EA040444B1 - METHOD AND SYSTEM FOR ROBOT-MANIPULATOR MOVEMENT PLANNING BY CORRECTION OF BASE TRAJECTORIES - Google Patents

Description

Technical field

The presented technical solution relates, in general, to the field of robotics, and in particular to the method and system for planning the movement of a robot-manipulator by correcting reference trajectories, and can be used to create control systems for robotic manipulators as a module for planning trajectories of movement of a robot-manipulator in conditions of a static collision scene, in particular, although not exclusively, for solving problems of manipulating objects.

State of the art

Robots have been widely used to automate production tasks. Their use allows you to free people from heavy and monotonous work. At the heart of any production task is the movement of the robotic arm. Currently, for the process of learning to capture objects, various approaches are proposed based on algorithms and machine learning models, in particular, using neural networks, which are used to train robotic manipulators for capturing objects using previously trained capture points. This approach is explored in Dense Object Nets: Learning Dense Visual Object Descriptors By and For Robotic Manipulation (Peter R. Florence et al., CSAIL, Massachusetts Institute of Technology, 09/07/2018). A well-known solution proposes to use point clouds of objects in the process of training robotic devices in order to generate object descriptors based on photographic and 3D models of such objects for the purpose of subsequently specifying capture points and interacting with objects at these points.

Also known is a method and system for capturing an object using a robotic device, disclosed in patent RU 2700246 C1, publ. 09/20/2019. In the known solution, a robotic device is trained using a machine learning algorithm for recognizing and remembering the capture points of objects, and training is performed on data characterizing photographic images of objects and corresponding 3D models of objects in various angles; form at least one point of capture of the object presented on the graphic image of the object; receiving a photographic image of the object using the camera of the robotic device and determine the angle that displays the formed capture point of the object; receiving a three-dimensional point cloud of the object in perspective using the depth sensor of the robotic device; determine the location of the capture point of the object based on the obtained point cloud; determining the orientation and position of the gripper of the robotic device at the gripping point by calculating rotation matrices of the robotic device; the object is captured using the robotic device based on the calculation of the average rotation matrix for a given capture point.

In comparison with existing approaches to planning the trajectories of a robotic arm, the proposed approach solves a similar problem faster due to the absence of the need to form a complete trajectory that does not allow collision with objects of a static collision scene.

The essence of the technical solution

The technical problem or technical challenge posed in this technical solution is to create a new efficient, simple and reliable way to plan the motion of a robotic arm.

The technical result achieved by solving the above technical problem is to reduce the planning time of the movement of the robotic arm. The specified technical result is achieved due to the implementation of a method for planning the movement of a robotic arm, performed by at least one computing device, comprising the steps in which:

receive information about the specified position and orientation of the working body, in which it is necessary to move the working body of the robotic arm;

determine the current position and orientation of the working body of the robotic arm;

determining, on the basis of information about the predetermined position and orientation of the working body, a cell of the area of interest into which it is necessary to move the working body of the robotic arm, the area of interest being inside the working space of the robotic manipulator;

on the basis of a set of possible orientations of the working body of the robotic arm when it is in the area of the said cell, the nearest possible orientation of the working body of the robotic manipulator to a given orientation is determined;

Based on the data on the cell and the nearest possible orientation of the working body of the robot manipulator, a pre-generated direct reference trajectory is retrieved from the list stored on the computing device, moreover, the direct reference trajectory describes the movement of the robot manipulator from the initial position corresponding to the area of interest to the position in which the working body of the robot - the manipulator is inside the area of interest;

determine a set of hinge angles at which the working body of the robotic arm is at a point of a given position with a given orientation;

remove from the end of the extracted direct reference trajectory successively points until the length of the trajectory is reduced by an amount equal to the size of the cell;

add a point to the end of the trajectory corresponding to the values of the set of angles of the robot's joints

- 1 040444 manipulators, in which the working body of the robotic manipulator is at a point of a given position with a given orientation, and to the beginning - a point corresponding to the current position of the working body of the robotic manipulator;

move the working body of the robot-manipulator in accordance with the trajectory obtained at the previous stage.

In one of the particular examples of the implementation of the method, the steps are additionally performed, at which:

determine that the point of a given position of the working body is inside the region of interest;

determining the amount of deviation of the angles of the hinges of the robot-manipulator in the current position from the angles of the hinges specified for the initial position corresponding to the region of interest;

the deviation value obtained at the previous stage is compared with the threshold value, and the step of determining the nearest possible orientation of the working body of the robot-manipulator to the given orientation is performed if the mentioned deviation value does not exceed the threshold value.

