each image is taken from a specific angle and can capture part of the chessboard and the chess pieces in that direction.

[0044] After acquiring the partial images, the desktop robot vision system automatically extracts feature points in each image. Feature points are points in an image that have unique properties, such as corner points, edge points, etc. They serve as anchor points in the image stitching process, helping to identify the correspondence between images. For example, assuming that the Nth image (N=1) captures a corner of the chessboard and part of the chess pieces, while the (N+1)th image (N+1=2) covers the other side of the chessboard. The vision system will identify the feature points of the chessboard corners and chess pieces in both images, then match them to find the same feature point pairs.

[0045] After the feature point matching is completed, the homography transformation matrix between the Nth image and the (N+1)th image is calculated according to the matched feature point pairs. It should be noted that homography transformation is an image transformation technique used to describe the geometric relationship between corresponding points in two images, especially in image stitching to adjust the image perspective so that two images can be aligned. In this embodiment, the calculated homography transformation matrix will be used for the alignment and stitching of subsequent images.

[0046] In the case of calculating the homographic transformation matrix, the calculated homographic transformation matrix is used to adjust the perspective of the N+1th image to align with the Nth image. The adjusted image is spliced with the Nth image to form a larger image. This process is repeated until all local images are spliced together to form a panoramic image covering the entire operating area.

[0047] For example, assuming that the target period is 120° rotation, the desktop robot will start rotating from 0°, and sequentially obtain three local images at 120°, 240°, and 360° (back to 0°). First, the feature points of the 1st image (0°) and the 2nd image (120°) are matched, and the corresponding homographic transformation matrix is calculated, and then the 2nd image is transformed and spliced onto the 1st image. Next, the 2nd and 3rd images (240°) repeat this process until all images are spliced together to form a panoramic image covering a 360° view.

[0048] After obtaining the panoramic image, the desktop robot can perform global observation and analysis on the entire operating area, supporting tabletop games with multiple users in different directions. For example, the desktop robot can simultaneously recognize a chessboard in front, a Gomoku board on the right, and a mahjong on the left, and then according to the game rules and strategies, it can play games with users in different directions respectively.

[0049] Through the above steps, the 360° rotating multi-task desktop robot can overcome the limitations of traditional desktop robots in terms of view angle, achieve omnidirectional visual perception of the operating area, and support simultaneous and multi-task tabletop games with multiple users, greatly enhancing the interactivity and entertainment of the desktop robot.

[0050] Optionally, the "determining the position to be operated by the desktop robot according to the panoramic image" in the above step S202 can be implemented by the following method: according to the first feature vector of each region in the panoramic image and the second feature vector of each region in the historical panoramic image, wherein the historical panoramic image is the previous image of the panoramic image, and each region corresponds to a different process of a tabletop game; determining whether the game state in each region has changed according to the first feature vector and the second feature vector of each region; in the case of determining that there is a target region where the game state has changed, determining whether the desktop robot needs to perform game operation according to the game rules of the tabletop game in the target region; in the case of determining that the desktop robot needs to perform game operation, determining the position corresponding to the target region as the position to be operated by the desktop robot.

[0051] In the embodiments of the present application, a first feature vector of each region is extracted from a current panoramic image, and a second feature vector of each region is extracted from a historical panoramic image (i.e., a previous frame image). The feature vector can include information such as color distribution, texture pattern, arrangement state of chess pieces or cards in the region, and is the basis for the system to determine the game state change.

[0052] The robot can determine whether the game state has changed by comparing the first feature vector and the second feature vector of each region. For example, if the position of a chess piece or a card in a region changes between two frames of images, the game state of the region is considered to have changed.

[0053] When a target region where the game state has changed is determined, the robot further analyzes the specific tabletop game (such as chess, cards, etc.) corresponding to the region and determines whether a game action needs to be performed to respond to the change. For example, a deck of playing cards is identified in the target region and the game state has changed, and the robot determines whether a card needs to be dealt or shuffled.

[0054] If it is determined that a game operation needs to be performed, the robot determines the specific action according to the game rules of the tabletop game in the target region. In a chess game scenario, the robot needs to determine whether to pick up a chess piece, move a chess piece, or put down a chess piece according to the chess rules. In a card game such as poker, the robot needs to determine whether to shuffle, deal, or collect cards according to the game rules.

