Force field compensation method and electronic equipment

Abstract

The invention provides a force field compensation method and electronic equipment, and relates to the technical field of robot control. The method comprises the following steps: acquiring mechanical information of a robot in a grid space of a preset dimension; according to the current position information of the robot, multiple adjacent grid points corresponding to the current position of the robot in the grid space are determined, and according to mechanical information on the adjacent grid points, environment force field information corresponding to the current position of the robot is determined; and determining a current compensation external force according to the environment force field information and the current sensor information, and performing force field compensation based on the current compensation external force. According to the application, the Cartesian space is gridded and the sampling force and moment information is traversed to establish the sampling force field model of the grid space, the sampling force field model is suitable for force field environments with various complexity degrees, and the force and moment information is efficiently acquired at any position with low calculation complexity by using the sampling force field model. Therefore, online real-time compensation of an external force field is realized.

Description

Technical Field

[0001] The present application relates to the technical field of robot control, and in particular, to a force field compensation method and an electronic device. Background Art

[0002] During the interaction between humans and robots, when an operator applies an external force to the robot or controls the robot to perform some force control operations, since there are multiple environmental force fields such as gravity, magnetism, and cable pulling in the environment where the robot is located, these environmental force fields will also exert forces on the end of the robot, thereby affecting the readings of the force sensors at the end of the robot, making it impossible for the robot to accurately distinguish the actual external force of the operator from the environmental force field, ultimately resulting in a decrease in force control accuracy. In addition, it is also prone to safety hazards. Therefore, ensuring precise force control is crucial for improving the operation accuracy and safety of robots.

[0003] In the prior art, gravity compensation is a commonly used force field compensation method. This method cancels the influence of gravity by pre-inputting the end load parameters. For example, when the robot is in different postures, by calculating the current gravity and subtracting the current gravity from the actual measured force of the sensor, the effect of being unaffected by the gravity field is achieved. However, the gravity compensation method is only applicable to gravity fields or other force fields with clear modeling, and is not applicable to application scenarios of complex non-linear force fields. Therefore, how to effectively identify and compensate for complex and variable environmental force fields to improve the operation accuracy and safety of robots in human-robot collaboration is a technical problem to be solved urgently. Summary of the Invention

[0004] The purpose of the present application is to provide a force field compensation method and an electronic device for solving the problem that the existing compensation method is not applicable to application scenarios of complex non-linear force fields in view of the above deficiencies in the prior art.

[0005] To achieve the above purpose, the technical solutions adopted in the embodiments of the present application are as follows:

[0006] In a first aspect, an embodiment of the present application provides a force field compensation method, and the method includes:

[0007] Obtain the mechanical information of the robot in the grid space of the preset dimension. A mechanical sensor is installed at the end of the robot, and the mechanical information includes the force and torque of the robot at each grid point in the grid space, where each grid point represents a specific position in the grid space;

[0008] Determine a plurality of adjacent grid points corresponding to the current position of the robot in the grid space according to the current position information of the robot, and determine the environmental force field information corresponding to the current position of the robot according to the mechanical information at each of the adjacent grid points;

[0009] Read the current sensor information of the mechanical sensor;

[0010] Determine the current compensating external force according to the environmental force field information corresponding to the current position of the robot and the current sensor information, and perform force field compensation based on the current compensating external force.

[0011] As a possible implementation manner, before obtaining the mechanical information of the robot in the grid space of the preset dimension, it further includes:

[0012] Determine the grid division parameters of the Cartesian space according to the spatial range information of the sampling force field, where the Cartesian space is a six-dimensional coordinate system space based on which the robot interacts with the outside world, and the grid division parameters include the boundary ranges, the number of grid points, and the grid point distribution order of each dimension;

[0013] Perform grid processing on the Cartesian space according to the grid division parameters to obtain the grid space, and the grid space is used to indicate the distribution of the sampling force field within the Cartesian space range.

[0014] As a possible implementation manner, the determining the multiple adjacent grid points corresponding to the current position of the robot in the grid space according to the current position information of the robot includes:

[0015] Determine the adjacent grid points in each dimension according to the fixed-point information of each dimension in the grid space in the current position information, where the current position information is six-dimensional coordinate information including translational motion information and rotational motion information;

[0016] Perform permutation and combination on the adjacent grid points in each dimension to determine the multiple adjacent grid points.

[0017] As a possible implementation manner, the determining the adjacent grid points in each dimension according to the fixed-point information of each dimension in the grid space in the current position information includes:

[0018] Determine the resolution of each dimension according to the number of grid points of each dimension;

[0019] Determine the grid point interval where the fixed-point information of each dimension is located according to the resolution of each dimension, the fixed-point information of each dimension, and the boundary range of each dimension;

[0020] Determine the adjacent grid points in each dimension according to the grid point interval where the fixed-point information of each dimension is located.

[0021] As a possible implementation manner, the determining the grid point interval where the fixed-point information of each dimension is located according to the resolution of each dimension, the fixed-point information of each dimension, and the boundary range of each dimension includes:

[0022] For each of the dimensions, determine a first difference between the resolution of the dimension and the fixed-point information of the dimension, and based on the first difference and the boundary range of the dimension, determine a grid interval parameter;

[0023] Round the grid interval parameter to determine the grid interval where the fixed-point information of the dimension is located.

[0024] As a possible implementation, adjacent grid points on each dimension include a first adjacent grid point and a second adjacent grid point. The arranging and combining the adjacent grid points on each dimension to determine the plurality of neighboring grid points includes:

[0025] Combine the first adjacent grid point on the target dimension with the first adjacent grid points and the second adjacent grid points on the remaining dimensions other than the target dimension respectively to obtain a first set of grid points, and combine the second adjacent grid point on the target dimension with the first adjacent grid points and the second adjacent grid points on the remaining dimensions other than the target dimension respectively to obtain a second set of grid points;

[0026] Arrange the grid points in the first set of grid points and the second set of grid points to obtain the plurality of neighboring grid points.

[0027] As a possible implementation, the determining the environmental force field information corresponding to the current position of the robot according to the mechanical information on each of the neighboring grid points includes:

[0028] Based on a preset interpolation strategy, perform dimensionality reduction interpolation processing on the mechanical information on each of the neighboring grid points to determine the environmental force field information corresponding to the current position of the robot.