In another particular example of the implementation of the method, the determination of the current position and orientation of the working body of the robot-manipulator is carried out by solving a direct kinematics problem based on its current values of the angles of the hinges. In another particular example of the implementation of the method, the stage at which the sets of hinge angles are determined, at which the working body of the robot-manipulator is at a point of a given position with a given orientation, contains the stages at which:

for a given position and orientation of the working body of the robot-manipulator, the inverse problem of kinematics is solved to obtain a set of sets of hinge angles at which the working body of the robotic manipulator is at a point of a given position with a given orientation;

determine sets of hinge angles that satisfy collision constraints;

choose as a set of hinge angles at which the working body of the robot-manipulator is at a point of a given position with a given orientation, from the set of hinge angles that satisfy collision restrictions, the set that is closest in terms of hinge angles to the last point of the extracted straight reference trajectory .

In another preferred embodiment of the claimed solution, a motion planning system for a robotic arm is provided, comprising:

robotic arm;

a computing device connected to the robotic arm;

at least one memory containing machine-readable instructions that, when executed by at least one computing device, perform the above method.

In another preferred embodiment of the claimed solution, a method for planning the motion of a robotic arm is presented, performed by at least one computing device, comprising the steps of:

receive information about the specified position and orientation of the working body, in which it is necessary to move the working body of the robotic arm;

determine the current position and orientation of the working body of the robotic arm;

determining, on the basis of information about the current position and orientation of the working body, a cell of the area of interest from which it is necessary to move the working body of the robotic arm, the area of interest being inside the working space of the robotic manipulator;

based on a set of possible orientations of the working body of the robotic arm when it is in the area of the said cell, the closest possible orientation of the working body of the robotic manipulator to the current orientation is determined;

Based on the data on the cell and the nearest possible orientation of the working body of the robotic arm, a previously generated reverse reference trajectory is retrieved from the list stored on the computing device, and the reverse reference trajectory describes the movement of the robot from the position in which the working member of the robotic manipulator is inside the area of interest, to the initial position , corresponding to the area of interest;

determine a set of hinge angles at which the working body of the robotic arm is at a point of a given position with a given orientation;

remove from the beginning of the retrieved reverse reference trajectory successively points until the length of the trajectory is reduced by an amount equal to the size of the cell;

add to the end of the trajectory a point corresponding to the values of the set of angles of the robotic arm hinges, at which the working body of the robotic manipulator is at a point of a given position with a given orientation, and to the beginning - a point corresponding to the current position of the working body of the robotic manipulator;

move the working body of the robot-manipulator in accordance with the trajectory obtained at the previous stage.

In one of the particular examples of the implementation of the method, the steps are additionally performed, at which:

- 2 040444 determine the amount of deviation of the specified position and the initial position of the working body of the robotic arm corresponding to the area of interest;

the deviation value obtained at the previous stage is compared with the threshold value, and the step of determining the nearest possible orientation of the working body of the robot-manipulator to the current orientation is performed if the said deviation value does not exceed the threshold value.

In another particular example of the implementation of the method, the determination of the current position and orientation of the working body of the robot-manipulator is carried out by solving a direct kinematics problem based on its current values of the angles of the hinges. In another particular example of the implementation of the method, the stage at which sets of hinge angles are determined, at which the working body of the robot-manipulator is at a point of a given position with a given orientation, contains the steps at which: for a given position and orientation of the working body of the robot-manipulator, the solution of the inverse kinematics tasks for obtaining a set of sets of hinge angles, in which the working body of the robotic arm is at a point of a given position with a given orientation;

determine sets of hinge angles that satisfy collision constraints;

choose as a set of hinge angles at which the working body of the robotic arm is at a point of a given position with a given orientation, from the set of hinge angles that satisfy collision constraints, the set that is closest in terms of hinge angles to the last point of the extracted back reference trajectory .

In another preferred embodiment of the claimed solution, a motion planning system for a robotic arm is provided, comprising:

robotic arm;

a computing device connected to the robotic arm;

at least one memory containing machine-readable instructions that, when executed by at least one computing device, perform the above method.

Brief description of the drawings

Signs of the advantage of the present technical solution will become apparent from the following detailed description and the accompanying drawings, in which:

in fig. 1 shows a general view of the motion planning system of a robotic arm;

in fig. 2 shows examples of the orientation of the working body along the vertical axis from the set of possible orientations of the working body of the robot-manipulator corresponding to the area of interest;

in fig. 3 shows examples of the orientation of the working body along the axes deviated from the vertical from the set of possible orientations of the working body of the robot-manipulator corresponding to the area of interest;

in fig. 4 shows an example of a straight reference path;

in fig. 5 shows an example of a reverse reference path;

in fig. 6 shows the first part of the operating mode flow diagram;

in fig. 7 shows the second part of the operating mode flow diagram;

in fig. 8 shows the third part of the operating mode flow diagram;

in fig. 9 shows the fourth part of the operating mode flow diagram;

in fig. 10 shows the fifth part of the operating mode flow diagram;

in fig. 11 shows the sixth part of the operating mode flowchart;

in fig. 12 shows an example of corrected reference paths;

in fig. 13 shows an example of the given position and orientation of the working body of the robotic arm;

in fig. 14 shows an example of a given orientation of the working body and a selected possible orientation of the working body;

in fig. 15 shows examples of solving the inverse problem of kinematics for a given position of the working body of the robotic arm;

in fig. 16 shows an example of the result of moving the robotic arm along the selected straight trajectory;

in fig. 17 shows an example of the result of moving the robotic arm along the corrected straight path;

in fig. 18 shows an example of a general view of a computing device.