[0055] After determining the specific game action, the robot determines the position corresponding to the target region as the operation position, i.e., the accurate tabletop coordinates where the game operation needs to be performed. For example, after determining that a card needs to be dealt in a deck of playing cards, the robot calculates the tabletop coordinates required for dealing the card and determines them as the operation position.

[0056] It should be noted that in the above scenario where the tabletop robot cannot accurately determine whether the tabletop robot needs to perform a game operation, the tabletop robot also sends question information to the target object through voice question and answer, and in the case where response information indicating that the tabletop robot needs to perform a game operation is received from the target object, the game operation is performed.

[0057] Suppose in a family gathering, a side of the table is playing a game of chess, and the other side has several users playing a game of poker. The robot continuously collects panoramic images of the table through the vision sensor of its head component.

[0058] When the robot detects that the feature vector of the chessboard region has changed, which may be because the user has moved the chess pieces. The robot determines this region as the target region of the game state change, and judges whether it needs to perform the operation of picking up the chess pieces, moving the chess pieces, etc. according to the rules of the chess game. If necessary, the robot will calculate the new position of the chess pieces, determine it as the to-be-operated position, and dispatch the mechanical arm to perform the corresponding chess piece moving action.

[0059] At the same time, if the robot detects a change in the feature vector in the poker game region, such as a change in the position or number of cards, which may mean that the user has dealt or shuffled the cards. The robot also determines this region as the target region of the game state change, and then judges whether it needs to perform the action of shuffling, dealing or collecting cards according to the rules of the poker game. If necessary, the robot will determine the accurate position of the cards as the to-be-operated position, and dispatch the mechanical arm to perform the operation of shuffling or dealing cards.

[0060] This method of analyzing the feature vector of continuous panoramic images and making intelligent decisions based on game rules enables the robot to efficiently respond to various actions in tabletop games, whether it is a chess game or a card game, and can achieve precise operation and intelligent game interaction, greatly enriching the application scenarios of the robot and the game experience of the user.

[0061] Optionally, the above step S204 can be implemented by: determining a first rotation angle of a head assembly corresponding to a first absolute value encoder when a target local image is collected by the visual sensor, wherein the first absolute value encoder is used to obtain the rotation angle of the head assembly, and the target local image contains the target region; and determining the first rotation angle as the target rotation angle.

[0062] When the tabletop robot head assembly rotates to a certain angle, its visual sensor will collect a local image. If this local image contains the target region that needs to be operated, this image is called the target local image. For example, assuming that the tabletop robot rotates to 240°, and the visual sensor captures the movement of the chess pieces on the chessboard, then the image collected at the 240° angle is the target local image.

[0063] An absolute value encoder is a sensor that can output the current position or angle. Unlike traditional incremental encoders, it can maintain the current position information after power failure without the need for repositioning. The first absolute value encoder equipped on the head assembly of the tabletop robot can monitor and record the rotation angle of the head assembly in real time, providing accurate information on the current viewing angle for the tabletop robot.

[0064] When the visual sensor captures the target local image, the first absolute value encoder records the current rotation angle of the head assembly, i.e., the first rotation angle. This rotation angle is directly related to the position to which the mechanical arm needs to move. If the current position of the mechanical arm is inconsistent with the target operation position (target area position), the desktop robot needs to calculate and determine a target rotation angle so that the mechanical arm can accurately move to the target position to perform the game action.

[0065] If the first rotation angle recorded by the first absolute value encoder is exactly the angle that the desktop robot needs to reach the position to be operated, the first rotation angle itself is directly determined as the target rotation angle. This means that the mechanical arm does not need to rotate additionally and can directly perform the game action.

[0066] After the target rotation angle is determined, the desktop robot controls the mechanical arm to rotate appropriately until the mechanical arm is aligned with the target area.

[0067] Through the embodiments of the present application, the desktop robot can realize efficient linkage of head assembly rotation and mechanical arm control, ensure that the mechanical arm can be aligned with the target area at the correct time and angle to perform the game action. The embodiments of the present application greatly improve the operation flexibility and accuracy of the desktop robot, enable it to accurately respond to the user's desktop game request in different directions, support multiple users to interact at the same time, and increase the interest and participation of the game.

[0068] For example, in a family game, when the desktop robot detects that the chess pieces on the chessboard in front have changed, the visual sensor on the head assembly will capture this change, and the first absolute value encoder will record the current rotation angle, i.e., 270°. If the current orientation of the mechanical arm is consistent with this angle, 270° is directly determined as the target rotation angle, and the mechanical arm does not need to be adjusted and can directly perform the operation, such as picking up or moving the chess pieces, to respond to the user's game action. If the mechanical arm is not at the position of 270°, the desktop robot will control the mechanical arm to rotate according to the target rotation angle until it is aligned with the target area to complete the game action. This precise control mechanism ensures that the desktop robot can flexibly and efficiently interact with the user in the 360° range for desktop games.

