A machine dog calibration rack and calibration method

Abstract

The present application relates to a kind of dog calibration frame and calibration method of machine, belong to robot technical field, the calibration frame includes torso support part and leg support part, the leg support part includes mutually connected thigh support plate and shank support plate, the shank support plate is vertically connected to the side of the torso support part and with the top surface of the torso support part parallel, the thigh support plate is vertically connected to the side of the torso support part and with the top surface of the torso support part vertical, the top surface of the torso support part is used to contact with the bottom surface of the torso of machine dog, the thigh support plate is used to contact with the back of the thigh of the machine dog, the shank support plate is used to contact with the bottom surface of the shank of the machine dog.The technical scheme provided by the present application can realize the efficient, accurate calibration of the thigh and shank drive motor of machine dog, to ensure that machine dog accurately executes gait action.

Description

Technical Field

[0001] This invention relates to the field of robotics, specifically to a robot dog calibration frame and calibration method. Background Technology

[0002] With the continuous development of robotics technology, more and more robots are appearing in people's daily lives. For example, quadruped robots, or robot dogs, are becoming increasingly common. To ensure the accuracy of robot operation, calibration is required. For example, after replacing motors or mechanical system parts, after exceeding mechanical limits, or during factory settings, the robot needs to be calibrated.

[0003] Taking the miniature quadruped robot Bittle as an example, its main moving parts are the thighs and lower legs, each driven by a corresponding servo motor. However, gear transmission may cause the thighs and lower legs to be in the correct zero position when they should be, leading to reduced gait and posture stability during subsequent movements. Current calibration methods for similar quadruped robots mainly involve resetting the thighs and lower legs to form an L-shape, then using an L-shaped tool as a reference to observe the deflection angle visually. Finally, the angles of the drive motors for the thighs and lower legs are adjusted based on this deflection angle to achieve calibration. However, this method is not only time-consuming but also has low accuracy. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a robot dog calibration frame and calibration method.

[0005] In a first aspect, the present invention provides a robot dog calibration frame, comprising a torso support and a leg support. The leg support includes a thigh support plate and a calf support plate connected to each other. The calf support plate is perpendicularly connected to the side of the torso support and parallel to the top surface of the torso support. The thigh support plate is perpendicularly connected to the side of the torso support and perpendicular to the top surface of the torso support. The top surface of the torso support is used to contact the bottom surface of the robot dog's torso. The thigh support plate is used to contact the back of the robot dog's thigh. The calf support plate is used to contact the bottom surface of the robot dog's calf.

[0006] Furthermore, the torso support includes a torso support body, a torso longitudinal limiting structure, and a torso lateral limiting structure. The two torso longitudinal limiting structures are respectively disposed at the longitudinal ends of the top surface of the torso support body, and the distance between the two torso longitudinal limiting structures matches the longitudinal dimension of the bottom surface of the robot dog. The two torso lateral limiting structures are respectively disposed at the lateral ends of the top surface of the torso support body, and the distance between the two torso lateral limiting structures matches the lateral dimension of the bottom surface of the robot dog.

[0007] Furthermore, the longitudinal limiting structure of the torso is smoothly connected to the torso support body, and / or the lateral limiting structure of the torso is smoothly connected to the torso support body.

[0008] Furthermore, the leg support also includes a lateral leg limiting structure, which is connected to the thigh support plate and / or the calf support plate and is located at the end of the leg support away from the torso support.

[0009] Secondly, the present invention provides a machine dog calibration method, based on the machine dog calibration frame described above, the machine dog calibration method comprising:

[0010] When the robot dog is placed on the robot dog calibration frame, the robot dog's thigh is controlled to rotate along the first direction by a first set angle, wherein the first set angle is greater than the preset maximum deflection angle.

[0011] The robot dog's thigh is controlled to rotate sequentially along a second direction by a preset angle, wherein the second direction is opposite to the first direction, the preset angle is smaller than the first set angle, and the angular velocity information of the robot dog is obtained each time the preset angle is rotated.