[0029] As a possible implementation, the based on a preset interpolation strategy, perform dimensionality reduction interpolation processing on the mechanical information on each of the neighboring grid points to determine the environmental force field information corresponding to the current position of the robot includes:

[0030] Divide the plurality of neighboring grid points according to the coordinate values corresponding to the first dimension of each of the neighboring grid points to obtain a first set of grid points and a second set of grid points;

[0031] Use the preset interpolation strategy to perform interpolation processing on the mechanical information of the first dimension of each grid point in the first set of grid points and the second set of grid points to obtain a first dimension interpolation result. The first dimension interpolation result includes a plurality of first neighboring grid points interpolated in the first dimension, and the number of the plurality of first neighboring grid points is one half of the number of the plurality of neighboring grid points;

[0032] Divide the multiple first neighboring lattice points according to the coordinate values corresponding to the second dimension of each of the first neighboring lattice points, to obtain a third lattice point group and a fourth lattice point group, and perform interpolation processing on the mechanical information of the second dimension of each lattice point in the third lattice point group and the fourth lattice point group by using the preset interpolation strategy, and iteratively execute until there is only one target lattice point in the lattice point group, and use the mechanical information corresponding to the target lattice point as the environmental force field information corresponding to the current position of the robot.

[0033] As a possible implementation manner, the determining the current compensation external force according to the environmental force field information corresponding to the current position of the robot and the current sensor information includes:

[0034] Determine a second difference between the environmental force field information and the current sensor information, and use the second difference as the current compensation external force.

[0035] In a second aspect, an embodiment of the present application provides a force field compensation device, and the device includes:

[0036] An acquisition module, configured to acquire mechanical information of a robot in a grid space of a preset dimension. A mechanical sensor is installed at the end of the robot, and the mechanical information includes forces and torques of the robot on each lattice point in the grid space, where each lattice point represents a specific position in the grid space;

[0037] A first determination module, configured to determine, according to the current position information of the robot, a plurality of neighboring lattice points corresponding to the current position of the robot in the grid space, and determine the environmental force field information corresponding to the current position of the robot according to the mechanical information on each of the neighboring lattice points;

[0038] A reading module, configured to read the current sensor information of the mechanical sensor;

[0039] A second determination module, configured to determine a current compensation external force according to the environmental force field information corresponding to the current position of the robot and the current sensor information, and perform force field compensation based on the current compensation external force.

[0040] In a third aspect, an embodiment of the present application provides an electronic device, including: a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of the force field compensation method according to any one of the first aspects described above.

[0041] Fourthly, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it executes the steps of the force field compensation method according to any one of the above first aspects.

[0042] According to the force field compensation method and electronic device of the embodiments of the present application, mechanical information of a robot in a grid space of a preset dimension is obtained. A mechanical sensor is installed at the end of the robot. The mechanical information includes the force and torque of the robot at each grid point in the grid space, where each grid point represents a specific position in the grid space; according to the current position information of the robot, a plurality of adjacent grid points corresponding to the current position of the robot in the grid space are determined, and according to the mechanical information at each adjacent grid point, the environmental force field information corresponding to the current position of the robot is determined; the current sensor information of the mechanical sensor is read; according to the environmental force field information and the current sensor information corresponding to the current position of the robot, the current compensation external force is determined, and force field compensation is performed based on the current compensation external force. According to the embodiments of the present application, the grid space of the preset dimension is obtained by performing grid processing on the Cartesian space. The construction of the grid space is directly based on sampling data rather than relying on a specific mathematical model. That is to say, the grid space of the present application is essentially a sampling force field model. Therefore, the grid space of the present application can be applied to force field environments of any complexity. On this basis, through the domain search algorithm and the interpolation algorithm, force and torque information can be quickly and accurately obtained at any position in the grid space. Furthermore, through the interpolation analysis of the sensor information and the environmental force field information, effective compensation for environmental interference is realized, and the accuracy and stability of robot operation are improved. Based on this, the present application establishes a detailed force field model of the grid space by performing grid processing on the Cartesian space and traversing and sampling force and torque information, and uses the grid space to efficiently obtain force and torque information at any position with a low computational complexity, so as to realize online real-time compensation for the external force field. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0044] Figure 1 It shows a schematic flowchart of a force field compensation method provided by an embodiment of the present application;

[0045] Figure 2 It shows a schematic flowchart of a grid processing method provided by an embodiment of the present application;

[0046] Figure 3 Shows a schematic flowchart of a method for determining adjacent grid points provided by an embodiment of the present application;

[0047] Figure 4 Shows a schematic flowchart of a method for determining adjacent grid points in each dimension provided by an embodiment of the present application;

[0048] Figure 5 Shows a schematic flowchart of a dimensionality reduction interpolation processing method provided by an embodiment of the present application;

[0049] Figure 6 Shows a schematic structural diagram of a force field compensation device provided by an embodiment of the present application;

[0050] Figure 7 Shows a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It should be understood that the accompanying drawings in the present application are only for the purposes of illustration and description, and are not used to limit the protection scope of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in the present application show operations implemented according to some embodiments of the present application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical context may be reversed or implemented simultaneously. In addition, those skilled in the art may add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of the present application.

[0052] In addition, the described embodiments are only some embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application claimed, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the protection scope of the present application.

[0053] It should be noted that the term "including" will be used in the embodiments of the present application to indicate the existence of the features stated hereinafter, but does not exclude the addition of other features.

[0054] In view of the problems existing in the prior art, the present application provides a force field compensation method. By meshing the Cartesian space and traversing and sampling force and torque information, an environment force field model based on sampling is constructed, that is, a grid space. This grid space can reflect the force field distribution in the entire sampling space. The advantage of its establishment is that it is directly established based on sampling data and does not depend on a specific mathematical expression to describe the force field. Therefore, the grid space can be applied to force field environments of any complexity, even those that are difficult to accurately describe with traditional mathematical formulas.

[0055] Furthermore, in order to efficiently obtain force and torque information at any position and thus achieve online real-time compensation for the external force field, the present application adopts a method combining a domain search algorithm and an interpolation algorithm. First, according to the current position of the robot, the nearest neighbor grid point group is found in the pre-constructed grid space, and then a selected one-dimensional or multi-dimensional interpolation strategy is used to estimate the force field characteristics at the current position of the robot based on the known force and torque information on these nearest neighbor grid points. Furthermore, by comparing the difference between the total force and torque measured by the sensor and the environmental force field information interpolated from the force field model, the external force that needs to be compensated can be determined, thus effectively eliminating the influence of the environmental force field on the operation and enabling the robot to respond more accurately to the user's intention.