Detailed description

This technical solution can be implemented on a computer in the form of an automated information system (AIS) or a machine-readable medium containing instructions for performing the above method.

Also, the technical solution can be implemented as a distributed computer system that can be installed on a centralized server (set of servers). User access to the system is possible both from the Internet and from the internal network of an enterprise/organization through a mobile communication device on which software is installed with an appropriate graphical user interface, or a personal computer with access to the web version of the system with an appropriate graphical interface user.

The terms and concepts necessary for the implementation of the present technical solution will be described below.

In this solution, a system means a computer system, computer (electronic computer), CNC (computer numerical control), PLC (programmable logic controller), computerized control systems and any other devices capable of performing a given, well-defined sequence of computational operations (actions, instructions ).

A command processing device is an electronic unit or an integrated circuit (microprocessor) that executes machine instructions (programs).

An instruction processing device reads and executes machine instructions (programs) from one or more data storage devices. The role of a storage device can be, but not limited to, hard disk drives (HDD), flash memory, ROM (read only memory), solid state drives (SSD), optical drives.

Program - a sequence of instructions intended for execution by a computer control device or a command processing device.

The working body of an industrial robot is an integral part of the manipulator, designed to interact with objects of manipulation or directly perform technological operations.

As shown in FIG. 1, the motion planning system 100 of the robotic arm contains a computing device 107 that provides the necessary processes for computing the algorithm for the operation of the robotic manipulator 103, a data transmission channel 106 that provides processes for receiving / transmitting data between elements of the system 100, as well as a remote computing device 105, for example, a cloud server for specialized tasks.

As a computing device 107 with which the user 108 interacts, for example, a personal computer, laptop, tablet, server, smartphone, server cluster, mainframe, supercomputer, thin client, system on a chip (eng. System on a Chip, abbr. .SoC), etc. The specific hardware implementation of the computing device 107 depends on the requirements and must provide the necessary digital information processing processes to ensure the proper functioning of the device 107 as part of the system 100.

Various types of communication can be used as a data transmission channel 106, for example, wired and / or wireless type, in particular: LAN, WAN, WLAN, Bluetooth, Wi-Fi, GSM/GPRS/LTE/5G, Ethernet, USB, RS232, RJ45 etc. Remote computing device 105 is typically a cloud storage or server for specialized tasks or program logic. A description of the operation of the device 105 will be disclosed later in this description.

As a robotic arm 103, it can be used as a stationary robotic arm for industrial or collaborative purposes, for example, Universal Robotics™, KUKA™, Fanuc™, ABB™, etc., and robotic manipulators mounted on mobile autonomous robotic devices, as limbs, in particular robotic arms. There are two modes of operation of the system (100) planning the motion of the robotic arm: the mode of formation of reference trajectories and the mode of operation.

In the mode of formation of reference trajectories in the working area of the robotic arm 103, the user 108 selects one or more areas of interest by means of the prior art methods known from the prior art, for example, in the area of the table 104 (Fig. 1). The region of interest is the region of the workspace of the robotic arm, which is, for example, a box 101, consisting of cells 102. In FIG. 1 cell, marked with position 102, is in the seventh row in length, in the first row in width, in the first row in height and has serial number 31. The user sets the position and orientation of the region of interest, as well as the dimensions and number of cells in height, length and width parallelepiped 101. The area of interest corresponds to the initial position of the robotic arm 103, which is also set by the user 108 by means of the computing device 107 in the form of a set of hinge angles that characterizes the position and orientation of the working body of the robotic arm 103. from the virtual model of the robotic arm implemented on the basis of the computing device 105, or from the controller of the real robotic manipulator 103, and from them the computing device 105 calculates the position and orientation of the working body of the robotic manipulator 103 by solving a direct kinematics problem. In this case, the initial position of the robotic arm must be such that the working body is outside the area of interest.

Also, the area of interest corresponds to a set of possible orientations of the working body of the robotic arm 103 (Fig. 2 and 3), which is formed by the user 108 and stored in the computer

- 4 040444 on the device 105 as a list. In the list for each cell 102 of the box 101, the user 108 defines/sets the set of orientation of the working body of the robot manipulator 103 when it is in the area of the cell 102. The orientation set of the working body of the robot manipulator 103 for the cell may contain, for example, vertical axis (FIG. 2, positions 201-208), as well as orientation along axes deviated from the vertical axis by user-specified angles 108 (FIG. 3, positions 301-308). These orientations, together with the coordinates of the cell centers, are used in the formation of straight reference trajectories as target positions and orientations into which the working body of the robot manipulator 103 should go from the initial position when moving along the reference trajectories (Fig. 4).