[0012] Based on the angular velocity information, determine whether the robot dog's thigh collides with the thigh support plate of the robot dog calibration frame;

[0013] When a collision is determined, the deflection angle of the robot dog's thigh is determined based on the first set angle and the number of times the preset division angle is rotated when the collision occurs.

[0014] The robot dog's thighs are calibrated based on the thigh deflection angle.

[0015] Furthermore, determining whether the robot dog's thigh collides with the thigh support plate of the robot dog calibration frame based on the angular velocity information includes:

[0016] When performing the action of rotating the preset division angle, the previous angular velocity information before rotation is obtained, and the subsequent angular velocity information after rotation is obtained;

[0017] Whether a collision occurs is determined based on the previous angular velocity information and the subsequent angular velocity information.

[0018] Furthermore, the angular velocity information includes at least one of lateral angular velocity, longitudinal angular velocity, and vertical angular velocity; when the angular velocity information includes lateral angular velocity and longitudinal angular velocity, determining whether a collision has occurred based on the previous angular velocity information and the subsequent angular velocity information includes:

[0019] Determine the first difference between the subsequent lateral angular velocity and the previous lateral angular velocity, and the second difference between the subsequent longitudinal angular velocity and the previous longitudinal angular velocity; determine the square root of the sum of the squares of the first difference and the second difference.

[0020] A collision is determined to have occurred when the square root value is greater than or equal to a preset threshold.

[0021] Furthermore, determining the thigh deflection angle of the robot dog based on the first set angle and the number of times the preset division angle rotates during the collision includes:

[0022] When the preset angle is rotated to the first preset angle and no collision occurs, the action of rotating the preset angle is continued and the first number of rotations is recorded until a collision occurs. The thigh deflection angle is determined based on the first number of rotations and the preset angle.

[0023] Furthermore, determining the deflection angle of the robot dog's thigh based on the first set angle and the number of times the preset division angle rotates during the collision also includes:

[0024] When a collision occurs before the preset angle is reached, the second number of rotations at which the preset angle is rotated is determined, and the thigh deflection angle is determined based on the second number of rotations, the preset angle, and the first preset angle.

[0025] Furthermore, the method also includes:

[0026] When the robot dog’s thigh is controlled to rotate along the first direction at the first set angle, the robot dog’s lower leg is controlled to rotate along the first direction at the second set angle, wherein the second set angle is greater than the preset maximum deflection angle;

[0027] The robot dog's lower legs are controlled to rotate sequentially along the second direction by the preset division angle, wherein the preset division angle is smaller than the second set angle, and the angular velocity information of the robot dog is obtained each time the preset division angle is rotated;

[0028] Based on the angular velocity information, determine whether the robot dog's lower leg collides with the lower leg support plate of the robot dog calibration frame;

[0029] When a collision is determined, the deflection angle of the robot dog's lower leg is determined based on the second set angle and the number of times the preset division angle is rotated when the collision occurs;

[0030] The robot dog's lower legs are calibrated based on the lower leg deflection angle.

[0031] The beneficial effects of the robot dog calibration frame and calibration method of the present invention include: Since the thighs and calves of the robot dog to be calibrated usually have a certain angular deviation, they can be adjusted to set angles and placed on the calibration frame, so that the robot dog's torso fits against the torso support, but the thighs and calves do not contact the calibration frame. Then, the thighs or calves are controlled to move towards the thigh support plate or calves support plate until a collision occurs. The deflection angle can be determined by the aforementioned set angle and the rotation angle of the thigh or calves at this time, and calibration can then be performed based on the deflection angle. Simply placing the robot dog on the calibration frame allows other operations to be completed automatically, unaffected by human error, and achieving high calibration accuracy. This enables efficient and accurate calibration of the robot dog's thigh and calves drive motors, ensuring the robot dog accurately executes gait movements. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a schematic diagram of the structure of the robot dog according to an embodiment of the present invention;

