Robot pose control methods, devices, storage media, and electronic devices

Abstract

The application discloses a pose control method and device of a robot, a storage medium and an electronic device, and relates to the technical field of robots. The method comprises the following steps: acquiring current motion data of the robot; obtaining a current position deviation value of the robot according to the current motion data by using a pre-established search database model, wherein the search database model comprises a corresponding relationship between historical motion data and historical position deviation values of the robot; and adjusting the pose of the robot according to the current position deviation value. The method adjusts the pose of the robot by using historical motion data fed back by a power mechanism of the robot, directly analyzes and learns the motion of the robot from a perception perspective, and predicts in advance, through self-learning, what kind of motion trend will be generated in the case that the robot is subjected to what kind of force during operation, so as to make a preliminary judgment and then perform fine correction actions, so that the robot achieves the purpose of intelligent, flexible and rapid control of the motion pose.

Description

Technical Field

[0001] This invention relates to the field of posture control technology, specifically to a posture control method, device, storage medium, and electronic device for a robot. Background Technology

[0002] In related technologies, robot pose correction methods often use sensor input (laser, vision, GNSS, inertial sensors, etc.) to obtain the robot's real-time position. Then, based on the theoretically designed route, the route deviation is calculated and the posture is corrected. The goal is to make the robot's actual running trajectory closely follow the theoretically designed route to achieve the target function (such as inspection or operation in a designated area). However, when the robot's movement deviates (such as turning or spinning), the correction is slow and it has poor adaptability to slopes or uneven environments. Summary of the Invention

[0003] This invention aims to at least partially solve one of the technical problems in related technologies. Therefore, one objective of this invention is to propose a robot pose control method that, through motion perception and self-learning, corrects deviations during movement, enabling the robot to achieve intelligent, flexible, and rapid control of its own motion pose.

[0004] The second objective of this invention is to provide a pose control device.

[0005] A third objective of this invention is to provide a computer-readable storage medium.

[0006] The fourth objective of this invention is to provide an electronic device.

[0007] To achieve the above objectives, a first aspect of the present invention provides a robot pose control method, the method comprising: acquiring current motion data of the robot; using a pre-established search database model to obtain a current position deviation value of the robot based on the current motion data, wherein the search database model includes a correspondence between historical motion data of the robot and historical position deviation values; and adjusting the pose of the robot based on the current position deviation value.

[0008] According to the robot pose control method of the present invention, the robot pose is adjusted by using historical motion data fed back by the robot's power mechanism. The robot's motion is directly analyzed and learned from the perspective of perception. Through self-learning, the robot can predict in advance what kind of force it is subjected to during operation and what kind of motion trend will occur. After making advance predictions, the robot can make subtle correction actions, so as to achieve the purpose of intelligent, flexible and rapid control of its own motion pose.

[0009] In addition, the robot pose control method proposed in the above embodiments of the present invention may also have the following additional technical features:

[0010] According to an embodiment of the present invention, the process of establishing the search database model includes: acquiring structure data, wherein the structure data includes a preset number of historical motion data, and the acquisition time and historical position deviation value corresponding to each historical motion data; performing deduplication processing on the structure data, and establishing the search database model based on the remaining structure data; or, performing group processing on the structure data, and establishing the search data model based on the grouped structure data.

[0011] According to an embodiment of the present invention, the deduplication process of the structure data includes: deduplicating structure data in which historical motion data and corresponding historical position deviation values ​​are repeated; the grouping process of the structure data includes: grouping the structure data according to the magnitude of the historical position deviation values.

[0012] According to one embodiment of the present invention, obtaining the current position deviation value of the robot based on the current motion data using a pre-established search database model includes: when a position deviation value is obtained based on the current motion data using the pre-established search database model, the position deviation value is used as the current position deviation value; when multiple position deviation values ​​are obtained based on the current motion data using the pre-established search database model, the current position deviation value is obtained based on the multiple position deviation values ​​and the current pose data of the robot.

[0013] According to one embodiment of the present invention, obtaining the current position deviation value based on the plurality of position deviation values ​​and the current pose data of the robot includes: obtaining the current route of the robot based on the current pose data; determining the robot's front orientation based on the current route and the theoretical route; and determining the current position deviation value based on the plurality of position deviation values ​​and the front orientation using the relationship between position deviation values ​​and front orientation in a pre-established search database model.