[0056] In summary, the present application uses grid sampling combined with an efficient interpolation algorithm to provide accurate force and torque information for force field environments of any complexity without relying on a specific mathematical model, thereby supporting the high-precision control and real-time response capabilities of the robot system.

[0057] Figure 1 The flowchart of a force field compensation method provided by an embodiment of the present application is shown. Refer to Figure 1 As shown, the method specifically includes the following steps:

[0058] S101. Obtain the mechanical information of the robot in the grid space of the preset dimension.

[0059] Optionally, the mechanical information includes the force and torque of the robot at each grid point in the grid space, where each grid point represents a specific position in the grid space. The grid space of the preset dimension is obtained by meshing the six-dimensional Cartesian space, that is, the grid space is the six-dimensional Cartesian space after meshing. Among them, the first three dimensions are used to describe the translational motion of the robot, and the last three dimensions use Euler angles to describe the rotational motion of the robot. It should be noted that since there are multiple ways to define rotation, that is, multiple different Euler angle sequences, for the sake of convenience of description, the present application takes the ZYZ Euler angle as an example for subsequent description.

[0060] Optionally, in the six-dimensional grid space, the translation dimensions such as (x, y, z) represent the specific position of the robot in the space. Among them, x represents the displacement along the X-axis, y represents the displacement along the Y-axis, and z represents the displacement along the Z-axis. The rotation dimensions are represented by three angle values to indicate the rotation of the robot around three main axes. Different combinations of rotations are called different convention forms. The above-mentioned ZYZ Euler angles are a common convention form. Among them, the first rotation represents the rotation around the Z-axis, usually called the yaw angle, the second rotation represents the rotation around the new Y-axis, usually called the pitch angle, and the third rotation represents the rotation around the updated Z-axis, usually called the roll angle. In this way, the orientation of the robot in the space can be uniquely determined.

[0061] Optionally, the robot in this application is a robot capable of measuring the six-dimensional force data at the end of the robot, or a mechanical sensor is installed at the end of the robot. This mechanical sensor is, for example, a six-dimensional force sensor. By sequentially traversing all the grid points in the grid space in the sampling force field that needs to be sampled, the force F=(F x ,F y ,F z ) and the moment M=(M x ,M y ,M z ) at each grid point can be measured. Further, after obtaining the force F=(F x ,F y ,F z ) and the moment M=(M x ,M y ,M z ) of the robot at each grid point in the grid space, the mechanical information of the force and moment of all grid points can be stored in a data set. Each piece of data in this data set contains at least the grid point coordinates and the measured force-moment two six-dimensional vectors. In addition, to ensure the consistency of any data access in terms of computational time complexity, the above data set can be stored in a data manner.

[0062] S102. Determine multiple neighboring grid points corresponding to the current position of the robot in the grid space according to the current position information of the robot, and determine the environmental force field information corresponding to the current position of the robot according to the mechanical information on each neighboring grid point.

[0063] Optionally, the current position information of the robot is also the real-time position, and this real-time position is six-dimensional coordinate information. For example, the real-time position Among them, the first three coordinates x r ,y r ,z r represent the translation position, and the last three coordinates θ r , ψ rIndicates the rotational attitude. For each dimension of the grid space, based on the current position information of the robot, the neighboring grid points closest to the current value need to be determined. It should be noted that for each dimension, two neighboring grid points closest to the current value can be determined. For example, in the x-dimension, the two neighboring grid points closest to the current value x r are x i and x i+1 , in the y-dimension, the two neighboring grid points closest to the current value y r are y i and y i+1 , in the z-dimension, the two neighboring grid points closest to the current value z r are z i and z i+1 , in the θ-dimension, the two neighboring grid points closest to the current value θ r are θ i and θ i+1 , in the dimension, the two neighboring grid points closest to the current value are and In the ψ-dimension, the two neighboring grid points closest to the current value ψ r are ψ i and ψ i+1 .

[0064] Optionally, based on the two neighboring grid points obtained in each of the above dimensions, all possible multiple neighboring grid points can be obtained by permuting and combining the two neighboring grid points in each dimension. Since the real-time position X of the robot is very likely not exactly on any grid point in the grid space but between grid points, interpolation calculations can be performed based on the mechanical information of the multiple neighboring grid points corresponding to the current position of the robot. And for the six-dimensional state, a recursive method can be used to perform interpolation calculations for each dimension in turn until the interpolation calculations for all six dimensions are completed, obtaining a comprehensive mechanical information F r , and this mechanical information F r is used as the environmental force field information corresponding to the current position of the robot.

[0065] S103. Read the current sensor information of the force sensor.

[0066] Optionally, the current sensor information can be directly read from the force / torque sensor installed on the end effector of the robot. And in this application, a six-dimensional force sensor is used as the force sensor, which can measure forces in three directions and torques in three directions simultaneously. In addition, before reading the current sensor, it is necessary to ensure that the six-dimensional force sensor is correctly connected to the control system of the robot, so as to ensure that the six-dimensional force sensor can work properly and start transmitting data.

[0067] S104. Determine the current compensation external force according to the environmental force field information corresponding to the current position of the robot and the current sensor information, and perform force field compensation based on the current compensation external force.

[0068] Optionally, the environmental force field information refers to various physical or virtual force fields existing in the environment where the robot is currently located. These force fields can attract or repel the robot. Therefore, after determining the environmental force field information corresponding to the current position of the robot and the current sensor information, the current compensation external force can be calculated based on the environmental force field information and the current sensor information. This current compensation external force can be used for force field compensation to offset or adjust the natural influence of the environmental force field on the robot, so as to ensure that the robot can move according to a predetermined path or strategy. For example, if the environmental force field causes the robot to deviate from the predetermined path, then force field compensation based on the compensation external force can guide the robot back to the correct path.

[0069] Based on this, according to the force field compensation method provided by the embodiments of the present application, the grid space of the preset dimension is obtained by meshing the Cartesian space. The construction of the grid space is directly based on sampling data rather than relying on a specific mathematical model. That is to say, the grid space of the present application is essentially a sampled force field model. Therefore, the grid space of the present application can be applied to force field environments of any complexity. On this basis, through the domain search algorithm and the interpolation algorithm, force and torque information can be quickly and accurately obtained at any position in the grid space. Furthermore, through the interpolation analysis of the sensor information and the environmental force field information, effective compensation for environmental interference is achieved, and the accuracy and stability of robot operation are improved. Based on this, the present application establishes a detailed force field model by meshing the Cartesian space and traversing and sampling force and torque information to establish a grid space, and uses this grid space to efficiently obtain force and torque information at any position with a low computational complexity, so as to realize online real-time compensation for the external force field.