[0034] Figure 2 This is a schematic diagram of the structure of the angular velocity sensor according to an embodiment of the present invention;

[0035] Figure 3 This is a schematic diagram of the structure of the machine dog calibration frame according to an embodiment of the present invention;

[0036] Figure 4 This is a schematic diagram of the structure of the robot dog placed on the robot dog calibration frame according to an embodiment of the present invention;

[0037] Figure 5 This is a schematic diagram illustrating the calibration of the robot dog's thigh according to an embodiment of the present invention;

[0038] Figure 6 This is a schematic diagram illustrating the calibration of the robot dog's lower legs according to an embodiment of the present invention;

[0039] Figure 7 This is a schematic flowchart of the machine dog calibration method according to an embodiment of the present invention;

[0040] Figure 8 This is a schematic diagram of collision detection in the robot dog calibration method according to an embodiment of the present invention. Detailed Implementation

[0041] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0042] The attached diagram includes XYZ coordinate axes. The X-axis represents the longitudinal direction, or front-to-back direction, with the positive direction representing the front and the negative direction representing the rear. The Y-axis represents the transverse direction, or left-to-right direction, with the positive direction representing the left and the negative direction representing the right. The Z-axis represents the vertical direction, or up-down direction, with the positive direction representing the top and the negative direction representing the bottom. It should be noted that the aforementioned representations of the X, Y, and Z axes are for ease of description and simplification of the invention, 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. Therefore, they should not be construed as limitations on the invention.

[0043] like Figure 1 As shown, taking the miniature quadruped robot Bittle as an example, it mainly consists of a torso 33, legs, and a head. The head can move relative to the torso 33 as follows: Figure 1 The rotation is in the direction indicated by the middle arrow, but the more significant movement is usually in the legs, and the calibration target of this invention is also primarily the legs. The legs include the thigh 31 and the lower leg 32, and the thigh 31 can rotate relative to the torso 33 as shown by the middle arrow. Figure 1 Rotation in the direction indicated by the middle arrow allows the lower leg 32 to move relative to the thigh 31 as follows: Figure 1 The rotation is in the direction indicated by the middle arrow. The thigh 31 and lower leg 32 are each driven by a corresponding servo motor, but the transmission mechanism includes gears, which may cause the thigh 31 and lower leg 32 to be in the zero position when they should be. Specifically, Figure 1 This can be considered the robot dog's reset state, or in other words, the state in which the robot dog is usually located at the factory or during calibration. In this state, the thigh 31 can be considered perpendicular to the torso 33, and the lower leg 32 can be considered perpendicular to the thigh 31. If accurate calibration is not performed, both the thigh 31 and the lower leg 32 may deviate from the ideal reset state, for example, by turning forward or upward at a certain angle. This will lead to a decrease in the stability of the robot dog's gait and posture when performing subsequent actions.

[0044] In addition, with Figure 1 In the state shown, the surface at the bottom of the torso 33 is defined as its bottom surface 331, the surface at the back of the thigh 31 is defined as its back surface 311, and the surface at the bottom of the calf 32 is defined as its bottom surface 321.

[0045] like Figure 3 and Figure 4As shown, the robot dog calibration frame of this embodiment includes a torso support 1 and a leg support 2. The leg support 2 includes a thigh support plate 21 and a calf support plate 22 connected to each other. The calf support plate 22 is vertically connected to the side of the torso support 1 and parallel to the top surface of the torso support 1. The thigh support plate 21 is vertically connected to the side of the torso support 1 and perpendicular to the top surface of the torso support 1. The top surface of the torso support 1 is used to contact the bottom surface 331 of the robot dog's torso 33. The thigh support plate 21 is used to contact the back 311 of the robot dog's thigh 31. The calf support plate 22 is used to contact the bottom surface 321 of the robot dog's calf 32.