[0014] According to one embodiment of the present invention, obtaining the current position deviation value of the robot based on the current motion data using a pre-established search database model includes: when there is no historical motion data identical to the current motion data in the search database model, determining, based on the search database model, the historical motion data closest to the current motion data from historical motion data smaller than the current motion data and historical motion data larger than the current motion data, respectively, and denoting them as the first historical motion data and the second historical motion data; obtaining the current position deviation value based on the first historical position deviation value corresponding to the first historical motion data and the second historical position deviation value corresponding to the second historical motion data.

[0015] According to one embodiment of the present invention, adjusting the pose of the robot based on the current position deviation value includes: establishing a virtual target coordinate point based on the current position deviation value; obtaining a desired linear velocity and a desired angular velocity based on the virtual target coordinate point and a theoretical positioning point; obtaining a desired rotational speed of the robot motor based on the desired linear velocity and the desired angular velocity using the robot's motion model; and controlling the motor to run at the desired rotational speed.

[0016] According to one embodiment of the present invention, the motion data includes at least one of pressure data and acceleration data.

[0017] To achieve the above objectives, a second aspect of the present invention provides a robot pose control device, the device comprising: an acquisition module for acquiring current motion data of the robot; a search module for obtaining a current position deviation value of the robot based on the current motion data using a pre-established search database model, wherein the search database model includes a correspondence between historical motion data of the robot and historical position deviation values; and a control module for adjusting the pose of the robot based on the current position deviation value.

[0018] To achieve the above objectives, a third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the pose control method of the robot as described in any one of claims 1-7.

[0019] To achieve the above objectives, a fourth aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the robot pose control method as proposed in the first aspect of the present invention.

[0020] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0021] Figure 1 This is a flowchart of a robot pose control method according to an embodiment of the present invention;

[0022] Figure 2 This is a flowchart illustrating the establishment of a search database model according to an embodiment of the present invention;

[0023] Figure 3 This is a flowchart illustrating the adjustment of a robot's pose according to an embodiment of the present invention;

[0024] Figure 4 This is a schematic diagram illustrating the correction of a robot during linear operation according to an embodiment of the present invention;

[0025] Figure 5 This is a schematic diagram illustrating the correction of a robot's spin in place according to an embodiment of the present invention.

[0026] Figure 6 This is a schematic diagram of a pose control device according to an embodiment of the present invention. Detailed Implementation

[0027] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0028] The robot pose control method, device, storage medium, and electronic device of the present invention will be described in detail below with reference to the specifications 1-6 and specific embodiments.

[0029] Figure 1 This is a flowchart illustrating a robot pose control method according to an embodiment of the present invention. Figure 1 As shown, the robot's pose control method may include:

[0030] S1, acquire the robot's current motion data.

[0031] In an embodiment of the present invention, the current motion data is data fed back from the robot's power mechanism. The robot's power mechanism may include a drive motor, a transmission mechanism, and sensors attached to it (e.g., force sensors, encoders, IMU (Inertial Measurement Unit), RTK (Real Time Kinematic) positioning sensors, etc.).

[0032] S2. Using a pre-established search database model, the robot's current position deviation value is obtained based on the current motion data. The search database model includes the correspondence between the robot's historical motion data and historical position deviation values.

[0033] In embodiments of the present invention, the pre-established search database model includes the correspondence between the robot's historical motion data and historical position deviation values. The historical motion data also includes data fed back from the robot's power mechanism.

[0034] It should be noted that the historical motion data type is consistent with the acquired current motion data type of the robot, or the historical motion data type includes the acquired current motion data type of the robot. This is to facilitate the search for the current position deviation value based on the current motion data.

[0035] It is feasible to use a pre-established search database model consistent with the current motion data type to search for the current position deviation value of the current motion data, and then adjust the robot's pose based on the current position deviation value.

[0036] S3, adjust the robot's pose based on the current position deviation value.

[0037] It is feasible to adjust the robot drive mechanism according to the current position deviation value to adjust the robot's pose so that the robot's actual running trajectory is consistent with the theoretical running trajectory.

[0038] The robot pose control method of this invention uses historical motion data fed back from the robot's power mechanism to adjust the robot's pose. It directly analyzes and learns the robot's motion from a perception perspective. Through self-learning, it can predict in advance what kind of force the robot will be subjected to during operation and what kind of motion trend will be generated. After making advance predictions, it can perform subtle correction actions, so as to achieve the purpose of intelligent, flexible and rapid control of the robot's own motion pose.