[0069] Optionally, the controlling the robot arm to move to the to-be-operated position according to the target rotation angle comprises: determining a second rotation angle of a second absolute value encoder corresponding to the robot arm; determining whether the second rotation angle is consistent with the target rotation angle; in the case that the second rotation angle is consistent with the target rotation angle, determining that the robot arm moves to the to-be-operated position; in the case that the second rotation angle is not consistent with the target rotation angle, controlling the robot arm to move to the to-be-operated position according to the target rotation angle.

[0070] The desktop robot first checks the rotation angle of the current position of the robot arm before moving the robot arm. The angle is provided by the second absolute value encoder corresponding to the robot arm. The absolute value encoder can provide direct information about the current position of the robot arm without the need for cumulative calculation of relative displacement.

[0071] After obtaining the second rotation angle, the desktop robot compares it with the target rotation angle, which is the ideal angle calculated by the desktop robot to move to the to-be-operated position.

[0072] If the second rotation angle is consistent with the target rotation angle, i.e., the robot arm is already aligned with the target position, no further rotation is needed. In this case, the desktop robot can directly control the robot arm to perform the game action.

[0073] If the second rotation angle is not consistent with the target rotation angle, the desktop robot will control the robot arm to make corresponding adjustments according to the target rotation angle to ensure that the robot arm can accurately align with the to-be-operated position. This usually involves calculating the difference angle that the robot arm needs to rotate and achieving this rotation by controlling the robot arm drive motor (such as a stepper motor). The desktop robot will continuously monitor the second absolute value encoder until the robot arm reaches the target rotation angle and aligns with the to-be-operated position.

[0074] Through the embodiments of the present application, it is ensured that the robot arm can accurately respond to the operation requirements of the desktop robot, and can quickly and accurately position to specific positions on the chessboard whether in the initial position or in the moving process. Through real-time monitoring of the second absolute value encoder, the desktop robot can achieve closed-loop control of the robot arm, i.e., adjusting the control signal according to the actual position feedback, which greatly improves the accuracy and stability of the movement of the robot arm.

[0075] In an example embodiment, after the mechanical arm is controlled to move to the to-be-operated position according to the target rotation angle, the method further comprises: determining a table game corresponding to the to-be-operated position; determining a target position of a picking component of the mechanical arm according to rules of the table game and a current state of the table game; and controlling the picking component to pick up a target object at the target position and / or move the target object to the target position, wherein the picking component is configured to grip or release the target object.

[0076] When the mechanical arm reaches the to-be-operated position, the table robot vision system further analyzes the features of the target region to determine the table game being played on the chessboard. Because different table games have different rules and playing methods. For example, the table robot needs to identify the layout of the chessboard, the style of the chess pieces, etc., to determine whether the game being played on the chessboard is chess, go, chess, or gomoku; in a poker game, the robot needs to understand the game round, the player's hand state, and the timing of card play or shuffling, etc.

[0077] After determining the table game, the table robot calculates the target position of the picking component of the mechanical arm according to the rules of the table game and the current state of the target region. That is, the table robot can analyze the current state of the game and evaluate the optimal operation strategy.

[0078] After determining the target position, the table robot controls the picking component on the mechanical arm to perform the action of picking up the target object (e.g., a chess piece, a card) and moving it to the target position. The picking component usually has the ability to grip and release the target object, thereby enabling the table robot to flexibly handle target objects such as chess pieces, cards, mahjong, etc., and move them from the current position to another position. For example, if the table robot identifies that the game being played on the target region is a go game, it will determine the best position to drop the stone according to the rules of go and the current game, then control the picking component to pick up a go stone and place it accurately at the calculated drop position.

[0079] Through the embodiments of the present application, the table robot can autonomously identify table games, analyze the current game state, and perform game actions according to game rules and strategies.

[0080] For example, in a family gathering, the tabletop robot detects that a user moves a chess piece on the chessboard on the left side. The tabletop robot first determines the mechanical arm moves to the operating position of the chessboard through the vision sensor and absolute value encoder system of the head component. After reaching the target position, the tabletop robot determines that the ongoing tabletop game is chess by analyzing the features of the chessboard. Then, the tabletop robot calculates the best position on the chessboard where the mechanical arm pick-up component should move a chess piece of its own to respond to the user's last move according to the rules of chess and the current game state. Finally, the tabletop robot controls the pick-up component to accurately pick up a chess piece and move it to the calculated position, completing a successful move.