[0046] Specifically, when the robot dog is in the reset state, or in other words, when the robot dog's thigh 31 is perpendicular to the torso 33 and the lower leg 32 is perpendicular to the thigh 31, the robot dog will be in contact with the corresponding parts of the calibration frame. That is, the back 311 of the thigh 31 is in contact with the thigh support plate 21, the bottom surface 321 of the lower leg 32 is in contact with the lower leg support plate 22, and the bottom surface 331 of the torso 33 is in contact with the torso support part 1. When calibrating the thigh 31 and lower leg 32, after the reset operation, the above-mentioned contact or fit is usually not achieved because there may be a certain deflection angle. Using a calibration frame, the thigh 31 and lower leg 32 can be rotated by a set angle to ensure that the bottom surface 331 of the torso 33 remains in contact with the torso support 1, but the thigh 31 and lower leg 32 do not contact the calibration frame. Taking the thigh 31 as an example, the thigh 31 is then rotated towards the thigh support plate 21, and the rotation angle of the thigh 31 when it collides with the thigh support plate 21 is recorded. Since the collision between the thigh 31 and the thigh support plate 21 represents that the two are just in contact, that is, in an ideal reset state, the deflection angle of the thigh before calibration can be determined by the above rotation angle and the set angle, for example, by subtracting or summing them, and then calibration can be performed using this deflection angle. The calibration process for the lower leg 32 is similar, and it can be performed after the thigh 31 calibration is completed.

[0047] In this embodiment, since the thighs and calves of the robot dog to be calibrated typically have a certain angular deviation, they can be adjusted to set angles and placed on the calibration frame, ensuring the robot dog's torso is in contact with the torso support, but the thighs and calves do not contact the calibration frame. Then, the thigh or calves are controlled to move towards the thigh support plate or calves support plate until a collision occurs. The deflection angle can be determined by the aforementioned set angle and the rotation angle of the thigh or calves at this point, and calibration can then be performed based on the deflection angle. Simply placing the robot dog on the calibration frame allows other operations to be completed automatically, unaffected by human error, resulting in high calibration accuracy. This enables efficient and accurate calibration of the robot dog's thigh and calves drive motors, ensuring the robot dog accurately executes gait movements.

[0048] Optionally, the torso support 1 includes a torso support body, a torso longitudinal limiting structure 11, and a torso lateral limiting structure 12. The two torso longitudinal limiting structures 11 are respectively disposed at the longitudinal ends of the top surface of the torso support body, and the distance between the two torso longitudinal limiting structures 11 matches the longitudinal dimension of the bottom surface of the robot dog. The two torso lateral limiting structures 12 are respectively disposed at the lateral ends of the top surface of the torso support body, and the distance between the two torso lateral limiting structures 12 matches the lateral dimension of the bottom surface of the robot dog.

[0049] Specifically, during the calibration process, the robot dog's thigh 31 will collide with the thigh support plate 21, and its lower leg 32 will collide with the lower leg support plate 22. Figure 3 As shown, longitudinal torso limiting structures 11 are respectively provided at the front and rear ends of the top surface of the torso support body. These structures can ensure that the torso 33 of the robot dog is relatively stable on the calibration frame when no collision occurs, and thus can be accurately identified when a collision occurs.

[0050] Meanwhile, the torso lateral limiting structures 12 set at the left and right rear ends of the top surface of the torso support body can also ensure that the robot dog's torso 33 is stably positioned on the calibration frame.

[0051] Optionally, the longitudinal limiting structure 11 of the torso is smoothly connected to the torso support body, and / or the lateral limiting structure 12 of the torso is smoothly connected to the torso support body.

[0052] Specifically, when the thigh 31 collides with the thigh support plate 21, or the lower leg 32 collides with the lower leg support plate 22, it can be detected by corresponding sensors. At this time, the robot dog's torso 33 will mainly move longitudinally and vertically. Setting the limiting structure and the torso support body as a smooth transition connection can effectively protect the robot dog's corresponding drive structure, such as the thigh drive motor or the lower leg drive motor and related structures, from damage due to collision.