[0039] In one embodiment of the present invention, motion data may include at least one of pressure data and acceleration data.

[0040] In one embodiment of the present invention, such as Figure 2As shown, the process of establishing a search database model may include:

[0041] S41, acquire structure data, wherein the structure data includes a preset number of historical motion data, as well as the acquisition time and historical position deviation value corresponding to each historical motion data.

[0042] Specifically, historical motion data from the robot's power mechanism is acquired for a predetermined number of purposes. This historical motion data includes one or more types of data, such as acquiring only pressure data, only acceleration data, or both pressure and acceleration data simultaneously. Each acquired historical motion data point includes its corresponding acquisition time and historical position deviation value.

[0043] It should be noted that the historical motion data type must be consistent with the acquired current motion data type of the robot, or the historical motion data type must include the acquired current motion data type of the robot. This is to obtain the robot's current position deviation value based on the robot's current motion data.

[0044] In this embodiment, when acquiring historical motion data, the robot can store the historical motion data fed back by the power mechanism locally. When a preset amount is stored, the locally stored historical motion data is packaged and uploaded to a local area network (LAN) server in batches, or the historical motion data fed back by the power mechanism is uploaded to the LAN server in real time. Alternatively, it can acquire the historical motion data directly from the robot's local storage without uploading to the LAN server. This embodiment of the invention does not limit the method of acquiring historical motion data.

[0045] S42, perform deduplication on the structure data and build a search database model based on the remaining structure data; or, perform grouping on the structure data and build a search data model based on the grouped structure data.

[0046] To improve the search efficiency of the search database model, the acquired structure data can be deduplicated, and a search database model can be built based on the remaining structure data to search for the current position deviation value of the current motion data. Alternatively, the acquired structure data can be grouped, and a search data model can be built based on the grouped structure data to search for the current position deviation value of the current motion data.

[0047] In one embodiment of the present invention, deduplication of structure data may include: deduplicating structure data in which historical motion data and corresponding historical position deviation values ​​are repeated; grouping structure data may include: grouping structure data according to the magnitude of historical position deviation values.

[0048] It is feasible to remove structure data that is identical to historical motion data and its corresponding historical position deviation value, and retain structure data that is different from historical motion data and its corresponding historical position deviation value, which can improve the search speed for the current position deviation value based on the current motion data.

[0049] It is feasible to group the structured data according to the magnitude of historical position deviation values. That is, to manage approximate data values ​​hierarchically, with different deviations in different levels, in order to improve the search speed for the current position deviation value based on the current motion data.

[0050] As a specific implementation, when the current motion data is pressure data, historical motion pressure data is inserted into a dictionary through appending or other methods to establish a pressure search database model. To distinguish deviations, a correlation value search is performed before inserting historical motion pressure data, and fuzzy processing is used to ensure that approximate values ​​are equivalent and not repeatedly inserted. Alternatively, the structured data can be grouped according to the magnitude of historical position deviation values.

[0051] As another specific embodiment, when the current motion data is acceleration data, historical motion acceleration data is inserted into a dictionary through appending or other methods to establish an acceleration search database model. To distinguish deviations, a correlation value search is performed before inserting historical motion acceleration data, and fuzzy processing ensures that approximate values ​​are equivalent and not repeatedly inserted. Alternatively, the structured data can be grouped according to the magnitude of historical position deviation values.

[0052] In one embodiment of the present invention, obtaining the robot's current position deviation value based on current motion data using a pre-established search database model may include:

[0053] When a position deviation value is obtained based on the current motion data using a pre-established search database model, that position deviation value is used as the current position deviation value.

[0054] When multiple position deviation values ​​are obtained based on the current motion data using a pre-established search database model, the current position deviation value is obtained based on these multiple position deviation values ​​and the robot's current pose data.

[0055] In some embodiments, when searching for the corresponding current position deviation value in a pre-established search database model based on the current motion data, one or more deviation values ​​may be obtained.

[0056] If the robot's motion mechanism only reports one current pressure data point, or reports two consistent current pressure data points, and a position deviation value is obtained based on the current pressure data using a pre-established pressure search database model, then that position deviation value is used as the current position deviation value.

[0057] If the robot's motion mechanism provides two inconsistent current pressure data, a pre-established pressure search database model can be used to obtain two position deviation values ​​based on the two current pressure data. Alternatively, the current position deviation value can be obtained based on these two position deviation values ​​and the robot's current pose data, such as the direction the robot is facing.