[0070] Figure 2 The flowchart of a meshing method provided by the embodiments of the present application is shown. As a possible implementation manner, referring to Figure 2 as shown, before obtaining the mechanical information of the robot in the grid space of the preset dimension, the method further includes:

[0071] S201. Determine the grid division parameters of the Cartesian space according to the spatial range information of the sampled force field.

[0072] Optionally, a sampling force field is usually constructed by collecting data within a specific spatial range, which describes the influence of forces or potential fields at different positions within this space. Therefore, the spatial range information of the sampling force field defines the spatial boundaries where the force field exists and is crucial for determining the grid division parameters in Cartesian space. Among them, Cartesian space is a six-dimensional coordinate system space based on the interaction between the robot and the external environment. Within this six-dimensional coordinate system space, the robot can accurately describe its position and orientation, thereby achieving precise interaction with the external environment.

[0073] Optionally, the grid division parameters include the boundary ranges of each dimension, the number of grid points, and the grid point distribution order. Among them, the boundary range is also the maximum and minimum values of each dimension. Taking the x dimension as an example, the boundary range includes the maximum value x max and the minimum value x min . The number of grid points in each dimension indicates the grid resolution of each dimension, that is, the number of segments divided in each dimension. For example, if the number of grid points on the x-axis is set to 11, the x-axis will be divided into 10 equally spaced small segments, and the spacing of each segment is Step x . The grid point distribution order is the distribution order of the grid points between the maximum and minimum values of each dimension, that is, the set direction, starting from the minimum value to the maximum value, or starting from the maximum value to the minimum value.

[0074] S202. Perform grid processing on the Cartesian space according to the grid division parameters to obtain a grid space.

[0075] Optionally, grid processing means that for each dimension, discrete grid point coordinates are generated according to the set grid resolution, set direction, and boundary range. For example, continuing with the example where the number of grid points on the x-axis is set to 11, 11 equally spaced grid points x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 can be generated on the x-axis. Among them, x1 is the minimum value of the defined boundary range, and x 11 is the maximum value of the defined boundary range. Finally, a six-dimensional grid can be formed, and each grid point is represented by (x i , y j , z k , θ l , φ m , ψ n ), where i, j, k, l, m, n respectively refer to the indices of each dimension. In this way, the grid space can be obtained, and this grid space can indicate the distribution of the sampling force field within the Cartesian space range.

[0076] Optionally, the grid space of the present application is obtained by gridding the Cartesian space. Specifically, by dividing the Cartesian space into many small and regular grids, each position in the sampled force field space is assigned to a specific grid. Then, each grid is traversed, the grid points in each grid are sampled for force and torque, and a model describing the internal force distribution of the entire space, that is, the grid space, is established based on the sampled data. It should be particularly noted that this grid space is a force field model established based on sampling and does not rely on a pre-set mathematical model to describe the force field, which means that this grid space can be applied to force field environments of any complexity level.

[0077] Based on this, the present application establishes the grid space, a force field model, by gridding the Cartesian space and sampling force and torque information. Since this force field model of the grid space is established by sampling rather than relying on a specific mathematical model, it can be applied to force field environments of various complexity levels, significantly improving the applicability of the force field compensation application scenarios.

[0078] Figure 3 The flowchart of a method for determining adjacent grid points provided by an embodiment of the present application is shown. As a possible implementation, referring to Figure 3 As shown, in step S102, according to the current position information of the robot, multiple adjacent grid points corresponding to the current position of the robot in the grid space are determined, which specifically includes the following steps:

[0079] S301. Determine the adjacent grid points in each dimension according to the fixed-point information of each dimension in the grid space in the current position information.

[0080] Optionally, the current position information is six-dimensional coordinate information including translational motion information and rotational motion information. The current position information is also the real-time position coordinate of the robot, and this real-time position coordinate is, for example, On this basis, based on the set grid resolution, that is, the number of grid points and the spacing Step in each dimension x , combined with the fixed-point information of each dimension in the grid space, the grid resolution, and the correlation relationship between the spacing Step x the closest grid points of the current position of the robot in each dimension are determined. Among them, the fixed-point information refers to the real-time position coordinate of the robot, such as the specific value of one dimension in, for example, x r .

[0081] Figure 4 The flowchart of a method for determining adjacent grid points in each dimension provided by an embodiment of the present application is shown. Referring to Figure 4As shown, the above step S301 determines the adjacent grid points in each dimension according to the fixed-point information of each dimension in the current position information, specifically including the following steps:

[0082] S401. Determine the resolution of each dimension according to the number of grid points in each dimension.

[0083] Optionally, the number of grid points refers to the number of sampling points set in a certain dimension. Continuing with the example where the number of grid points on the set x-axis is 11, it means that there are 11 discrete measurement points or positions including the starting point and the ending point. And the number of grid points determines how many small segments the dimension is divided into. For example, if the number of grid points on the x-axis is set to 11, the x-axis will be divided into 10 equally spaced small segments, that is, a small segment is defined between every two adjacent grid points, and the length of each small segment is called the pitch, representing the distance between two adjacent grid points. On this basis, if the total length of the boundary range of the entire x-axis is known, the pitch of each segment can also be calculated by dividing the total length by the number of small segments, which is Step x .

[0084] Optionally, the resolution refers to the minimum level of detail that can be distinguished. The resolution is usually directly related to the number of grid points and shows a positive correlation, that is, a higher number of grid points means a higher resolution. To a certain extent, the number of grid points in each dimension indicates the resolution of each dimension, that is, the number of divisions in each dimension. For example, if there are 11 grid points on the x-axis, the resolution on the x-axis is correspondingly 11.

[0085] S402. Determine the grid point interval where the fixed-point information of each dimension is located according to the resolution of each dimension, the fixed-point information of each dimension, and the boundary range of each dimension.

[0086] Optionally, the resolution refers to the number of divisions of the grid in each dimension, the fixed-point information refers to the specific value of a dimension in the real-time position coordinates of the robot, and the grid point interval indicates between which two adjacent grid points the fixed-point information of this dimension is located.