[0053] Optionally, the leg support 2 further includes a leg lateral limiting structure 23, which is connected to the thigh support plate 21 and / or the calf support plate 22 and is located at the end of the leg support 2 away from the torso support 1.

[0054] Specifically, the lateral leg limiting structure 23 can be located at the end of the leg support 2 away from the torso support 1. It can be connected to the thigh support plate 21 and the lower leg support plate 22. Together with the torso support 1, the three form the receiving space or support area for the robot dog's legs, thus limiting the robot dog's lateral movement and facilitating accurate detection of collisions. The thigh support plate 21 and the lower leg support plate 22 can also be triangular or trapezoidal to increase the overall structural strength.

[0055] like Figure 7 As shown, another embodiment of the present invention, a machine dog calibration method based on the machine dog calibration frame described above, includes the following steps:

[0056] When the robot dog is placed on the robot dog calibration frame, the robot dog's thigh 31 is controlled to rotate along the first direction by a first set angle, wherein the first set angle is greater than the preset maximum deflection angle.

[0057] Specifically, due to potential angular deviations in the joints of the robot dog, when the robot dog is adjusted to the reset state, its thigh 31 may not be perpendicular to the torso 33, or in other words, it may not be able to fit or contact the thigh support plate 31 of the calibration frame. For example... Figure 5 As shown, taking the left hind leg of the robot dog as an example, the vertical downward direction of the thigh drive motor is set as the zero point. At this time, the thigh 31 is controlled to rotate clockwise by a first set angle as shown in the figure. According to past data, the maximum deflection angle of the robot dog's thigh is, for example, 10°. Therefore, when the first set angle is set to, for example, 15°, regardless of whether the thigh 31 was previously deflected clockwise or counterclockwise relative to the zero point, it is now located in the clockwise position of the thigh support plate 21 and does not interfere with the thigh support plate 21.

[0058] The robot dog's thigh 31 is controlled to rotate sequentially along a second direction by a preset angle, wherein the second direction is opposite to the first direction and the preset angle is smaller than the first set angle, and the angular velocity information of the robot dog is obtained each time the preset angle is rotated.

[0059] Specifically, the preset division angle can be set according to the parameters of the robot dog's drive device. For example, if the angle of the robot dog's servo gear is 1° per revolution, the preset division angle can be set to 1°. Then, the thigh 31 is rotated counterclockwise one after another, or its servo is controlled to rotate, and the angular velocity information of the robot dog is monitored during the rotation.

[0060] Based on the angular velocity information, determine whether the robot dog's thigh 31 collides with the thigh support plate 21 of the robot dog calibration frame.

[0061] Specifically, a device such as... can be installed on the robot dog's torso 33. Figure 2The angular velocity monitoring chip shown can monitor longitudinal angular velocity (X-axis angular velocity), which can be calculated as roll angle (Roll); lateral angular velocity (Y-axis angular velocity), which can be calculated as pitch angle (Pitch); and vertical angular velocity (Z-axis angular velocity), which can be calculated as yaw angle (Yaw). Changes in angular velocity can determine whether the robot dog's state has changed, specifically whether the thigh 31 has collided with the thigh support plate 21 of the robot dog calibration frame.

[0062] When a collision is determined, the deflection angle of the robot dog's thigh is determined based on the first set angle and the number of times the preset division angle is rotated when the collision occurs.

[0063] Specifically, if the thigh 31 rotates to fit against the thigh support plate 21, it can be considered that no collision has occurred. However, if the thigh 31 continues to rotate by a preset division angle, the two will collide. The state before the collision can be considered as the theoretical reset state of the thigh 31. At this time, the thigh deflection angle of the robot dog can be determined by the total angle of the first set angle and the preset division angle that has been rotated.

[0064] The thigh 31 of the robot dog is calibrated according to the thigh deflection angle.

[0065] Specifically, after determining the thigh deflection angle, calibration can be performed through the thigh motor controller. For example, the actual thigh in the reset state can be rotated by a certain angle to achieve the theoretical reset state. However, the actual thigh in the reset state has a certain angle deviation compared to the theoretical reset state.