[0081] When the robot detects the need for shuffling in the card game area (possibly due to the end of the previous round), it calculates the target position of the pick-up component from the card deck to the shuffling area according to the rules of the card game. Then, the robot controls the pick-up component to perform the shuffling action, mixes the cards evenly, and then according to the game rules, distributes the shuffled cards to each player to ensure the fairness and smoothness of the game.

[0082] Optionally, based on the panoramic image of the operating area of the tabletop robot obtained by the vision sensor located on the head component of the tabletop robot, the panoramic image of the operating area of the tabletop robot is obtained by the vision sensor located on the head component of the tabletop robot, including: determining whether multiple rounds of tabletop games exist in the operating area at the same time; in the case that multiple rounds of tabletop games do not exist in the operating area at the same time, determining a target operating area for the tabletop game, and obtaining an image of the target operating area of the tabletop robot based on the vision sensor of the head component of the tabletop robot; in the case that multiple rounds of tabletop games exist in the operating area at the same time, obtaining a panoramic image of the operating area of the tabletop robot based on the vision sensor located on the head component of the tabletop robot.

[0083] In the embodiments of the present application, the tabletop robot checks whether multiple rounds of tabletop games exist in the operating area (usually the tabletop or game area) at the same time. Optionally, by analyzing the image information obtained by the current vision sensor, the distribution of different chessboards in the area is identified and counted to determine. For example, the tabletop robot vision system can identify the presence of objects on the tabletop, thereby determining whether multiple rounds of tabletop games are being played at the same time.

[0084] If the tabletop robot determines that multiple rounds of tabletop games do not exist in the operating area at the same time, i.e., the tabletop robot only needs to focus on and handle the game on a single area. In this case, the tabletop robot will determine the target operating area for the tabletop game. Then, the tabletop robot will only obtain the image of the target operating area based on the vision sensor of the head component to save computing resources and improve processing efficiency.

[0085] If the tabletop robot detects multiple rounds of tabletop games in the operation area at the same time, the tabletop robot will obtain a panoramic image of the entire operation area based on the visual sensor of its head component. The panoramic image can cover all the information in the area, that is, the tabletop robot can simultaneously process games on multiple areas, identify layout changes of games on each area, and calculate response strategies respectively.

[0086] Through the embodiments of the present application, the tabletop robot can intelligently adjust the image acquisition range of its visual sensor, efficiently analyze the game state and calculate the game action according to the actual situation of the tabletop game in the operation area. The ability of the tabletop robot to intelligently adjust the image acquisition range not only improves the processing efficiency of the tabletop robot, but also ensures that it can accurately and timely respond to the game actions of the user in a complex game environment, supporting simultaneous games with multiple users on different boards.

[0087] In order to better understand the process of the above-mentioned mechanical arm control method, the implementation method flow of the above-mentioned mechanical arm control will be described in combination with optional embodiments below, but not used to limit the technical solutions of the embodiments of the present application.

[0088] In the embodiments of the present application, a tabletop robot is provided, as shown in Figure 3 , Figure 4 and Figure 5 ,

[0089] Figure 3 1-1 in the above-mentioned is a head component (including a visual sensor) of the tabletop robot; 1-2 is a mechanical arm of the tabletop robot; and 1-3 is a center of the head component of the tabletop robot / rotation center of the mechanical arm.

[0090] Figure 4 2-1 in the above-mentioned is a mechanical arm driving step motor module; 2-2 is a motor mounting bracket; 2-3 is a large arm driving shaft support bearing; 2-4 is a mechanical arm skeleton; 2-5 is a motor movement limiting frame; 2-6 is a mechanical arm skeleton shaft support bearing; 2-7 is a mechanical arm shell; 2-8 is an upper damping piece of the motor; 2-9 is a lower damping piece of the motor; and 2-10 is a machine shell.

[0091] Figure 5 3-1 in the above-mentioned is a head component driving step motor module; 3-2 is a head driving gear (17 teeth); 3-3 is a head driven gear (51 teeth); 3-4 is a machine body shell; 3-5 is a head rotation bearing; 3-6 is a head rotation contact spring piece; and 3-7 is a machine body spring piece seat.