[0087] Optionally, for each dimension, determine the first difference between the resolution of the dimension and the fixed-point information of the dimension, and based on the first difference and the boundary range of the dimension, determine the grid point interval parameter; perform rounding processing on the grid point interval parameter to determine the grid point interval where the fixed-point information of the dimension is located.

[0088] Exemplarily, taking the x-dimension as an example, the boundary range includes the maximum value x max and the minimum value x min , the number of grid points is n x , then the pitch For the given dimension fixed-point information, such as x r , calculate the fixed-point information x rThe first difference between the starting point of the x-dimension (such as x min ), that is, the first difference diff = x r - x min . Among them, if the number of grid points on the x-axis is 11, then 11 equally spaced grid points x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 are correspondingly generated on the x-axis. Then the starting point x min of the above x-dimension is x1.

[0089] Exemplarily, dividing the above first difference diff by the spacing Step x , a preliminary grid point interval parameter can be obtained, that is, i = diff / Step x , where i represents the starting grid point index of the grid point interval where the fixed-point information x r is located.

[0090] Furthermore, through the rounding operation, ensure that i is an integer, that is, it is determined that the starting grid point index of the grid point interval where the fixed-point information x r is located is i, and the ending grid point index is i + 1, that is, the grid point interval is determined as [x i , x i+1 .

[0091] S403. Determine the adjacent grid points on each dimension according to the grid point intervals where the fixed-point information of each dimension is located.

[0092] Exemplarily, continue to take the x-dimension as an example. On the basis of determining that the grid point interval corresponding to the fixed-point information x r on the dimension is [x i , x i+1 , it can be determined that the fixed-point information x r on the x-dimension is located between x i and x i+1 , that is, the adjacent grid points on the x-dimension are x i and x i+1 . And so on, the adjacent grid points on each dimension can be determined respectively.

[0093] Based on this, since the grid points in the grid space are equally spaced, the position where the fixed-point information is located can be quickly determined directly through simple arithmetic operations, such as subtraction, division, and rounding operations, without the need for complex searches or iterations. And the time complexity of this method for determining adjacent grid points is O(1), because it does not depend on the grid size or specific values, but is based on fixed mathematical relationships.

[0094] S302. Perform permutations and combinations on the adjacent grid points on each dimension to determine multiple neighboring grid points.

[0095] Optionally, the adjacent lattice points in each dimension include a first adjacent lattice point and a second adjacent lattice point, that is, for the fixed-point information in each dimension, there are two corresponding adjacent lattice points. For example, for a given dimension such as the x dimension, according to the fixed-point information x in the x dimension r the first adjacent lattice point is calculated as x i , in the same dimension, the next lattice point adjacent to the fixed-point information x r , that is, the second adjacent lattice point is x i+1 .

[0096] Optionally, the first adjacent lattice points in the target dimension are respectively combined with the first adjacent lattice points and the second adjacent lattice points in the remaining dimensions except the target dimension to obtain a first set of lattice points, and the second adjacent lattice points in the target dimension are respectively combined with the first adjacent lattice points and the second adjacent lattice points in the remaining dimensions except the target dimension to obtain a second set of lattice points; the lattice points in the first set of lattice points and the second set of lattice points are arranged to obtain a plurality of neighboring lattice points.

[0097] Exemplarily, the target dimension is any one of the six dimensions. Taking the target dimension as the x dimension as an example, the first adjacent lattice point x i in the target dimension is combined with the first adjacent lattice points and the second adjacent lattice points in the remaining five dimensions except the target dimension to form a set of lattice points as the first set of lattice points. Similarly, the second adjacent lattice point x i+1 in the target dimension is combined with the first adjacent lattice points and the second adjacent lattice points in the remaining five dimensions except the target dimension to form a set of lattice points as the first set of lattice points.

[0098] Furthermore, all the lattice points in the first set of lattice points and the second set of lattice points are sorted to ensure no duplicates and are arranged in a certain logical order, such as arranging them according to the distance from the current position, to obtain a plurality of neighboring lattice points.

[0099] It should be noted that since there are two possible choices (the first adjacent lattice point and the second adjacent lattice point) in each dimension, in a six-dimensional grid space, there are a total of 2 6 that is, 64 different combination methods, and each combination method represents a potential neighboring state. The number of the finally obtained neighboring lattice points is also 64.

[0100] Based on this, by using the two closest lattice points found in each dimension, such as the first adjacent lattice point and the second adjacent lattice point, and then combining these adjacent lattice points, all the neighboring lattice points around the current position of the robot can be effectively found, which helps to more accurately simulate the environment and predict the influence of the force field.

[0101] As a possible implementation, in step S102 above, the environmental force field information corresponding to the current position of the robot is determined according to the mechanical information on each neighboring grid point, including: performing dimensionality reduction interpolation processing on the mechanical information on each neighboring grid point based on a preset interpolation strategy to determine the environmental force field information corresponding to the current position of the robot.

[0102] Optionally, the preset interpolation strategy can be selected according to different application scenarios. The preset interpolation strategy includes linear interpolation, bilinear interpolation, trilinear interpolation, spline interpolation, etc. Since the grid space in this application is six-dimensional and the data form of the mechanical information on each neighboring grid point is relatively complex, if interpolation calculation is directly performed in the six-dimensional space, it is too complex and resource-consuming. Therefore, in this application, the mechanical information on each neighboring grid point is first subjected to dimensionality reduction processing, and then the interpolation operation is performed in the low-dimensional space, so as to effectively utilize the mechanical information on the surrounding grid points to estimate the environmental force field information of the current position of the robot.

[0103] Figure 5 The flowchart of a dimensionality reduction interpolation processing method provided by an embodiment of the present application is shown. Refer to Figure 5 As described above, in the above step, dimensionality reduction interpolation processing is performed on the mechanical information on each neighboring grid point based on a preset interpolation strategy to determine the environmental force field information corresponding to the current position of the robot, which specifically includes the following steps:

[0104] S501. Divide multiple neighboring grid points according to the coordinate values corresponding to the first dimension of each neighboring grid point to obtain a first grid point group and a second grid point group.

