WO2018129705A1

WO2018129705A1 - Method and apparatus for use in determining inverse solution for robots in series connection

Info

Publication number: WO2018129705A1
Application number: PCT/CN2017/071083
Authority: WO
Inventors: 任岭雪; 王磊; 熊璟
Original assignee: 中国科学院深圳先进技术研究院
Priority date: 2017-01-13
Filing date: 2017-01-13
Publication date: 2018-07-19

Abstract

A method and an apparatus for use in determining an inverse solution for robots in series connection. The method comprises: acquiring a target coordinate of a terminal joint of a robot in series connection, wherein at least one front end joint is connected in series above the terminal joint (110); randomly generating a plurality of random joint variable sets, wherein each random joint variable set comprises therein a joint variable which corresponds to each front end joint (120) respectively; calculating random coordinates of the terminal joint according to each random joint variable set (130); selecting an approximate coordinate of the terminal joint from the random coordinates according to the target coordinate (140); and determining an inverse solution of the target coordinate according to a random joint variable set which corresponds to the approximate coordinate (150); the approximate coordinate is a random coordinate wherein an error of the random coordinate and the target coordinate is within an error threshold.

Description

Method and device for determining reverse solution result of serial robot

Technical field

Embodiments of the present disclosure relate to robot control techniques, for example, to a method and apparatus for determining a reverse solution result of a tandem robot.

Background technique

Bionic robots have been applied to many fields. Bionic robots can replace or partially replace people in difficult environments, such as earthquake disasters or the medical industry. In the medical industry, robots can be made small enough to enter spaces that cannot be reached by human hands. The natural cavity of the human body (such as the esophagus) is completed or administered at a fixed point. However, some environments do not allow the robot to operate completely autonomously, such as a medical operating environment or a very complex earthquake-stricken environment. It requires experienced people to manipulate these robots to perform the corresponding tasks. The master-slave control system can be used to control the robot.

In the master-slave control system, it is generally necessary to have a master operator, a slave operator, and a control system. The main operator is generally operated by a person, and the operator is generally a tandem robot. In the control system of the master-slave robot, the real-time requirements between the master and the slave are high. For high real-time performance, the response time of the motor and controller in the control system of the master-slave robot is short, and the communication time between the master and slave robots is very short. It is easy to meet the real-time requirements. The difficulty is that the calculation result of the inverse solution of the tandem robot is longer. .

In the related art, for the tandem robot, the master-slave mode can be used to perform the master-slave operation, and the master-slave heterogeneous robot mostly uses the Jacobian algorithm to control the speed of the master robot in the three-dimensional space. Kinematic modeling, calculating the Jacobian matrix from the robot, inverts the Jacobian matrix. The Jacobian matrix control algorithm is suitable for the case where the degree of freedom of the robot (the number of joints of the tandem robot) can be reduced to 3 or 6, but the cumulative error of the displacement is large, and the Jacobian matrix will be invalid at the singular position, thus for high redundancy. The tandem robot (12 degrees of freedom or more), this method does not apply. In addition, the cyclic coordinate descent method, the Newton method, the quasi-Newton method in the numerical method, or the combination of two or two of these three, neural network, etc., although they can be used, the slow speed of these algorithms is limited, and there are limitations. It is expensive to calculate and store, so it is not practical for serial robots.

Summary of the invention

Embodiments of the present disclosure provide a method and apparatus for determining a reverse solution result of a serial robot, which can be fast The result of the inverse solution of the tandem robot is accurately obtained.

In a first aspect, an embodiment of the present disclosure provides a method for determining a reverse solution result of a serial robot, including:

Obtaining target coordinates of the end joint of the tandem robot, wherein at least one front end joint is connected in series above the end joint;

Randomly generating a plurality of random joint variable groups, wherein each random joint variable group includes a joint variable corresponding to each of the front end joints;

Calculating random coordinates of the end joint according to each of the random joint variable groups;

Selecting approximate coordinates of the end joints in the random coordinates according to the target coordinates;

Determining an inverse solution result of the target coordinate according to a random joint variable group corresponding to the approximate coordinate;

The approximate coordinate is a random coordinate in which the error of the random coordinate and the target coordinate is within an error threshold range.