[0058] As a specific embodiment, obtaining the current position deviation value based on the multiple position deviation values ​​and the robot's current pose data includes: obtaining the robot's current route based on the current pose data; determining the robot's front orientation based on the current route and the theoretical route; and determining the current position deviation value based on the relationship between position deviation values ​​and front orientation in a pre-established search database model, using the multiple position deviation values ​​and the front orientation.

[0059] When multiple positional deviation values ​​are obtained based on two inconsistent current pressure data, multiple positional deviation values ​​are generally obtained because of the difference in the vehicle's facing angle. The current positional deviation value of the robot can be determined by using the relationship between positional deviation values ​​and vehicle facing angle in a pre-established search database model. Specifically, based on the robot's current pose data, such as pressure data, acceleration data, and drive motor current data, the robot's current path is obtained. By comparing the robot's current path with the theoretical path, it is determined whether the robot's current body is on the left or right side of the theoretical path, and thus the robot's facing angle for correction is determined. Using the relationship between positional deviation values ​​and vehicle facing angle in the pre-established search database model, the current positional deviation value is obtained based on multiple positional deviation values ​​and the determined vehicle facing angle.

[0060] As another specific embodiment, when the current position deviation value is obtained based on the multiple position deviation values, the position deviation values ​​within the preset range of the Gaussian distribution center can be selected from the multiple position deviation values, and the average value of the position deviation values ​​within the preset range of the Gaussian distribution center can be calculated to obtain the current position deviation value.

[0061] It should be noted that the process of calculating the current deviation value described above does not need to be performed on the robot itself.

[0062] In one embodiment of the present invention, obtaining the robot's current position deviation value based on current motion data using a pre-established search database model may include:

[0063] When there is no historical motion data that is the same as the current motion data in the search database model, the historical motion data that is closest to the current motion data is determined from the historical motion data that is less than the current motion data and the historical motion data that is greater than the current motion data, and is denoted as the first historical motion data and the second historical motion data, based on the search database model.

[0064] The current position deviation value is obtained based on the first historical position deviation value corresponding to the first historical motion data and the second historical position deviation value corresponding to the second historical motion data.

[0065] In some embodiments, when searching for the corresponding current position deviation value based on the current motion data in a pre-established search database model, the pre-established search database model may not contain corresponding current motion data, resulting in the inability to find the corresponding position deviation value. Based on the search database model, the historical motion data closest to the current motion data can be determined from historical motion data less than the current motion data, denoted as the first historical motion data. Similarly, the historical motion data closest to the current motion data can be determined from historical motion data greater than the current motion data, denoted as the second historical motion data. The first historical position deviation value corresponding to the first historical motion data and the second historical position deviation value corresponding to the second historical motion data are obtained, and the current position deviation value is calculated using the least squares method.

[0066] In embodiments of the present invention, current motion data that does not have a corresponding value and the current position deviation value calculated based thereon can be inserted into the search database model to further expand the search database model.

[0067] In one embodiment of the present invention, such as Figure 3 As shown, the robot's pose is adjusted based on the current position deviation value, including:

[0068] S31, Establish virtual target coordinate points based on the current position deviation value;

[0069] S32, based on the virtual target coordinates and the theoretical positioning point, obtain the desired linear velocity and desired angular velocity;

[0070] S33, using the robot's motion model, the desired rotational speed of the robot motor is obtained based on the desired linear velocity and desired angular velocity;

[0071] S34 controls the motor to run at the desired speed.

[0072] In some embodiments, to reduce the deviation between the robot's current position and its theoretical position, a virtual target coordinate point is established based on the current position deviation value and the theoretical target route, so that the robot's actual running route approximates the theoretical target route. Based on the virtual target coordinate point and the theoretical positioning point (the theoretical next position of the robot) determined by the theoretical target route, the deviation between the virtual target coordinate point and the theoretical positioning point is determined. This deviation is then used to determine the desired linear velocity and desired angular velocity of the robot's drive motors. Then, using the robot's motion model, the desired rotational speed of the robot motors is obtained based on the desired linear velocity and desired angular velocity. The drive motors are then controlled to run at the desired rotational speed, allowing the robot to make corrections during movement.