[0066] In this embodiment, since the thighs and calves of the robot dog to be calibrated typically have a certain angular deviation, taking the thigh as an example, it can be adjusted to a set angle and placed on the calibration frame, so that the robot dog's torso is in contact with the torso support, but the thigh does not contact the calibration frame. Then, the thigh is controlled to move towards the thigh support plate until a collision occurs. The deflection angle can be determined by the aforementioned set angle and the rotation angle of the thigh at this time, and then calibration is performed based on the deflection angle. Simply placing the robot dog on the calibration frame allows other operations to be completed automatically, without being affected by human error. The calibration accuracy is high, enabling efficient and accurate calibration of the robot dog's thigh and calf drive motors to ensure that the robot dog accurately executes gait movements.

[0067] Optionally, determining whether the robot dog's thigh 31 collides with the thigh support plate 21 of the robot dog calibration frame based on the angular velocity information includes:

[0068] When performing the action of rotating the preset division angle, the previous angular velocity information before rotation is obtained, and the subsequent angular velocity information after rotation is obtained.

[0069] Specifically, since the process of rotating the thigh 31 counterclockwise is usually a continuous process, or rather, it usually rotates multiple preset division angles, the angular velocity information of the robot dog during each rotation is acquired.

[0070] Whether a collision occurs is determined based on the previous angular velocity information and the subsequent angular velocity information.

[0071] Specifically, when the thigh 31 collides with the thigh support plate 21, the robot dog itself will move accordingly, or in other words, the robot dog's torso 33 will change from its contact state with the torso support 1. At this time, its angular velocity will change significantly. Therefore, the change in angular velocity information between the two previous and subsequent collisions can be used to determine whether a collision has occurred.

[0072] Optionally, the angular velocity information includes at least one of lateral angular velocity, longitudinal angular velocity, and vertical angular velocity; when the angular velocity information includes lateral angular velocity and longitudinal angular velocity, determining whether a collision has occurred based on the previous angular velocity information and the subsequent angular velocity information includes:

[0073] Determine the first difference between the subsequent lateral angular velocity and the previous lateral angular velocity, and the second difference between the subsequent longitudinal angular velocity and the previous longitudinal angular velocity, and determine the square root of the sum of the squares of the first difference and the second difference.

[0074] A collision is determined to have occurred when the square root value is greater than or equal to a preset threshold.

[0075] Specifically, whether a collision has occurred can be determined by one, two, or all three of the lateral angular velocity, longitudinal angular velocity, and vertical angular velocity.

[0076] For example, the determination is made using both lateral angular velocity and longitudinal angular velocity. Figure 8 As shown, both the longitudinal angular velocity and the lateral angular velocity, or the roll angle and pitch angle obtained by calculation, exhibit a step change when a collision occurs. However, since there is some noise before a collision occurs, both the lateral angular velocity and the longitudinal angular velocity are used for collision judgment.

[0077] More specifically, it can be calculated using the following formula:

[0078]

[0079] Where pr represents the square root value, p l p represents the subsequent transverse angular velocity or the corresponding angle. l-1 r represents the previous lateral angular velocity or corresponding angle. l Indicates the subsequent longitudinal angular velocity or corresponding angle, r l-1 This indicates the previous longitudinal angular velocity or the corresponding angle.

[0080] This method can more accurately determine whether a collision has occurred.

[0081] Optionally, determining the thigh deflection angle of the robot dog based on the first set angle and the number of times the preset division angle rotates during the collision includes:

[0082] When the preset angle is rotated to the first preset angle and no collision occurs, the action of rotating the preset angle is continued and the first number of rotations is recorded until a collision occurs. The thigh deflection angle is determined based on the first number of rotations and the preset angle.