In a second aspect, an embodiment of the present disclosure further provides an apparatus for determining a result of a reverse solution of a serial robot, including:

a target coordinate acquiring module, configured to acquire target coordinates of an end joint of the tandem robot, wherein at least one front end joint is connected in series above the end joint;

a random joint variable group generating module is connected to the target coordinate acquiring module, and is configured to randomly generate a plurality of random joint variable groups, wherein each random joint variable group includes a joint variable corresponding to each of the front end joints respectively;

a random coordinate calculation module, connected to the random joint variable group generating module, configured to calculate a random coordinate of the end joint according to each of the random joint variable groups;

An approximate coordinate selection module, coupled to the random coordinate calculation module, configured to select an approximate coordinate of the end joint in the random coordinate according to the target coordinate;

The inverse solution result determining module is connected to the approximate coordinate selecting module, and is configured to determine an inverse solution result of the target coordinate according to the random joint variable group corresponding to the approximate coordinate;

Wherein the approximate coordinate is an error of the random coordinate and the target coordinate in an error threshold Random coordinates within the perimeter.

The present disclosure also provides a non-transitory computer readable storage medium storing computer executable instructions arranged to perform the methods described above.

The present disclosure also provides a tandem robot comprising:

At least one processor;

a memory communicatively coupled to the at least one processor; wherein

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the method described above.

In the embodiment of the present disclosure, by randomly generating a plurality of random joint variable groups, the random coordinates of the end joints are calculated, and the approximate coordinates of the end joints are selected in the random coordinates to determine the inverse solution result of the target coordinates. The method in the embodiment of the present disclosure can be applied to a series robot with arbitrary degrees of freedom, and the result of inverting the solution is fast, the precision of the inverse solution is high, the overhead on calculation and storage is small, and the cumulative error is small.

DRAWINGS

1 is a flowchart of a method for determining a reverse solution result of a serial robot according to Embodiment 1 of the present disclosure;

2 is a flow chart of a method for determining a reverse solution result of a serial robot according to Embodiment 2 of the present disclosure;

3 is a flowchart of a method for determining a reverse solution result of a serial robot according to Embodiment 3 of the present disclosure;

4 is a flow chart of a method for determining a reverse solution result of a serial robot according to Embodiment 4 of the present disclosure;

5 is a schematic structural diagram of an apparatus for determining a reverse solution result of a serial robot according to Embodiment 5 of the present disclosure;

6 is a schematic diagram of a closed-loop control system of a master-slave robot for implementing a method for determining a result of a reverse solution of a serial robot according to Embodiment 6 of the present disclosure;

7 is a schematic diagram of a joint coordinate system in a method for determining a reverse solution result of a tandem robot according to Embodiment 6 of the present disclosure;

FIG. 8 is a schematic structural diagram of an electronic device according to Embodiment 8 of the present disclosure.

detailed description

The present disclosure will be described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the disclosure and are not intended to be limiting. In addition, it should be noted that, for the convenience of description, only some but not all of the structures related to the present disclosure are shown in the drawings. The features of the following embodiments and embodiments may be arbitrarily combined with each other without conflict.

Embodiment 1

1 is a flowchart of a method for determining an inverse solution result of a tandem robot according to Embodiment 1 of the present disclosure. This embodiment is applicable to a case where a joint robot coordinates are known to solve each joint variable, and the method can determine a tandem robot. The device that performs the inverse solution is implemented by software, hardware, or software and hardware. The device can be integrated into a device that provides storage and computing functions, such as a typical personal computer (Personal Computer, PC). )Wait.

In step 110, target coordinates of the end joints of the tandem robot are acquired, wherein at least one front end joint is connected in series above the end joints. The tandem robot is connected in series with a plurality of joints, and the end joint is an end joint, and the joint above the end joint can be referred to as a front joint.

When controlling the robot, the control result can be expressed by the position and posture of the end joint of the tandem robot, but in order to enable the end joint to reach a preset position and posture, each front joint needs to cooperate with each other.