[0073] It should be noted that the position deviation value in the search database model is the difference between the theoretical positioning point coordinates and the current position coordinates. By continuously establishing virtual target coordinate points, and based on the theoretical positioning point coordinates determined by the theoretical target route, as well as the target linear velocity, angular velocity, deviation, and vehicle heading of the drive mechanism, the real-time linear velocity and expected angular velocity of the drive mechanism are obtained. Then, the motion model is used to calculate the expected speed of each motor movement, forming a correction trend, so that the correction can be completed within a certain period of time.

[0074] In one embodiment of the present invention, when the robot has two walking wheels, the desired rotational speed of the robot motor is obtained by using the robot's motion model based on the desired linear velocity and desired angular velocity, which can be obtained according to the following equation.

[0075] F 上 =0.5*G*(1-K), F 下 =0.5*G*(1+K).

[0076] Among them, F 上 and F 下 These represent the forces acting on the robot's two wheels, where G represents gravity and K is a calculation coefficient that can be obtained from the static pressure value.

[0077] The robot's total energy equation is: E1 = W1 + E2 + E3. Here, E1 represents electrical energy loss, W1 represents kinetic energy, E2 represents internal losses (such as heat generation from electronic components), and E3 represents external losses (mainly energy loss due to friction). This total energy equation is the theoretical energy conservation equation. It should be noted that external forces include gravity and friction; here, friction is considered the driving force.

[0078] It should be noted that E1 is the simultaneous equation for real-time measurement of battery voltage, current, and capacity consumption, W1 is the simultaneous equation for real-time measurement of speed through an accelerometer, E2 is the simultaneous equation for real-time measurement of current, and E3 is the frictional force simultaneous equation. When the movement is bumpy, the frictional force changes, generating two component forces along the movement direction and perpendicular to the movement direction, and this component force is the reason for the movement deviation.

[0079] The total motion equation of the robot: ∑Fx*Δt = ∑M*Vx; ∑Fy*Δt = ∑M*Vy; ∑Fz*Δt = ∑M*Vz.

[0080] The total mechanical equation of the robot: ∑Fx = ∑M*Ax; ∑Fy = ∑M*Ay; ∑Fz = ∑M*Az.

[0081] Among them, M is the mass of each part, Vx is the velocity component in the x direction, Vy is the velocity component in the y direction, Vz is the velocity component in the z direction, Ax is the acceleration component in the x direction, Ay is the acceleration component in the y direction, and Az is the acceleration component in the z direction.

[0082] Specifically, for different motion processes, such as straight-line and turning, different motion equations, mechanical equations, and energy equations can be established, so that the robot can perceive how much offset force it is subjected to and through what proportion of speed adjustment can eliminate the offset of the robot body relative to the theoretical route, and then adjust the attitude to continue driving.

[0083] In an embodiment of the present invention, when the search database model has not been established and the robot is running straight, according to the straight-line process in the self-learning motion equation based on speed and force feedback, the conversion coefficient is calculated to obtain the current position deviation value, and the next-round motion correction is updated according to the previous-round motion deviation to achieve optimization. See Figure 4 , the actual deviation correction control process is exemplified as follows: when a lateral deviation occurs, it is judged that when Δy (the position deviation value in the y direction) > y_thred (the preset threshold in the y direction), deviation correction is performed, that is, the curve motion control is realized by controlling the motion components (motors) of the robot to accelerate and decelerate, so that Δy < y_thred, and then continue to run.

[0084] In an embodiment of the present invention, when the search database model has not been established and the robot spins in place (a two-wheel differential-like scenario), according to the spin process in the self-learning motion equation based on speed and force feedback, the conversion coefficient is calculated to obtain the current position deviation value, and the next-round motion correction is updated according to the previous-round motion deviation to achieve optimization. See Figure 5, the actual deviation correction control process is exemplified as follows: When the rotation center deviates, if it is determined that Δx (the position deviation value in the x direction) > x_thred (the preset threshold in the x direction) or Δy (the position deviation value in the y direction) > y_thred (the preset threshold in the y direction), motion deviation correction is performed. That is, curve motion control is achieved by controlling the motion component (motor) to accelerate and decelerate (the ordinary method does not perform motion control during the spin and curve processes. Here, the method provided by the embodiment of the present invention is used to perform deviation correction during the spin and curve processes, with higher real-time deviation correction and reduced possibility of "derailing" directly after the spin or curve motion), so that Δx < x_thred and Δy < y_thred, and then continue to run. Here, "derailing" refers to the maximum threshold of the projection distance of the current robot position deviating from the theoretical route.