For each cell 102 and a set of possible orientations of the working body of the robotic manipulator 103 given for them, the computing device 105 generates a forward and reverse reference trajectory of the robotic manipulator 103 (Fig. 4 and 5), which are set in space by a set of angles of the hinges of the robotic manipulator 103 for each points of the mentioned trajectory. A direct path is formed by executing any available known planning algorithms, such as the algorithm disclosed in RRT-connect: An efficient approach to single-query path planning (J. Kuffner et. al., Computer Science Dept, Stanford University, 2000) or the algorithm described in CHOMP: Gradient Optimization Techniques for Efficient Motion Planning (N. Ratliffy et. al., Robotics Institute, Carnegie Mellon University, 2009). The direct trajectory describes the movement of the robotic arm 103 from the initial position corresponding to the area of interest to the position in which the working body of the robotic arm 103 is located in the center of the cell 102 with a given (possible) orientation for this cell 102 in accordance with the list stored in the computing device 105 Accordingly, the direct path information generated by the computing device 105 contains a sequence of path points, each of which corresponds to a set of angles, velocities and accelerations of the joints of the robot manipulator 103.

The reverse trajectory describes the movement of the robot arm 103 from the cell 102 to the initial position of the robot arm 103 corresponding to the region of interest. To form a reverse trajectory, a sequence of points of a direct trajectory, each of which corresponds to a set of angles, velocities and accelerations of the robotic arm joints, is set by the computing device 105 in reverse order. At the same time, the angles of the hinges remain unchanged, and the speeds and accelerations of the hinges of the robot-manipulator 103 are recalculated in such a way that when the robot-manipulator 103 moves along the trajectory, its movement is smooth. All reference trajectories are recorded in a database, which can optionally be equipped with the computing device 105.

In the operating mode (Fig. 6-11) from a higher-level control system (not shown in the drawings) or from a computing device 107, information about a given (desired) position and orientation 601 (Fig. 6) is supplied to the input of the computing device 105 (Fig. 6), with which it is necessary to move the working body of the robot-manipulator. At the beginning, by solving the direct kinematics problem, using information about the angles of the robotic arm joints coming from the virtual model of the robot or from the controller of the real robot, the current position and orientation of the working body 602 are calculated. which is the point of the desired or current position of the working body. If both the point of the desired and the point of the current position of the working body are outside the areas of interest, then any available known algorithm for planning trajectories for the desired position and orientation of the working body 1101 (Fig. 11) is performed. Otherwise, two options for the operation of the system 100 are possible.

If inside the area of interest there is a point of the specified position of the working body 604 (the case shown in Fig. 4), then the serial number of the cell 701 (Fig. 7) containing this point is calculated, and the proximity of the angles of the hinges of the robot-manipulator in the current and initial position 702 corresponding to the region of interest. If the difference between the angles of the hinges exceeds a threshold value, then any available known path planning algorithm is executed for the desired position and orientation of the end effector 1101 (FIG. 11). If the difference between the angles of the hinges does not exceed the threshold value, then among all the direct trajectories recorded in the database for the found cell, a trajectory 703 is searched for, at the last point of which the possible orientation of the working body of the robot manipulator is closest to the specified orientation. For a given position and orientation of the working body, the system solves the inverse kinematic problem (IKP) 704. Its solutions are sets of angles of the hinges of the robot-manipulator, in which the working body is in a given position with a given orientation. The virtual robot model is set up in a configuration corresponding to the hinge angles of each solution (j) of the kinematics inverse problem 901 (FIG. 9) and checked for self-intersection and intersection with objects in the virtual scene (for example, table 104 in FIG. 1) by performing any known collision detection algorithms 902. If no solution (j) satisfies the collision restrictions imposed by objects located in the working area of the robotic arm, then any available known trajectory planning algorithm is performed for the desired position and orientation of the working body 1101 (Fig. 11 ). Otherwise, among the solutions (j) satisfying the collision constraints that

- 5 040444 objects are located in the working area of the robotic arm, one is selected, the hinge angles of which are closest to the hinge angles of the robotic arm at the end of the selected trajectory 903. in FIG. 12 shows the selected reference path in dashed line, 1202 the first point of the selected reference path, 1203 the last point of the selected reference path. Several points are removed from the end of the selected reference trajectory so that the length of the trajectory is reduced by an amount equal to the size of the cell 904. A point 1204 is added to the end of this trajectory, containing the angles of the hinges of the selected solution of the inverse problem of kinematics 905 (Fig. 9), and to the beginning - point 1201 containing the angles of the robotic arms at the current position 906. The resulting trajectory is shown in FIG. 12 as a solid line. Then, for the points of the obtained trajectory, the velocities and accelerations of the hinges of the robot-manipulator are calculated so that when the robot-manipulator moves along the trajectory, its movement is smooth 1102 (Fig. 11). The values of speeds and accelerations of the joints of the robotic arm depend on the desired speed profile. There are many well-known algorithms that solve this problem. The choice of a specific algorithm does not affect the operation of the system under consideration. After that, the trajectory is ready to be sent for execution 1103.