[0105] Exemplarily, for a six-dimensional grid space, interpolation can be performed in a recursive manner. Taking 64 neighboring grid points as an example, since there are two possible choices for each dimension, namely the first adjacent grid point and the second adjacent grid point, the 64 neighboring grid points are divided into two groups according to the coordinate values of the first dimension, such as the x dimension. Some grid points will include x1 as the coordinate value of their x-axis, while some grid points will include x2 as the coordinate value of their x-axis.

[0106] Exemplarily, all neighboring grid points are divided into two groups according to whether their coordinate values on the x-axis are x1 or x2, so as to obtain a first grid point group and a second grid point group. Among them, the coordinate values of the grid points in the first grid point group on the x-axis are x1, and the coordinate values of the grid points in the first grid point group on the x-axis are x2. For example, when interpolation is required for the mechanical information Assume that the 64 adjacent grid points of X and their corresponding force field sampling values are Taking interpolation of the x dimension as an example, all 64 neighboring grid points can be divided into the following two groups:

[0107]

[0108] S502. Interpolate the mechanical information in the first dimension of each lattice point in the first lattice point group and the second lattice point group using a preset interpolation strategy to obtain the interpolation result in the first dimension.

[0109] Optionally, the difference result in the first dimension includes multiple first neighboring lattice points interpolated in the first dimension, and the number of multiple first neighboring lattice points is one-half of the number of multiple neighboring lattice points.

[0110] Exemplarily, the preset interpolation strategy can select a one-dimensional interpolation algorithm, such as a linear interpolation algorithm, to determine the mechanical information (such as the force field value) at the fixed point x. r The principle of the linear interpolation algorithm is that it is assumed that there are paired arrays (x1, V1), (x2, V2) of the lattice point x vector V. Then, for the actual point with coordinates x, the interpolation is as shown in the following expression (1):

[0111]

[0112] Where V represents the result vector obtained by interpolation, V1 represents the vector corresponding to the coordinate point x1, V2 represents the vector corresponding to the coordinate point x2, x represents the coordinate point at which the mechanical information needs to be interpolated, and x1 and x2 respectively represent the coordinate points of the known mechanical information.

[0113] Exemplarily, taking the above first lattice point group and second lattice point group as an example, for a group (j, k, l, m, n), applying the one-dimensional interpolation formula to the paired array (x, F) represented by the above expression (1), the result of one-dimensional interpolation in the x direction as shown in the following expression (2) can be obtained:

[0114]

[0115] Where F jklmn represents the force that needs to be interpolated at the coordinate point F 1jklmn represents the known force value at the coordinate point F 2jklmn represents the known force value at the coordinate point x r represents the x coordinate of the coordinate point at which interpolation is required, and x1 and x2 represent the coordinate points of the known mechanical information.

[0116] Exemplarily, based on the above expression (2), the mechanical information interpolated in the x-axis direction can be obtained, thereby forming the interpolation result in the first dimension. The interpolation result in the first dimension includes 32 new lattice points. That is, after interpolation in the x-axis direction, the original 64 lattice points are reduced to 32. The new lattice points contain the interpolated mechanical information, and the number of lattice points is half of the original number of lattice points.

[0117] S503. Divide the multiple first neighboring grid points according to the coordinate values corresponding to the second dimension of each first neighboring grid point to obtain a third grid point group and a fourth grid point group, and perform interpolation processing on the mechanical information of the second dimension of each grid point in the third grid point group and the fourth grid point group by using a preset interpolation strategy. Iteratively execute until there is only one target grid point in the grid point group, and use the mechanical information corresponding to the target grid point as the environmental force field information corresponding to the current position of the robot.

[0118] Exemplarily, the second dimension is, for example, the y-dimension. That is, repeat the steps of S501 to S502 above on the y-axis. For the obtained 32 new grid points, divide the 32 new grid points into two groups according to the coordinate values of the second dimension such as the y-dimension. Some grid points will include y1 as the coordinate value of their y-axis, while some other grid points will include y2 as the coordinate value of their y-axis.

[0119] Exemplarily, divide the 32 new grid points into two groups according to whether their coordinate values on the y-axis are y1 or y2, so as to obtain a third grid point group and a fourth grid point group. Among them, the coordinate value of each grid point in the third grid point group on the y-axis is y1, and the coordinate value of each grid point in the fourth grid point group on the y-axis is y2. On this basis, apply the one-dimensional interpolation algorithm shown in the above formula (2) to the third grid point group and the fourth grid point group for interpolation operation, so as to obtain the interpolation result in the y-axis direction. And so on, perform interpolation processing on the remaining dimensions in turn, each time halving the number of grid points until finally there is only one target grid point left, and this target grid point represents the current position of the robot of the environmental force field information.

[0120] Based on this, the dimensionality reduction interpolation processing adopted in this application is essentially a recursive or iterative multi-dimensional interpolation technique, which simplifies the high-dimensional interpolation problem by gradually reducing the dimension. And each iteration reduces one dimension of the data, while maintaining the effectiveness and accuracy of the mechanical information. The final result obtained is a mechanical information value that accurately reflects the environmental force field of the current position of the robot.

[0121] As a possible implementation manner, the above step S104 determines the current compensation external force according to the environmental force field information corresponding to the current position of the robot and the current sensor information, including: determining a second difference between the current environmental force field information and the current sensor information, and using the second difference as the current compensation external force.

[0122] Optionally, the compensating external force is determined mainly to remove the influence of the inherent force field in the environment on the robot operation, so as to ensure that the robot only responds to the actual input or intention of the user. During the real-time control process, the environmental force field information obtained through the dimensionality reduction interpolation process is subtracted from the total force and torque measured by the sensor, and the current compensating external force can be obtained. This current compensating external force is the real external force applied by the operator. In this way, it can be ensured that the robot only responds to the intention of the operator and is not affected by the environmental force field.

[0123] Exemplarily, the environmental force field information of the current position of the robot has been obtained according to the dimensionality reduction interpolation process The environmental force field information includes all the forces and torques acting on the current position, and the six-axis force sensor installed at the end of the robot can measure and provide the current force state of the robot in real time, including the forces and torques from all sources. By calculating the second difference between the environmental force field information and the total force and torque measured by the sensor, and taking this second difference as the current compensating external force, this current compensating external force represents the external force truly applied by the user after removing the environmental influence. Further, applying the obtained compensating external force to the control system of the robot can reduce the interference force in the environment, so that the robot can respond more precisely to the instructions of the user.