[0092] As shown in Figure 3 and Figure 4As shown, the robotic arm drive stepper motor module (2-1) drives the robotic arm (1-2) to rotate around the robotic arm rotation center (1-3): The robotic arm drive stepper motor module (2-1) is mounted on the motor mounting bracket (2-2), and the motor shaft is directly connected to the robotic arm frame (2-4) to realize the driving of the robotic arm. The transmission ratio between the motor and the robotic arm is 1:1.

[0093] like Figure 5 As shown, the head assembly drive stepper motor module (3-1) drives the head assembly (1-1) to rotate around the head assembly center (1-3). The head assembly drive stepper motor module (3-1) is mounted on the head assembly (1-1), and the motor shaft is connected to the head drive gear (3-2), which drives the head driven gear (3-3), thereby realizing the rotation of the head assembly. The transmission ratio of the head gear is 1:3. The head rotation contact spring (3-6) and the body spring seat (3-7) can maintain reliable contact and transmit signals throughout the 360° rotation of the head.

[0094] In this embodiment of the application, to address the risk of breakage caused by the winding of traditional cables during the 360° rotation of the desktop robot's head, a spring contact technology is employed, such as... Figure 5 As shown, through the elastic connection between the rotating contact spring (3-6) on the head assembly and the spring seat (3-7) on the fuselage, the signal transmission can be kept stable even in endless rotation cycles, effectively avoiding the risk of cable breakage, thus achieving an infinite rotation capability of more than 360°.

[0095] To ensure synchronization between the rotation of the head vision sensor and the position of the robotic arm, a multi-encoder system, particularly a combination of absolute encoders, was employed. Each encoder monitors and records the precise angle of the robotic arm's rotation, ensuring accurate matching with the field of view of the vision sensor after rotation in any direction. This design not only enhances the overall flexibility of the system but also guarantees consistency and accuracy between visual inspection and robotic arm execution.

[0096] During task execution, the robotic arm can be instantly moved to the target position by utilizing angle information provided by multiple encoders when the vision system identifies and locates the area to be operated. This ensures the smooth completion of the task even if the vision sensor and the robotic arm are in different rotation states. In scenarios that require simulating a human-versus-many game, a synchronous control strategy is adopted to ensure that the robotic arm and the head vision system maintain the same rotation direction, allowing the desktop robot to seamlessly switch between different games and engage in continuous tabletop game battles with multiple users.

[0097] To overcome the limitation of limited field of view of the camera, in the embodiments of the present application, the camera automatically takes a picture when reaching a certain angle during rotation, capturing the image on the current area. By continuously taking K images, each image shares a part of the field of view with its adjacent image, and by using image processing technology, key feature points such as specific identifiers of SIFT feature points are extracted. These feature points serve as anchor points for image stitching, and through the calculation of homography transformation matrix, K images can be accurately aligned, and finally a complete 360° panoramic image is synthesized. This method not only solves the problem of field of view limitation, but also provides the tabletop robot with full-range game detection capability, enhancing the system's adaptability to complex game environments.

[0098] Through the embodiments of the present application, not only the problem of synchronization between mechanical arm rotation and vision system in traditional design is solved, but also the overall efficiency and flexibility of the system is further improved. By decoupling the vision system and the mechanical arm control, the tabletop robot can continuously position and detect in real time within a 360° range. When the task requirement is clear, the system can quickly position and adjust the mechanical arm to the required position, quickly respond and execute the operation. In the embodiments of the present application, not only the application range of the tabletop robot in multi-task environment is expanded, but also the interactive experience and intelligent performance of the tabletop robot in the tabletop game scene are significantly improved. Through the complex multi-angle task time scheduling algorithm, the task processing process is optimized, realizing the dual improvement of efficiency and accuracy, thereby bringing users a smoother and more intelligent tabletop game experience.

[0099] Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present application.

[0100] In the present embodiment, a tabletop robot is also provided, which is used to implement the above embodiments and preferred embodiments, which have been described. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the tabletop robot described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware implementation is also possible and contemplated.

[0101] Figure 6is a structural block diagram of a desktop robot according to an embodiment of the present application, as shown in the figure, the desktop robot comprises: Figure 6

[0102] a visual sensor 62, configured to acquire a panoramic image of an operable area of the desktop robot;

[0103] a driving module 64, configured to determine a to-be-operated position of the desktop robot according to the panoramic image; determine a target rotation angle of a mechanical arm of the desktop robot moving to the to-be-operated position; and control the mechanical arm to move to the to-be-operated position according to the target rotation angle, wherein the to-be-operated position is used to indicate a position at which the desktop robot is to perform a game action.