If inside the area of interest there is a point of the current position of the working body 801 (the case presented in Fig. 5), then in accordance with the algorithm presented in Fig. 8, the sequence number of the cell 802 containing this point is calculated, and the proximity of the specified position points and the initial position of the robotic arm 803 corresponding to the region of interest is checked. If the distance between these points exceeds a threshold value, then any available known trajectory planning algorithm is executed for the desired position and orientation of the end effector 1101 (FIG. 11). If the distance between these points does not exceed the threshold value, then among all the return trajectories recorded for the found cell, a trajectory 804 is searched for, at the first point of which the hinge angles are closest to the current robotic arm hinge angles. For a given position and orientation of the worker, the system solves the inverse problem of kinematics 805. Its solutions are sets of angles of the robot-manipulator hinges, in which the working body is in a given position with a given orientation.

The virtual robot model is set up in a configuration corresponding to the hinge angles of each solution to the inverse kinematics problem 1001 (FIG. 10) and checked for self-intersection and intersection with objects in the virtual scene (for example, with the table in FIG. 1) by running any available collision detection algorithms 1002. If no solution (j) satisfies the collision constraints imposed by objects located in the working area of the robotic arm, then any available known trajectory planning algorithm is executed for the desired position and orientation of the working body 1101 (Fig. 11). Otherwise, among the solutions that satisfy the collision constraints that are imposed by objects located in the working area of the robot manipulator, one is selected, the hinge angles of which are closest to the angles of the hinges of the robot manipulator at the end of the selected trajectory 1003. in FIG. 12 shows the selected reference path in dashed line, 1203 the first point of the selected reference path, 1202 the last point of the selected reference path. Several points are removed from the beginning of the selected trajectory in such a way that the length of the trajectory is reduced by an amount equal to the size of cell 1004. A point 1204 is added to the beginning of this trajectory, containing the angles of the joints of the robot-manipulator in the current position 1005, and to the end - a point 1201 containing the angles hinges of the chosen solution of the inverse problem of kinematics 1006. Then, for the points of the obtained trajectory, the velocities and accelerations of the robot-manipulator hinges are calculated so that when the robot-manipulator moves along the trajectory, its movement is smooth 1102 (Fig. 11). The velocities and accelerations of the robotic arm hinges depend on the desired velocity profile set, for example, by the user 108. There are many well-known algorithms that solve this problem. The choice of a specific algorithm does not affect the operation of the system under consideration. After that, the trajectory is ready to be sent for execution 1103.

An example of the implementation of the invention

Let us consider the operation of the system (100) for planning the motion of a robotic arm using a specific example. In FIG. 1 shows the working area of the robot manipulator 103, inside which the table 104 is located. The Z axis of the inertial coordinate system is directed upwards, the X axis is oriented in the direction from the center of the table to the robot manipulator, the Y axis complements the X and Z axes to the right trio of vectors. The position of the robotic arm 103 coincides with the position of the inertial coordinate system. The base of the robot is deployed relative to the inertial coordinate system around the Z axis at an angle equal to -42.5°.

In the mode of generating reference trajectories, the user 108 selects one region of interest 101 by means of a computing device 107 above the table 104. For example, the center of the region of interest can be specified by the vector [-0.730; 0.0; 0.230] and the quaternion [0.0; 0.0; 0.0; 1,0] in the inertial coordinate system. Each cell of the region of interest 102 may be in the form of a cube with an edge length of 0.088 m. The region of interest may contain, for example, 7 cells in length, 5 cells in width, and 4 cells in height of box 101.

Next, the user 108 by means of the computing device 107 selects the initial

- 6 040444 position of the robotic arm 103 corresponding to the region of interest. For example, as the mentioned initial position, the angles of the manipulator hinges can be set sequentially from the base of the robot to the working body - [50.0; -80.0; -80.0; -115.0; 90.0; 0.0]. Information about the initial position from the computing device 107 enters the computing device 105, which on its basis, using methods known from the prior art, calculates the position and orientation of the working body of the robot-manipulator 103 in the inertial coordinate system, information about which can be represented as a vector [-0.543 ; 0.075; 0.716] and quaternion [0.737; 0.0; -0.676; 0.016].

The set of possible orientations of the end effector of the robot manipulator 103, specified by the user 108 for each cell, may consist of 16 rotations shown in FIG. 2 and 3. They are 8 orientations along the vertical axis and 8 orientations along the axes, the angle between each of which and the vertical axis is 30°.

Further, based on the information about the initial position and the coordinates of the centers of cells 102, the computing device 105 generates reference trajectories for each cell of the parallelepiped 101 and for each possible orientation of the working body of the robot-manipulator 103 from the previously generated set specified for cells 102. When forming direct reference trajectories, the coordinates the centers of the cells 102 of the region of interest and the possible orientations of the end effector are used as target positions and orientations to which the end effector of the robotic arm 103 exits the start position corresponding to the area of interest. The reverse reference paths are obtained by the system by inverting the forward reference paths. The reference trajectories are recorded in the database with the following fields: unique identifier of the region of interest, cell serial number, ordinal number of the working body orientation, identifier of the reference trajectory type (forward or reverse) and reference trajectory. In the operating mode, the input of the system 100, for example, from the user 108 receives information about the specified position and orientation of the working body, in which it is necessary to move the working body of the robot-manipulator 103. The computing device 105 requests the current values of the angles of the hinges from the virtual model of the robot or from the robot controller -manipulator 103. According to the current values of the angles of the hinges, the computing device 105 performs the direct task of kinematics, the solution of which is information about the current position and orientation of the working body of the robot-manipulator 103.