[0124] Based on this, the present application enables the robot to more accurately respond to the external force input of the operator or control the output external force by sampling and learning and compensating in real time the traction force generated by any environmental force field, avoiding unnecessary movements caused by being pulled by the external force field. In this way, not only the physical burden of the operator is reduced, but also the force control accuracy of the robot is improved.

[0125] Based on the same inventive concept, an embodiment of the present application also provides a force field compensation device corresponding to the force field compensation method. Since the principle of solving problems by the force field compensation device in the embodiment of the present application is similar to that of the above force field compensation method in the embodiment of the present application, the implementation of the force field compensation device can refer to the implementation of the force field compensation method, and the repeated parts will not be described again.

[0126] Referring to Figure 6 As shown, it is a schematic structural diagram of a force field compensation device provided by an embodiment of the present application. The force field compensation device 600 includes: an acquisition module 601, a first determination module 602, a reading module 603, and a second determination module 604, where:

[0127] The acquisition module 601 is configured to acquire the mechanical information of the robot in the grid space of the preset dimension. A mechanical sensor is installed at the end of the robot, and the mechanical information includes the forces and torques of the robot at each grid point in the grid space, where each grid point represents a specific position in the grid space;

[0128] The first determination module 602 is configured to determine a plurality of adjacent grid points corresponding to the current position of the robot in the grid space according to the current position information of the robot, and determine the environmental force field information corresponding to the current position of the robot according to the mechanical information on each adjacent grid point;

[0129] The reading module 603 is configured to read the current sensor information of the mechanical sensor;

[0130] The second determination module 604 is configured to determine the current compensation external force according to the environmental force field information and the current sensor information corresponding to the current position of the robot, and perform force field compensation based on the current compensation external force.

[0131] Based on this, for the force field compensation device according to the embodiments of the present application, the grid space of the preset dimension is obtained by performing grid processing on the Cartesian space. The construction of the grid space is directly based on the sampling data rather than relying on a specific mathematical model. That is to say, the grid space of the present application is essentially a sampling force field model. Therefore, the grid space of the present application can be applied to force field environments of any complexity. On this basis, through the domain search algorithm and the interpolation algorithm, the force and torque information can be quickly and accurately obtained at any position in the grid space. Furthermore, through the interpolation analysis of the sensor information and the environmental force field information, the effective compensation for environmental interference is realized, and the accuracy and stability of the robot operation are improved. Based on this, the present application establishes a detailed force field model by performing grid processing on the Cartesian space and traversing and sampling the force and torque information to establish the grid space, and uses the grid space to efficiently obtain the force and torque information at any position with a low computational complexity, so as to realize the online real-time compensation for the external force field.

[0132] In a possible implementation manner, the above-mentioned acquisition module 601 is further configured to:

[0133] Determine the grid division parameters of the Cartesian space according to the spatial range information of the sampling force field. The Cartesian space is a six-dimensional coordinate system space based on which the robot interacts with the outside world. The grid division parameters include the boundary range, the number of grid points, and the grid point distribution order of each dimension;

[0134] Perform grid processing on the Cartesian space according to the grid division parameters to obtain a grid space, where the grid space is used to indicate the distribution of the sampling force field within the range of the Cartesian space.

[0135] In a possible implementation manner, the above-mentioned first determination module 602 is specifically configured to:

[0136] Determine the adjacent grid points on each dimension according to the fixed-point information of each dimension in the grid space in the current position information. The current position information is six-dimensional coordinate information including translational motion information and rotational motion information;

[0137] Arrange and combine adjacent lattice points in each dimension to determine multiple neighboring lattice points.

[0138] In a possible implementation manner, the above-mentioned first determination module 602 is specifically configured to:

[0139] Determine the resolution of each dimension according to the number of lattice points in each dimension;

[0140] Determine the lattice point interval where the fixed-point information of each dimension is located according to the resolution of each dimension, the fixed-point information of each dimension, and the boundary range of each dimension;

[0141] Determine the adjacent lattice points in each dimension according to the lattice point interval where the fixed-point information of each dimension is located.

[0142] In a possible implementation manner, the above-mentioned first determination module 602 is specifically configured to:

[0143] For each dimension, determine the first difference between the resolution of the dimension and the fixed-point information of the dimension, and determine the lattice point interval parameter based on the first difference and the boundary range of the dimension;

[0144] Perform rounding processing on the lattice point interval parameter to determine the lattice point interval where the fixed-point information of the dimension is located.

[0145] In a possible implementation manner, the adjacent lattice points in each dimension include first adjacent lattice points and second adjacent lattice points. The above-mentioned first determination module 602 is specifically configured to:

[0146] Combine the first adjacent lattice points in the target dimension with the first adjacent lattice points and the second adjacent lattice points in the other dimensions except the target dimension respectively to obtain a first lattice point set, and combine the second adjacent lattice points in the target dimension with the first adjacent lattice points and the second adjacent lattice points in the other dimensions except the target dimension respectively to obtain a second lattice point set;

[0147] Arrange the lattice points in the first lattice point set and the second lattice point set to obtain multiple neighboring lattice points.

[0148] In a possible implementation manner, the above-mentioned first determination module 602 is specifically configured to:

[0149] Based on a preset interpolation strategy, perform dimensionality reduction interpolation processing on the mechanical information at each neighboring lattice point to determine the environmental force field information corresponding to the current position of the robot.

[0150] In a possible implementation manner, the above-mentioned first determination module 602 is specifically configured to:

[0151] Divide the multiple neighboring lattice points according to the coordinate values corresponding to the first dimension of each neighboring lattice point to obtain a first lattice point group and a second lattice point group;

[0152] Interpolate the mechanical information of the first dimension of each lattice point in the first lattice point group and the second lattice point group by using a preset interpolation strategy to obtain a first-dimension interpolation result. The first-dimension interpolation result includes a plurality of first neighboring lattice points interpolated in the first dimension, and the number of the plurality of first neighboring lattice points is one-half of the number of the plurality of neighboring lattice points.

[0153] Divide the plurality of first neighboring lattice points according to the coordinate values corresponding to the second dimension of each first neighboring lattice point to obtain a third lattice point group and a fourth lattice point group, and interpolate the mechanical information of the second dimension of each lattice point in the third lattice point group and the fourth lattice point group by using the preset interpolation strategy, and iterate until there is only one target lattice point in the lattice point group, and use the mechanical information corresponding to the target lattice point as the environmental force field information corresponding to the current position of the robot.