[0104] With the desktop robot, the visual sensor located at the head assembly of the desktop robot acquires a panoramic image of an operable area of the desktop robot, and the driving module determines a to-be-operated position of the desktop robot according to the panoramic image, wherein the to-be-operated position is used to indicate a position at which the desktop robot is to perform a game action; determines a target rotation angle of a mechanical arm of the desktop robot moving to the to-be-operated position; and controls the mechanical arm to move to the to-be-operated position according to the target rotation angle. In the embodiment of the present application, the desktop robot continuously rotates the head assembly to capture images from multiple angles by using the visual sensor, and then fuses these images into a panoramic image (i.e., a 360° image) by using image stitching technology; based on the panoramic image, the desktop robot can determine a to-be-operated position by using image analysis and pattern recognition technology, and then control the desktop robot to play chess with multiple people within a 360° range. Therefore, the problem that the traditional desktop robot can only play chess with one person and has certain limitations can be solved.

[0105] In an exemplary embodiment, the desktop robot comprises a head assembly and a body spring piece, wherein the rotating contact spring piece of the head assembly is in contact with the body spring piece seat through elastic force.

[0106] ​In an example embodiment, the visual sensor 62 is configured to acquire partial images respectively captured by the visual sensor at different rotation angles, wherein adjacent two partial images have a common field of view; perform a stitching step of extracting feature points in each partial image and matching the feature points in an Nth partial image with the feature points in an (N+1)th partial image; calculate a homographic transformation matrix of the Nth partial image and the (N+1)th partial image according to the matched feature points; perform homographic transformation on the (N+1)th partial image according to the homographic transformation matrix of the Nth partial image and the (N+1)th partial image, and stitch the homographically transformed (N+1)th partial image with the Nth partial image, wherein N is a positive integer; and repeatedly perform the stitching step until the homographically transformed Mth partial image is stitched with the (M-1)th partial image to obtain the panoramic image, wherein the Mth partial image is the last partial image acquired by the visual sensor in a target period, and M is a positive integer greater than 1.

[0107] In an example embodiment, the driving module 64 is configured to determine whether a game state in each region corresponding to a different process of a tabletop game changes according to a first feature vector of each region in the panoramic image and a second feature vector of each region in a historical panoramic image, wherein the historical panoramic image is a previous image of the panoramic image; determine whether the tabletop robot needs to perform a game operation according to a game rule of a tabletop game in a target region in which it is determined that the game state changes; and determine a position corresponding to the target region as a to-be-operated position of the tabletop robot in a case where it is determined that the tabletop robot needs to perform the game operation.

[0108] In an example embodiment, the driving module 64 is configured to determine a first rotation angle of a head assembly corresponding to a first absolute value encoder for acquiring a rotation angle of the head assembly when a target partial image in which the target region exists is acquired by the visual sensor, and determine the first rotation angle as the target rotation angle.

[0109] In an example embodiment, the driving module 64 is configured to determine a second rotation angle of a second absolute value encoder corresponding to the mechanical arm; determine whether the second rotation angle is consistent with the target rotation angle; determine that the mechanical arm moves to the to-be-operated position in a case where the second rotation angle is consistent with the target rotation angle; and control the mechanical arm to move to the to-be-operated position according to the target rotation angle in a case where the second rotation angle is not consistent with the target rotation angle.

[0110] In an example embodiment, the driving module 64 is configured to determine a tabletop game corresponding to the position to be operated, determine a target position of a picking component of the mechanical arm according to rules of the tabletop game and a current state of the tabletop game, control the picking component to grasp a target object at the target position, and / or control the picking component to move the target object to the target position, wherein the picking component is configured to grasp or release the target object.

[0111] In an example embodiment, the driving module 64 is configured to determine whether multiple rounds of tabletop games exist in the operable region simultaneously, determine a target operating region for playing the tabletop game in a case where the multiple rounds of tabletop games do not exist in the operating region simultaneously, and acquire an image of the target operating region of the tabletop robot by the visual sensor 62, and acquire a panoramic image of the operable region of the tabletop robot in a case where the multiple rounds of tabletop games exist in the operating region simultaneously.

[0112] It should be noted that each of the above modules can be implemented by software or hardware, and for the latter, the following implementation manners can be used, but are not limited thereto: all of the above modules are located in the same processor; or the above modules are located in different processors in any combination.