For example, in accordance with FIG. 13, information 1301 about the target position and orientation of the end effector to which the end effector of the robot arm 1302 needs to be moved can be represented in the inertial coordinate system by the vector [-0.892; 0.061; 0.205] and quaternion [-0.539; -0.280; 0.770; -0.196]. Information about the current position of the robot arm 1302, requested by the computing device 105 from the virtual model of the robot or from the controller of the robot arm 1302, may contain a set of angles of the robot arm 1302, for example, the values of which can be represented by a sequence from the robot base to the end effector - [49,466; -80.140; -81.054; -113.738; 91.489; 2.133]. The current position and orientation of the robotic arm 1302, calculated by the computing device 105 as a result of solving the direct kinematics problem for the specified set of angles, in the inertial coordinate system will be represented by the vector [-0.544; 0.073; 0.715] and quaternion [0.739; 0.002; -0.671; 0.019].

In FIG. 13 example, the point of the specified position of the working body, to which it is necessary to move the working body of the robot manipulator 1302, is located inside the region of interest 1303, namely, inside the cell 1304, which is in the sixth row in length, in the third row in width, in the second row in height and has serial number 63.

Next, the computing device 105 determines the amount of deviation of the angles of the joints of the robot arm 1302 at the initial position from the angles of the joints of the robot arm 1302 at the current position, after which the amount of deviation by the computing device 105 of the robot arm 1302 is compared with a threshold value. In the example shown, the articulated angles of the robotic arm 1302 at the initial position corresponding to the region of interest differ from the articulated angles of the robotic arm 1302 at the current position by an amount not exceeding a threshold value.

At the next stage, the computing device 105 for the found cell 1304 retrieves from the database, which the said device 105 can be equipped with, a set of possible orientations of the working body of the robot-manipulator 1302 when it is in the area of the cell 1304, after which the computing device 105, based on the values of the angles of the hinges , contained in the information about the specified position and orientation of the working body, in which it is necessary to move the working body of the robotic arm 1302, and the values of the angles of the hinges of the mentioned extracted set of possible orientations of the working body determines the nearest possible orientation of the working body of the robotic arm 1302, available for its location in the area of the cell 1304.

In the presented example, the closest orientation of the working body of the robot manipulator 1302, available for its location in the cell area 1304, is the possible orientation of the working number at number 10 (position 302 in Fig. 3), specified in the inertial coordinate system by the quaternion

- 7 040444

[-0.456; -0.195; 0.794; -0.350]. Those. the angle by which it is necessary to turn the working body from the selected possible orientation to the orientation of the working body specified by the user 108 in accordance with the information about the position and orientation of the working body is minimal among similar angles from the set of possible orientations. In FIG. 14 shows examples of a given orientation of the end effector 1401 and a selected possible orientation of the end effector 1402.

Next, according to the cell number 1304, in which the point of the specified position of the working body is located, and the data of the nearest possible orientation of the working body of the robot manipulator 1302, the direct reference trajectory is retrieved from the database by the computing device 105.

After that, for a given position and orientation of the working body of the robot-manipulator 1302, the inverse kinematics problem is solved by the computing device 105 to obtain sets of angles of the hinges of the robot-manipulator 1302, in which the working body is in a given position with a given orientation. In accordance with the presented example, for a given robot manipulator 1302, in the general case, there are 8 solutions, which are sets of hinge angles (Fig. 15), in which the end effector of the robot manipulator is at a given position with a given orientation. The four solutions do not satisfy the collision constraints because, in this configuration, the elbow of the robot arm 1302 is in contact with the table. At the last point of the selected trajectory, the angles of the joints of the robotic arm 1302 have the values [45.745; -115.115; -105.229; -27.290; 109.951; -39,276]. Among the solutions of the inverse problem of kinematics that satisfy collision constraints, the solution [48,960; -114.466; -101.212; -35.291; 87.414; -33.630] (position 1505 in Fig. 15), the set of hinge angles of which is selected by the computing device 105 as a solution to the inverse problem of kinematics.