[0085] In an embodiment of the present invention, when the search database model has not been established and the robot runs in an arc curve, according to the curve motion process in the self-learning motion equation based on speed and force feedback, combined with the theoretical rotation radius conversion coefficient, the motion deviation is obtained, and the next-round motion deviation correction is updated according to the motion deviation of the previous round to achieve optimization. The actual deviation correction control process is similar to the scenario where the robot spins in place.

[0086] It should be noted that when some sensors are to be removed after a period of self-learning, the self-learning function (autonomous identification of data sources) will be turned off, and motion deviation correction optimization will be performed relying on the completed action database.

[0087] It should be noted that according to the position deviation value obtained from the unestablished search database model, multiplying the position deviation value by the probability coefficient can obtain the real motion deviation. Here, the probability coefficient is the conversion coefficient, that is, the probability distribution coefficient obtained in the above multi-solution situation. For example, when there are multiple deviation solutions under an approximate external load (associated with pressure and acceleration feedback), these multiple deviation solutions will have a certain distribution law, such as a Gaussian distribution, that is, there will be a deviation center point. Taking the ratio of the point closer to the center point (the distance from the center point is a preset threshold) to the total number of all solutions as the probability distribution coefficient. If this probability distribution coefficient is high, it is considered that this central deviation value is credible, and this central deviation value is used as the final real deviation.

[0088] It should be noted that the above process of solving the position deviation value may not be performed on the robot body.

[0089] It should be noted that after the search database model is established, the search database model can be directly used for searching to determine the current position deviation value.

[0090] It should be noted that after optimizing the model through self-learning, from a practical standpoint, some of the sensors initially deployed (such as RTK sensors, cameras, and other sensors used for positioning and navigation) can be removed later, reducing equipment maintenance costs. Before removing any sensors, a judgment must be made. After removing the sensors, without positioning data, the corrective action (specifying the wheel speed of the differential wheel) is directly matched to the appropriate action by searching the database model.

[0091] The robot pose control method provided in this embodiment of the invention obtains the "motion-deviation" relationship based on the search database model. That is, it obtains multiple deviation solutions under approximate external force (considering the robot's front orientation, etc.), performs data analysis, leaves the solutions near the center that conform to the Gaussian distribution, and calculates the average value to obtain the final deviation value.

[0092] The robot pose control method provided in this invention performs curve fitting on the deviation based on raw data such as force, acceleration, and current. Then, it performs curve fitting on the correction action based on the deviation (calculating the curve equation using a polynomial after statistical plotting points) to find the relationship between what kind of "motion" the robot should perform and what kind of "correction action" it should perform. Furthermore, by observing changes in motion (force, current, etc.) and the deviation patterns in the established search database model, it anticipates the necessary correction actions and performs corrections when the deviation is small, thus optimizing the process.

[0093] The pose control method of this invention can improve motion correction in sloping or uneven environments, increase the straight-line correction response speed, and optimize the correction actions during turning and spinning, making the motion closer to the theoretical trajectory.

[0094] The robot pose control method provided in this embodiment of the invention is based on the principle that "force is the cause of changes in the motion state of an object". By learning in advance what kind of force (pressure sensor) the robot is subjected to during operation, what kind of motion trend (predicted result) will be generated, so as to make a subtle correction action after making a prediction in advance. In this way, the robot's running trajectory is better aligned with the theoretical route, and the running efficiency is also improved at the same time.

[0095] This invention provides a robot pose control device.

[0096] Figure 6 This is a schematic diagram of a robot pose control device according to an embodiment of the present invention. Figure 6 As shown, the robot's pose control device 100 may include: an acquisition module 10, a search module 20, and a control module 30.

[0097] The acquisition module 10 is used to acquire the robot's current motion data; the search module 20 is used to obtain the robot's current position deviation value based on the current motion data using a pre-established search database model, wherein the search database model includes the correspondence between the robot's historical motion data and historical position deviation values; and the control module 30 is used to adjust the robot's pose based on the current position deviation value.

[0098] It should be noted that other specific embodiments of the robot pose control device provided in the embodiments of the present invention can be found in other specific embodiments of the robot pose control method in the above embodiments of the present invention.