[0154] In a possible implementation manner, the above-mentioned second determination module 604 is specifically configured to:

[0155] Determine a second difference between the environmental force field information and the current sensor information, and use the second difference as the current compensation external force.

[0156] For the processing flow of each module in the device and the interaction flow between each module, reference may be made to the relevant descriptions in the above method embodiments, which will not be elaborated here.

[0157] The embodiment of the present application further provides an electronic device 700, as Figure 7 shown, which is a schematic structural diagram of the electronic device 700 provided by the embodiment of the present application, including: a processor 701, a memory 702, and optionally, a bus 703 may also be included. The memory 702 stores machine-readable instructions executable by the processor 701. When the electronic device 700 runs, the processor 701 communicates with the memory 702 through the bus 703, and when the machine-readable instructions are executed by the processor 701, the steps in the force field compensation method described in any one of the above are executed.

[0158] The embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the steps in the force field compensation method described in any one of the above are executed.

[0159] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the method embodiments, and will not be elaborated herein. In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some communication interfaces. The indirect couplings or communication connections of the devices or modules can be in electrical, mechanical, or other forms.

[0160] In addition, in each embodiment of the present application, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, and other various media that can store program codes.

[0161] The above are only the specific implementation manners of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all should be covered by the protection scope of the present application.

Claims

1. A force field compensation method, characterized in that: include: Acquire mechanical information of the robot in a grid space of preset dimensions, wherein a mechanical sensor is installed at the end of the robot, and the mechanical information includes force and torque of the robot at each grid point in the grid space, wherein each grid point represents a specific position in the grid space; Determine, according to the current position information of the robot, a plurality of adjacent grid points corresponding to the current position of the robot in the grid space, and determine, according to the mechanical information on each of the adjacent grid points, the environmental force field information corresponding to the current position of the robot; Reading current sensor information of the mechanical sensor; According to the environmental force field information corresponding to the current position of the robot and the current sensor information, a current compensation external force is determined, and force field compensation is performed based on the current compensation external force.

2. The method according to claim 1, characterized in that Before obtaining the mechanical information of the robot in the grid space of the preset dimension, the method further includes: Determine the grid division parameters of the Cartesian space according to the spatial range information of the sampled force field, wherein the Cartesian space is a six-dimensional coordinate system space based on which the robot interacts with the outside world, and the grid division parameters include the boundary range of each dimension, the number of grid points, and the order of grid point distribution; The Cartesian space is gridded according to the grid division parameters to obtain the grid space, and the grid space is used to indicate the distribution of the sampling force field within the Cartesian space.

3. The method according to claim 1, characterized in that Determining, based on the current position information of the robot, a plurality of adjacent grid points corresponding to the current position of the robot in the grid space includes: Determine adjacent grid points in each dimension according to the fixed point information in each dimension in the grid space in the current position information, wherein the current position information is six-dimensional coordinate information including translation motion information and rotation motion information; Adjacent grid points in each dimension are arranged and combined to determine the plurality of adjacent grid points.

4. The method according to claim 3, characterized in that The step of determining adjacent grid points in each dimension according to the fixed point information in each dimension in the grid space in the current position information includes: Determining the resolution of each dimension according to the number of grid points in each dimension; Determine the grid point interval where the fixed point information of each dimension is located according to the resolution of each dimension, the fixed point information of each dimension and the boundary range of each dimension; Adjacent grid points on each dimension are determined according to the grid point interval where the fixed point information of each dimension is located.

5. The method according to claim 4, characterized in that The determining, according to the resolution of each dimension, the fixed-point information of each dimension, and the boundary range of each dimension, the grid point interval where the fixed-point information of each dimension is located includes: For each of the dimensions, determine a first difference between the resolution of the dimension and the fixed point information of the dimension, and determine a grid point interval parameter based on the first difference and a boundary range of the dimension; The grid point interval parameters are rounded to determine the grid point interval where the fixed point information of the dimension is located.

6. The method according to claim 3, characterized in that The adjacent grid points in each dimension include a first adjacent grid point and a second adjacent grid point, and the adjacent grid points in each dimension are arranged and combined to determine the plurality of adjacent grid points, including: Combining the first adjacent grid points on the target dimension with the first adjacent grid points and the second adjacent grid points on the remaining dimensions except the target dimension to obtain a first grid point set, and combining the second adjacent grid points on the target dimension with the first adjacent grid points and the second adjacent grid points on the remaining dimensions except the target dimension to obtain a second grid point set; The grid points in the first grid point set and the second grid point set are arranged to obtain the plurality of adjacent grid points.

7. The method according to claim 1, characterized in that Determining the environmental force field information corresponding to the current position of the robot according to the mechanical information on each of the adjacent grid points includes: Based on a preset interpolation strategy, dimension reduction interpolation processing is performed on the mechanical information on each of the adjacent grid points to determine the environmental force field information corresponding to the current position of the robot.

8. The method according to claim 7, characterized in that The method of performing dimensionality reduction interpolation processing on the mechanical information at each of the adjacent grid points based on a preset interpolation strategy to determine the environmental force field information corresponding to the current position of the robot includes: Dividing the plurality of adjacent grid points according to the coordinate value corresponding to the first dimension of each adjacent grid point to obtain a first grid point group and a second grid point group; The preset interpolation strategy is used to interpolate the first-dimensional mechanical information of each grid point in the first grid point group and the second grid point group to obtain a first-dimensional interpolation result, wherein the first-dimensional interpolation result includes a plurality of first neighboring grid points interpolated in the first dimension, and the number of the plurality of first neighboring grid points is half of the number of the plurality of neighboring grid points; The multiple first neighboring grid points are divided according to the coordinate values ​​corresponding to the second dimension of each of the first neighboring grid points to obtain a third grid point group and a fourth grid point group, and the preset interpolation strategy is used to interpolate the second dimensional mechanical information of each grid point in the third grid point group and the fourth grid point group, and the process is iteratively executed until there is only one target grid point in the grid point group, and the mechanical information corresponding to the target grid point is used as the environmental force field information corresponding to the current position of the robot.

9. The method according to claim 1, characterized in that: The determining the current compensation external force according to the environmental force field information corresponding to the current position of the robot and the current sensor information includes: A second difference between the environmental force field information and the current sensor information is determined, and the second difference is used as the current compensation external force.

10. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor executes the machine-readable instructions to perform the steps of the force field compensation method as described in any one of claims 1 to 9.

Images