[0113] Embodiments of the present application also provide a computer readable storage medium, which stores a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

[0114] Optionally, in the present embodiment, the storage medium can be configured to store program code for executing the following steps:

[0115] S1, acquiring a panoramic image of an operable region of a tabletop robot based on a visual sensor located at a head component of the tabletop robot, and determining a position to be operated of the tabletop robot according to the panoramic image, wherein the position to be operated is used to indicate a position at which the tabletop robot is to perform a game action;

[0116] S2, determining a target rotation angle of a mechanical arm of the tabletop robot to move to the position to be operated;

[0117] S3, controlling the mechanical arm to move to the position to be operated according to the target rotation angle.

[0118] In an example embodiment, the computer readable storage medium described above can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0119] Embodiments of the present application also provide an electronic device, comprising a memory and a processor, the memory storing a computer program, and the processor being configured to execute the computer program to perform the steps in any of the method embodiments described above.

[0120] In an example embodiment, the electronic device described above can further comprise a transmission device connected to the processor and an input / output device connected to the processor.

[0121] Optionally, in the embodiment, the processor can be configured to perform the following steps by the computer program:

[0122] S1, acquiring a panoramic image of an operable area of a desktop robot based on a visual sensor of a head assembly of the desktop robot, and determining a to-be-operated position of the desktop robot according to the panoramic image, wherein the to-be-operated position is used to indicate a position at which the desktop robot is to perform a game action;

[0123] S2, determining a target rotation angle of a mechanical arm of the desktop robot moving to the to-be-operated position;

[0124] S3, controlling the mechanical arm to move to the to-be-operated position according to the target rotation angle.

[0125] Embodiments of the present application also provide a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps in any of the method embodiments described above.

[0126] Embodiments of the present application also provide another computer program product, which comprises a non-volatile computer readable storage medium, and the non-volatile computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps in any of the method embodiments described above.

[0127] Embodiments of the present application also provide a computer program, which comprises computer instructions stored in a computer readable storage medium; a processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to perform the steps in any of the method embodiments described above.

[0128] Optionally, in the embodiment, the processor can be configured to execute the following steps by a computer program:

[0129] S1, acquiring a panoramic image of an operable area of the desktop robot based on a visual sensor of a head assembly of the desktop robot, and determining a to-be-operated position of the desktop robot according to the panoramic image, wherein the to-be-operated position is used to indicate a position at which the desktop robot is to perform a game action;

[0130] S2, determining a target rotation angle of a mechanical arm of the desktop robot moving to the to-be-operated position;

[0131] S3, controlling the mechanical arm to move to the to-be-operated position according to the target rotation angle.

[0132] The specific examples in the embodiment can refer to the examples described in the above embodiments and exemplary embodiments, and will not be repeated here.

[0133] Obviously, those skilled in the art should understand that each module or each step of the present application described above can be realized by a general computing device, which can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, which can be realized by program codes executable by the computing device, so that they can be stored in the storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in different order, or they can be manufactured into each integrated circuit module, or multiple modules or steps among them can be manufactured into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.

[0134] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A control method of a robot arm, characterized by, The method comprises: acquiring a panoramic image of an operable area of the tabletop robot based on a vision sensor of a head assembly of the tabletop robot, and determining a to-be-operated position of the tabletop robot according to the panoramic image, wherein the to-be-operated position is used to indicate a position at which the tabletop robot is to perform a game action, and the panoramic image is an image in a range of 0°-360°; determining a target rotation angle at which a mechanical arm of the tabletop robot is to move to the to-be-operated position; controlling the mechanical arm to move to the to-be-operated position according to the target rotation angle; wherein determining the to-be-operated position of the tabletop robot according to the panoramic image comprises: determining, according to a first feature vector of each region in the panoramic image and a second feature vector of each region in a historical panoramic image, whether a game state in each region changes, wherein the historical panoramic image is a previous image of the panoramic image, and each region corresponds to a different process of a tabletop game; determining, according to the first feature vector and the second feature vector of each region, whether a game state in each region changes; in a case where it is determined that there is a target region in which a game state changes, determining whether the tabletop robot needs to perform a game operation according to a game rule of a tabletop game in the target region; in a case where it is determined that the tabletop robot needs to perform a game operation, determining that a position corresponding to the target region is the to-be-operated position of the tabletop robot.