[0099] The robot posture control device of this invention uses historical motion data fed back from the robot's power mechanism to adjust the robot's posture. It directly analyzes and learns the robot's movement mode from a perception perspective. Through self-learning, it can predict in advance what kind of force the robot will be subjected to during operation and what kind of movement trend will be generated. After making advance predictions, it can perform subtle correction actions, so as to achieve the purpose of intelligent, flexible and rapid control of the robot's own movement posture.

[0100] This invention provides a computer-readable storage medium.

[0101] In this embodiment, a computer program is stored on a computer-readable storage medium. When the computer program is executed by a processor, it implements the robot pose control method described above.

[0102] This invention provides an electronic device.

[0103] In this embodiment, the electronic device may include a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it implements the robot pose control method described above.

[0104] The computer-readable storage medium and electronic device of this invention utilize a robot pose control method to enable the robot to achieve intelligent, flexible, and rapid control of its own motion pose.

[0105] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0106] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0107] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0108] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to 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 this invention.

[0109] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0110] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0111] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0112] 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 pose control method of a robot, characterized by, The method comprises: acquiring current motion data of the robot; obtaining a current position deviation value of the robot according to the current motion data by using a pre-established search database model, wherein the search database model comprises a corresponding relationship between historical motion data and historical position deviation values of the robot; adjusting a pose of the robot according to the current position deviation value; the step of obtaining the current position deviation value of the robot according to the current motion data by using the pre-established search database model comprises: when one position deviation value is obtained according to the current motion data by using the pre-established search database model, taking the position deviation value as the current position deviation value; when multiple position deviation values are obtained according to the current motion data by using the pre-established search database model, obtaining the current position deviation value according to the multiple position deviation values and current pose data of the robot; the step of obtaining the current position deviation value according to the multiple position deviation values and the current pose data of the robot comprises: obtaining a current route of the robot according to the current pose data; determining a vehicle head orientation of the robot according to the current route and a theoretical route; determining the current position deviation value according to the multiple position deviation values and the vehicle head orientation by using a relationship between position deviation values and vehicle head orientations in the pre-established search database model.

2. The pose control method of the robot according to claim 1, characterized in that, the process of establishing the search database model comprises: acquiring structure data, wherein the structure data comprises a preset number of historical motion data, and acquisition time and historical position deviation values corresponding to each historical motion data; performing deduplication processing on the structure data, and establishing the search database model based on the remaining structure data; or performing grouping processing on the structure data, and establishing the search data model based on the grouped structure data.

3. The pose control method of the robot according to claim 2, characterized in that, the step of performing deduplication processing on the structure data comprises: performing deduplication processing on structure data in which historical motion data and corresponding historical position deviation values are repeated; the step of performing grouping processing on the structure data comprises: performing grouping processing on the structure data according to the value size of the historical position deviation values.

4. The pose control method according to claim 1, wherein the step of obtaining the current position deviation value of the robot according to the current motion data by using the pre-established search database model comprises: when there is no historical motion data identical to the current motion data in the search database model, determining historical motion data closest to the current motion data from historical motion data smaller than the current motion data and historical motion data larger than the current motion data based on the search database model, and recording the historical motion data as first historical motion data and second historical motion data; obtaining the current position deviation value according to a first historical position deviation value corresponding to the first historical motion data and a second historical position deviation value corresponding to the second historical motion data.

5. The pose control method of the robot according to claim 1, characterized in that, the step of adjusting the pose of the robot according to the current position deviation value comprises: establishing a virtual target coordinate point according to the current position deviation value; According to the virtual target coordinate point and the theoretical positioning point, a desired linear velocity and a desired angular velocity are obtained; According to the desired linear velocity and the desired angular velocity, a desired rotating speed of a motor of the robot is obtained by using a motion model of the robot; The motor is controlled to operate according to the desired rotating speed.

6. The pose control method of claim 1, wherein, The motion data includes at least one of pressure data and acceleration data.

7. A pose control device of a robot, characterized by, A device for implementing the pose control method of the robot as claimed in any one of claims 1-6, the device comprising: an acquisition module configured to acquire current motion data of the robot; a search module configured to obtain a current position deviation value of the robot according to the current motion data by using a pre-established search database model, wherein the search database model comprises a corresponding relationship between historical motion data and historical position deviation values of the robot; a control module configured to adjust a pose of the robot according to the current position deviation value.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that The computer program, when executed by a processor, implements the pose control method of the robot as claimed in any one of claims 1-6.

9. An electronic device comprising a memory, a processor, the memory having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the pose control method of the robot as claimed in any one of claims 1-6.

Images