[0083] Specifically, continuing the previous example, if the actual deflection angle of thigh 31 is 8° clockwise relative to zero, then after rotating clockwise by the first preset angle of 15°, the deflection angle of the thigh relative to zero is 23° clockwise. Then, rotating counterclockwise 23 times with a preset deflection angle of 1°, that is, exceeding 15°, allows thigh 31 to adhere to or contact thigh support plate 21; a collision occurs after 24 rotations. At this point, multiplying 1° by 23 and subtracting from 15° yields a thigh deflection angle of 8°. Alternatively, one can multiply 8 (the number of rotations from 15° to the point of collision) by 1° to obtain a thigh deflection angle of 8°, with the direction being clockwise.

[0084] Optionally, determining the thigh deflection angle of the robot dog based on the first set angle and the number of times the preset division angle rotates during the collision further includes:

[0085] When a collision occurs before the preset angle is reached, the second number of rotations at which the preset angle is rotated is determined, and the thigh deflection angle is determined based on the second number of rotations, the preset angle, and the first preset angle.

[0086] Specifically, continuing the previous example, if the actual deflection angle of thigh 31 is 9° counterclockwise relative to zero, then after rotating clockwise by the first preset angle of 15°, the thigh's deflection angle relative to zero is 6° clockwise. Then, rotating counterclockwise 6 times with a preset deflection angle of 1°, before reaching 15°, achieves the contact or fit between thigh 31 and thigh support plate 21; a collision occurs after 7 rotations. At this point, multiplying 1° by 6 and subtracting from 15° yields a thigh deflection angle of 9°, with the orientation being counterclockwise.

[0087] Optionally, the method further includes the following steps:

[0088] When the robot dog’s thigh 31 is controlled to rotate along the first direction at the first set angle, the robot dog’s lower leg 32 is controlled to rotate along the first direction at the second set angle, wherein the second set angle is greater than the preset maximum deflection angle.

[0089] The robot dog's lower leg 32 is controlled to rotate sequentially along the second direction by the preset division angle, wherein the preset division angle is smaller than the second set angle, and the angular velocity information of the robot dog is obtained each time the preset division angle is rotated.

[0090] Based on the angular velocity information, determine whether the lower leg 32 of the robot dog collides with the lower leg support plate 22 of the robot dog calibration frame.

[0091] When a collision is determined, the deflection angle of the robot dog's lower leg is determined based on the second set angle and the number of times the preset division angle is rotated when the collision occurs.

[0092] The lower leg 32 of the robot dog is calibrated according to the lower leg deflection angle.

[0093] Specifically, such as Figure 6 As shown, the lower leg 32 can be calibrated after the thigh 31 is calibrated. Alternatively, the lower leg 32 can be rotated clockwise first, and then counterclockwise by a preset angle. The specific operation process is similar to that of calibrating the thigh 32.

[0094] Readers should understand that in the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0095] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method of calibrating a robot dog, characterized by, Based on the robot dog calibration frame, the robot dog calibration frame includes a torso support (1) and a leg support (2). The leg support (2) includes a thigh support plate (21) and a lower leg support plate (22) connected to each other. The lower leg support plate (22) is vertically connected to the side of the torso support (1) and parallel to the top surface of the torso support (1). The thigh support plate (21) is vertically connected to the side of the torso support (1) and perpendicular to the top surface of the torso support (1). The top surface of the torso support (1) is used to contact the bottom surface (331) of the robot dog's torso (33). The thigh support plate (21) is used to contact the back (311) of the robot dog's thigh (31). The lower leg support plate (22) is used to contact the bottom surface (321) of the robot dog's lower leg (32). The machine dog calibration method includes: When the robot dog is placed on the robot dog calibration frame, the robot dog's thigh (31) is controlled to rotate along the first direction by a first set angle, wherein the first set angle is greater than the preset maximum deflection angle; Control the thigh (31) of the robot dog to rotate a preset division angle in a second direction, wherein the second direction is opposite to the first direction, the preset division angle is smaller than the first set angle, and obtain the angular velocity information of the robot dog when rotating the preset division angle once; Based on the angular velocity information, determine whether the thigh (31) of the robot dog collides with the thigh support plate (21) of the robot dog calibration frame; When a collision is determined, the deflection angle of the robot dog's thigh is determined based on the first set angle and the number of times the preset division angle is rotated when the collision occurs. The thigh (31) of the robot dog is calibrated according to the thigh deflection angle; When the robot dog’s thigh (31) is controlled to rotate along the first direction at the first set angle, the robot dog’s lower leg (32) is controlled to rotate along the first direction at the second set angle, wherein the second set angle is greater than the preset maximum deflection angle; The robot dog's lower leg (32) is controlled to rotate the preset division angle sequentially along the second direction, wherein the preset division angle is smaller than the second set angle, and the angular velocity information of the robot dog is obtained when the preset division angle is rotated once; Based on the angular velocity information, determine whether the lower leg (32) of the robot dog collides with the lower leg support plate (22) of the robot dog calibration frame; When a collision is determined, the deflection angle of the robot dog's lower leg is determined based on the second set angle and the number of times the preset division angle is rotated when the collision occurs; The lower leg (32) of the robot dog is calibrated according to the lower leg deflection angle.