In this embodiment, the target coordinates of the end joint may be represented by coordinates of a position that the desired end joint can reach and coordinates of the posture. After acquiring the target coordinates of the end joints of the tandem robot, the joint variables of each front joint (the joint variables may be positions and postures) are calculated by the following steps, and the end joints can be adjusted above the target coordinates.

In step 120, a plurality of random joint variable groups are randomly generated, wherein each random joint variable group includes joint variables respectively corresponding to each of the front end joints.

Each joint has a corresponding joint variable. The joint variable can indicate the motion of the joint. The motion of all joints of the entire tandem robot can be represented by a joint variable group, and the joint variable group includes joint variables of each front joint. When a plurality of random joint variable groups are randomly generated, the position and posture of the corresponding end are also random, and the position and posture of the obtained end may be ideal, or may be unsatisfactory, and may be ideal after screening. The position and posture. The random joint variable group can be generated by a random function, and the random joint variable can be limited to a certain range according to the range of motion of the joint. For example, when the joint variable is a joint angle, a plurality of random joint angles can be generated by the formula θ _n = θ _Min-n + θ _Max-n × Random(), and Random() is a random function for generating a random number. . θ _n is the joint angle of the nth joint, θ _Min-n is the minimum joint angle of the nth joint, θ _Max-n is the maximum joint angle of the nth joint, and constitutes a random joint angle group (θ ₁ , θ ₂ ,...,θ _N ),n∈[1,N], N is the total number of joint angles included in the random joint variable, and N is consistent with the total number of front joints.

The number of groups of random joint variable groups generated can be set according to the accuracy of the solution, and the number of groups of random joint variable groups can be increased, which can improve the accuracy of the results.

When the number of joints of the tandem robot is increased by one, only one joint variable is added to the generated random joint variable group, and the computational complexity is linearly increased. Therefore, the present embodiment is applicable to a tandem robot of any joint number (degree of freedom).

In step 130, random coordinates of the end joint are calculated based on each of the random joint variable sets.

When the joint variable of each front joint is determined, the position and posture of the end joint are determined, that is, the coordinates of the end joint are determined. For example, a base coordinate system and a joint coordinate system can be established, and a transformation matrix between each two coordinate systems and a coordinate transformation matrix of the end joint to the base coordinate system are determined by the connection relationship between the joints, by substituting each random joint variable group into Go to the coordinate transformation matrix and get the random coordinates of the corresponding multiple end joints.

In step 140, approximate coordinates of the end joint are selected in the random coordinates according to the target coordinates.

Because there are multiple sets of random joint variable groups, there are multiple random coordinates of the corresponding end joints. The difference between the random coordinates and the target coordinates is large or small, the random joint variable group is randomly generated, and the probability that the random coordinates are the same as the target coordinates is small. The error threshold can be set, and the random coordinates within the error threshold range are used as the approximate coordinates of the end joint, which is convenient for screening.

Since the error threshold can be set artificially, the error of the result can be controlled. The results in this example were tried by random experiments, not by calculations, so there is no cumulative error.

The number of approximate coordinates selected may be one or more. Wherein, a random coordinate with the smallest error value between the target coordinates may be used as the approximate coordinate; or a plurality of random coordinates that satisfy the preset error condition with the error value between the target coordinates may be used as the approximate coordinate.

In step 150, an inverse solution result of the target coordinates is determined according to a random joint variable group corresponding to the approximate coordinates.

The approximate coordinate is a random coordinate in which the error of the random coordinate and the target coordinate is within an error threshold range. The approximate coordinates of the end joints have corresponding random joint variable groups. When each joint moves according to the joint variables in the random joint variable group, the end joints can be placed at positions corresponding to the approximate coordinates close to the target coordinates. Therefore, the random joint variable group corresponding to the approximate coordinates can be used as the inverse solution result of the target coordinates. The genetic algorithm can be used to take into account the principle of "multiple moving small joints, less moving large joints" and "the joint angle of each tandem robot is smaller than the last moving angle", and the inverse solution result of the target coordinates is determined. Wherein the large joint is the front joint and the small joint is the end joint.