2. The method of claim 1, wherein, Acquiring a panoramic image of an operable area of the tabletop robot based on a vision sensor of a head assembly of the tabletop robot comprises: acquiring local images respectively collected by the vision sensor at different rotation angles, wherein there is a common field of view between two adjacent local images; performing a stitching step: extracting feature points in each local image, and matching feature points in an Nth local image with feature points in an (N+1)th local image; calculating a homographic transformation matrix of the Nth local image and the (N+1)th local image according to the matched feature points; performing homographic transformation on the (N+1)th local image according to the homographic transformation matrix of the Nth local image and the (N+1)th local image, and stitching the homographically transformed (N+1)th local image with the Nth local image, wherein N is a positive integer; recursively performing the stitching step until the homographically transformed Mth local image is stitched with an (M-1)th local image to obtain the panoramic image, wherein the Mth local image is a last local image acquired by the vision sensor within a target period, and M is a positive integer greater than 1.

3. The method of claim 1, wherein, Determining a target rotation angle at which a mechanical arm of the tabletop robot is to move to the to-be-operated position comprises: determining a first rotation angle of a head assembly corresponding to a first absolute value encoder when a target local image is collected by the vision sensor, wherein the first absolute value encoder is used to acquire a rotation angle of the head assembly, and the target region is present in the target local image; determining that the first rotation angle is the target rotation angle.

4. The method of claim 1, wherein, controlling the mechanical arm to move to the target operation position according to the target rotation angle, comprises: determining a second rotation angle of a second absolute value encoder corresponding to the mechanical arm; determining whether the second rotation angle is consistent with the target rotation angle; in the case that the second rotation angle is consistent with the target rotation angle, determining that the mechanical arm moves to the target operation position; in the case that the second rotation angle is not consistent with the target rotation angle, controlling the mechanical arm to move to the target operation position according to the target rotation angle.

5. The method of claim 1, wherein, After controlling the mechanical arm to move to the target operation position according to the target rotation angle, the method further comprises: determining a tabletop game corresponding to the target operation position; determining a target position of a picking assembly of the mechanical arm according to rules of the tabletop game and a current state of the tabletop game; controlling the picking assembly to grab a target object at the target position, and / or controlling the picking assembly to move the target object to the target position, wherein the picking assembly is used to grab or release the target object.

6. The method of claim 1, wherein, acquiring a panoramic image of an operable area of the tabletop robot based on a visual sensor of a head assembly of the tabletop robot, comprises: determining whether multiple rounds of tabletop games exist in the operable area at the same time; in the case that multiple rounds of tabletop games do not exist in the operable area at the same time, determining a target operation area for playing the tabletop game, and acquiring an image of the target operation area of the tabletop robot based on the visual sensor of the head assembly of the tabletop robot; in the case that multiple rounds of tabletop games exist in the operable area at the same time, acquiring the panoramic image of the operable area of the tabletop robot based on the visual sensor of the head assembly of the tabletop robot.

7. A table-top robot characterized by comprises: a visual sensor, configured to acquire a panoramic image of an operable area of the tabletop robot, wherein the panoramic image is an image in a range of 0° to 360°; a driving module, configured to determine a target operation position of the tabletop robot according to the panoramic image, determine a target rotation angle of a mechanical arm of the tabletop robot for moving to the target operation position, and control the mechanical arm to move to the target operation position according to the target rotation angle, wherein the target operation position is used to indicate a position at which the tabletop robot is to perform a game action; wherein the driving module is further configured to determine, according to a first feature vector of each region in the panoramic image and a second feature vector of each region in a historical panoramic image, whether a game state in each region changes, wherein the historical panoramic image is a previous image of the panoramic image, the each region corresponds to a different round of tabletop game; determine, according to the first feature vector and the second feature vector of each region, whether a game state in each region changes; in the case that a target region in which a game state changes is determined to exist, determine, according to game rules of the tabletop game in the target region, whether the tabletop robot needs to perform a game operation; in the case that it is determined that the tabletop robot needs to perform a game operation, determine that a position corresponding to the target region is the target operation position of the tabletop robot.

8. The desktop robot of claim 7, wherein, comprises: The head assembly and the fuselage elastic sheet, wherein the rotating contact elastic sheet of the head assembly and the fuselage elastic sheet seat are in contact by elastic force.

9. An electronic device comprising a memory and a processor, characterized in that The memory stores a computer program, and the processor is configured to execute the method in any one of claims 1 to 6 by the computer program.

Images