2. The machine dog calibration method of claim 1, wherein, The torso support part (1) includes a torso support body, a torso longitudinal limiting structure (11) and a torso lateral limiting structure (12). The two torso longitudinal limiting structures (11) are respectively disposed at the longitudinal ends of the top surface of the torso support body. The distance between the two torso longitudinal limiting structures (11) matches the longitudinal dimension of the bottom surface of the robot dog. The two torso lateral limiting structures (12) are respectively disposed at the lateral ends of the top surface of the torso support body. The distance between the two torso lateral limiting structures (12) matches the lateral dimension of the bottom surface of the robot dog.

3. The machine dog calibration method of claim 2, wherein, The longitudinal limiting structure (11) of the torso is smoothly connected to the torso support body, and / or the lateral limiting structure (12) of the torso is smoothly connected to the torso support body.

4. The machine dog calibration method of claim 1, wherein, The leg support (2) further includes a leg lateral limiting structure (23), which is connected to the thigh support plate (21) and / or the calf support plate (22) and is located at the end of the leg support (2) away from the torso support (1).

5. The machine dog calibration method of claim 1, wherein, The step of determining whether the robot dog's thigh (31) collides with the thigh support plate (21) of the robot dog calibration frame based on the angular velocity information includes: When performing the action of rotating the preset division angle, the previous angular velocity information before rotation is obtained, and the subsequent angular velocity information after rotation is obtained; Whether a collision occurs is determined based on the previous angular velocity information and the subsequent angular velocity information.

6. The machine dog calibration method according to claim 5, wherein, The angular velocity information includes at least one of lateral angular velocity, longitudinal angular velocity, and vertical angular velocity; when the angular velocity information includes lateral angular velocity and longitudinal angular velocity, determining whether a collision has occurred based on the previous angular velocity information and the subsequent angular velocity information includes: Determine the first difference between the subsequent lateral angular velocity and the previous lateral angular velocity, and the second difference between the subsequent longitudinal angular velocity and the previous longitudinal angular velocity; determine the square root of the sum of the squares of the first difference and the second difference. A collision is determined to have occurred when the square root value is greater than or equal to a preset threshold.

7. The machine dog calibration method of claim 1, wherein, Determining the thigh deflection angle of the robot dog based on the first set angle and the number of times the preset division angle rotates during the collision includes: When the preset angle is rotated to the first set angle and no collision occurs, the action of rotating the preset angle is continued and the first number of rotations is recorded until a collision occurs. The thigh deflection angle is determined based on the first number of rotations and the preset angle.

8. The machine dog calibration method according to claim 7, wherein, The step of determining the thigh deflection angle of the robot dog based on the first set angle and the number of times the preset division angle rotates during the collision further includes: When a collision occurs before the preset division angle is reached, the second number of rotations at which the preset division angle is rotated is determined, and the thigh deflection angle is determined based on the second number of rotations, the preset division angle, and the first set angle.

Images