Next, the computing device 105 sequentially removes points from the end of the extracted straight trajectory until the length of the trajectory decreases beyond the length of the edge of cell 1304 (Fig. 13). Then, a point is added to the end of this trajectory corresponding to the values of the angles of the joints of the robotic arm 1302 of the selected solution of the inverse kinematics problem, and to the beginning - a point corresponding to the current angles of the joints of the robot manipulator 1302. At the end, the computing device 105 recalculates the values of the velocities and accelerations of the angles of the joints at each point to ensure smooth movement of the robot manipulator along the trajectory. For comparison, in Fig. 16 shows the result of moving the robotic arm along the selected straight trajectory, and FIG. 17 - along the same trajectory after correction. Accordingly, if the point of the current position of the working body of the robotic arm 1302 (Fig. 13) is in the region of interest 1303, then the computing device 105 in the previously described way: determines, based on information about the current position and orientation of the working body, the cell of the region of interest from which it is necessary to move working body of the robotic manipulator; based on a set of possible orientations of the working body of the robotic arm when it is in the area of the said cell, determines the closest possible orientation of the working body of the robotic manipulator to the current orientation; on the basis of data about the cell and the nearest possible orientation of the working body of the robot-manipulator, extracts from the list stored on the computing device a previously generated back reference trajectory, and the back reference trajectory describes the movement of the robot from the position in which the working body of the robot-manipulator is inside the area of interest, to initial position corresponding to the area of interest; determines, on the basis of information about the specified position and orientation of the working body, the cell of the area of interest into which it is necessary to move the working body of the robotic arm; defines a set of hinge angles at which the working body of the robotic arm is at a point of a given position with a given orientation; removes successively points from the beginning of the extracted reverse reference trajectory until the length of the trajectory is reduced by an amount equal to the size of the cell; adds to the end of the trajectory a point corresponding to the values of the set of angles of the robot-manipulator's hinges, at which the working body of the robot-manipulator is at a point of a given position with a given orientation, and to the beginning - a point corresponding to the current position of the working body of the robot-manipulator. Finally, the computing device 105 similarly recalculates the values of the velocities and accelerations of the angles of the joints at each point to ensure smooth movement of the robotic arm along the path. The trajectories obtained by the computing device 105 can then be sent to the robotic arm controller 103 to move the robotic arm in accordance with the obtained trajectory. Thus, due to the fact that in the presented solution, when planning the movement of the robot-manipulator, only the start and end points of the reference trajectories, previously generated for the robot-manipulator, are corrected, the time for planning the movement of the robot-manipulator is reduced, and the speed of its control is also increased.

In general terms (see Fig. 18), the computing device 1600 includes one or more processors 1601, memory means such as RAM 1602 and ROM 1603, input/output interfaces 1604, input/output devices 1605, and a device for networking 1606. The processor 1601 (or multiple processors, multi-core processor, etc.) may be selected from a variety of devices currently widely used, for example, such

-

Claims (7)

manufacturers like: Intel™, AMD™, Apple™, Samsung Exynos™, MediaTEK™, Qualcomm Snapdragon™, etc. Under the processor or one of the processors used in the device 1600, it is also necessary to take into account the graphics processor, for example, NVIDIA GPU or Graphcore, the type of which is also suitable for full or partial execution of the method, and can also be used to train and apply machine learning models in various information systems . RAM 1602 is a random access memory and is designed to store machine-readable instructions executable by the processor 1601 to perform the necessary logical data processing operations. RAM 1602 typically contains executable instructions for the operating system and associated software components (applications, program modules, and the like). In this case, the RAM 1602 may be the available memory of the graphics card or graphics processor. ROM 1603 is one or more persistent storage devices such as a hard disk drive (HDD), a solid state drive (SSD), flash memory (EEPROM, NAND, etc.), optical storage media (CD-R/RW, DVD-R/RW, BlueRay Disc, MD), etc. Various types of I/O interfaces 1604 are used to organize the operation of the components of the device 1600 and organize the operation of external connected devices. The choice of appropriate interfaces depends on the specific design of the computing device, which can be, but are not limited to: PCI, AGP, PS / 2, IrDa, FireWire , LPT, COM, SATA, IDE, Lightning, USB (2.0, 3.0, 3.1, micro, mini, type C), TRS/Audio jack (2.5, 3.5, 6.35), HDMI, DVI, VGA, Display Port, RJ45, RS232 etc. Various I/O information means 1605 are used to provide user interaction with the device 1600, for example, a keyboard, a display (monitor), a touch screen, a touchpad, a joystick, a mouse, a light pen, a stylus, a touchpad, a trackball, speakers, a microphone, augmented reality tools, optical sensors, tablet, light indicators, projector, camera, biometric identification tools (retinal scanner, fingerprint scanner, voice recognition module), etc. The networking tool 1606 provides data communication over an internal or external computer network, such as an Intranet, Internet, LAN, and the like. One or more means 1606 can be used, but not limited to: Ethernet card, GSM modem, GPRS modem, LTE modem, 5G modem, satellite communication module, NFC module, Bluetooth and/or BLE module, Wi-Fi module, etc. Additionally, satellite navigation tools in the device 1600 can also be used, such as GPS, GLONASS, BeiDou, Galileo. The specific choice of elements of the device 1600 for implementing various hardware and software architectural solutions may vary while maintaining the desired functionality provided. Modifications and improvements to the above described embodiments of the present technical solution will be clear to experts in this field of technology. The foregoing description is provided by way of example only and is not intended to be limiting in any way. Thus, the scope of the present technical solution is limited only by the scope of the appended claims. CLAIM 1. A method for planning the movement of a robotic arm, performed by at least one computing device, comprising the steps of: receive information about the specified position and orientation of the working body, in which it is necessary to move the working body of the robotic arm; determine the current position and orientation of the working body of the robotic arm; determining, on the basis of information about the predetermined position and orientation of the working body, a cell of the area of interest into which it is necessary to move the working body of the robotic arm, the area of interest being inside the working space of the robotic manipulator; on the basis of a set of possible orientations of the working body of the robotic arm when it is in the area of the said cell, the nearest possible orientation of the working body of the robotic manipulator to a given orientation is determined; Based on the data on the cell and the nearest possible orientation of the working body of the robot manipulator, a pre-generated direct reference trajectory is retrieved from the list stored on the computing device, moreover, the direct reference trajectory describes the movement of the robot manipulator from the initial position corresponding to the area of interest to the position in which the working body of the robot - the manipulator is inside the area of interest; determine a set of hinge angles at which the working body of the robotic arm is at a point of a given position with a given orientation; remove from the end of the extracted direct reference trajectory successively points until the length of the trajectory is reduced by an amount equal to the size of the cell; add to the end of the trajectory a point corresponding to the values of the set of angles of the robotic arm hinges, at which the working body of the robotic manipulator is at a point of a given position with a given orientation, and to the beginning - a point corresponding to the current position of the working body of the robotic manipulator; - 9 040444 move the working body of the robot-manipulator in accordance with the trajectory obtained at the previous stage. 2. The method according to claim 1, characterized in that it further comprises the steps of: determine that the point of a given position of the working body is inside the region of interest; determining the amount of deviation of the angles of the hinges of the robot-manipulator in the current position from the angles of the hinges specified for the initial position corresponding to the region of interest; the deviation value obtained at the previous stage is compared with the threshold value, and the step of determining the nearest possible orientation of the working body of the robot-manipulator to the given orientation is performed if the mentioned deviation value does not exceed the threshold value. 3. The method according to claim 1, characterized in that the determination of the current position and orientation of the working body of the robotic arm is carried out by solving a direct kinematics problem based on its current values of the angles of the hinges. 4. The method according to claim 1, characterized in that the stage at which sets of hinge angles are determined, at which the working body of the robotic arm is at a point of a given position with a given orientation, comprises stages at which: for a given position and orientation of the working body of the robot-manipulator, the inverse problem of kinematics is solved to obtain a set of sets of hinge angles at which the working body of the robotic manipulator is at a point of a given position with a given orientation; determine sets of hinge angles that satisfy collision constraints; choose as a set of hinge angles at which the working body of the robot-manipulator is at a point of a given position with a given orientation, from the set of hinge angles that satisfy collision restrictions, the set that is closest in terms of hinge angles to the last point of the extracted straight reference trajectory . 5. A system for planning the movement of a robotic arm, comprising: robotic arm; a computing device connected to the robotic arm; at least one memory containing machine-readable instructions, which, when executed by at least one computing device, performs the method according to any one of claims 1-4. 6. A method for planning the movement of a robotic arm, performed by at least one computing device, comprising the steps of: receive information about the specified position and orientation of the working body, to which it is necessary to move the working body of the robot-manipulator; determine the current position and orientation of the working body of the robotic arm; determining, based on information about the current position and orientation of the working body, a cell of the area of interest from which it is necessary to move the working body of the robot manipulator, the area of interest being inside the working space of the robot manipulator; based on a set of possible orientations of the working body of the robotic arm when it is in the area of the said cell, the closest possible orientation of the working body of the robotic manipulator to the current orientation is determined; Based on the data on the cell and the nearest possible orientation of the working body of the robotic arm, a pre-generated reverse reference trajectory is retrieved from the list stored on the computing device, and the reverse reference trajectory describes the movement of the robot from the position in which the working member of the robotic manipulator is inside the area of interest, to the initial position , corresponding to the area of interest; determine the set of hinge angles at which the working body of the robot-manipulator is at a point of a given position with a given orientation; points are removed from the beginning of the retrieved reverse reference trajectory in succession until the length of the trajectory is reduced by an amount equal to the size of the cell; add to the end of the trajectory a point corresponding to the values of the set of angles of the robotic arm hinges, at which the working body of the robotic manipulator is at a point of a given position with a given orientation, and to the beginning - a point corresponding to the current position of the working body of the robotic manipulator; move the working body of the robot-manipulator in accordance with the trajectory obtained at the previous stage. 7. The method according to claim 6, characterized in that it further comprises the steps of: determine the amount of deviation of the specified position and the initial position of the working body of the robotic arm corresponding to the area of interest; the deviation value obtained at the previous stage is compared with the threshold value, and the step of determining the nearest possible orientation of the working body of the robot-manipulator to the current orientation is performed if the said deviation value does not exceed the threshold value. -