In this embodiment, by randomly generating a plurality of random joint variable groups, the random coordinates of the end joints are calculated, and the approximate coordinates of the end joints are selected in the random coordinates to determine the inverse solution result of the target coordinates. The method in this embodiment is applicable to a series robot with arbitrary degrees of freedom, and the result of inverting the solution is fast, the precision of the inverse solution is high, the overhead on calculation and storage is small, and there is no accumulated error.

Embodiment 2

FIG. 2 is a flowchart of a method for determining a reverse solution result of a serial robot according to Embodiment 2 of the present disclosure, and Embodiment 2 is based on the foregoing embodiment.

The method for determining the inverse solution result of the serial robot provided by this embodiment includes steps 210 to 280. The

steps

210, 220, 270, and 280 are the same as the

steps

110, 120, 140, and 150 in the first embodiment, respectively.

In step 210, target coordinates of the end joints of the tandem robot are acquired, wherein at least one front end joint is connected in series above the end joints.

In step 220, a plurality of random joint variable groups are randomly generated, wherein each random joint variable group includes joint variables respectively corresponding to each of the front end joints.

In step 230, a base coordinate system of the tandem robot and a joint coordinate system for each joint are established.

For example, according to the link parameters of the space robot, the DH modeling method can be applied to establish the base coordinate system O ₀ (x ₀ , y ₀ , z ₀ ) of the tandem robot, and the joint coordinate system O _i (x _i , y of each joint). _i , z _i ), where i ∈ [1, n], n is the total number of joints included in the tandem robot. The joint variable in each joint coordinate system represents the motion of the joint. The joint variables in the joint coordinate system can be uniformly converted into the base coordinate system and represented by the coordinates in the base coordinate system.

In step 240, a transformation matrix between adjacent two coordinate systems is determined based on the base coordinate system and the joint coordinate system.

The connection relationship between joints determines the transformation matrix between two adjacent coordinate systems

The variables of the joint coordinate system can be represented by the adjacent joint coordinate system, and the joint coordinate system can be converted into a base coordinate system after several transformations.

In step 250, a transformation matrix of the end joints in the base coordinate system is calculated as the preset coordinate transformation matrix.

Each set of random joint variable groups corresponds to the random coordinates of the end joints. This correspondence relationship may be the transformation matrix of the end joints in the base coordinate system. After the tandem robot is determined, the connection relationship between each joint is also determined. The preset coordinate transformation matrix is determined. The coordinate transformation matrix of the coordinate system of the end joint to the base coordinate system O ₀ (x ₀ , y ₀ , z ₀ )

Can be transformed matrix between two adjacent coordinate systems

Multiply item by item,

Wherein, d _1x , d _1y , d _1z , d _2x , d _2y , d _2z , d _3x , d _{3y ,} and d _{3z are} used to represent the posture of the end joint, and p _x , p _{y ,} and p _{z are} used to represent the end joint. position.

In step 260, each random joint variable group is substituted into a preset coordinate transformation matrix to obtain a plurality of random coordinates corresponding to the end joint.

Because of the preset coordinate transformation matrix

Can be transformed matrix between two adjacent coordinate systems

By item by right multiplication, so d _1x , d _1y , d _1z , d _2x , d _2y , d _2z , d _3x , d _3y , d _3z , p _x , p _{y ,} and p _{z in the} preset coordinate transformation matrix The value is a function of each joint variable, and each random joint variable group is substituted into a preset coordinate transformation matrix to obtain a plurality of random coordinates corresponding to the end joint. The random joint variable group represents the motion of each joint, and the preset coordinate transformation matrix represents the connection relationship between the joints, and the combination of the two can obtain the corresponding random coordinates.

In step 270, approximate coordinates of the end joint are selected in the random coordinates according to the target coordinates.

In step 280, an inverse solution result of the target coordinates is determined according to a random joint variable group corresponding to the approximate coordinates.

In this embodiment, by substituting each random joint variable group into a preset coordinate transformation matrix, a plurality of random coordinates corresponding to the end joint are obtained, and the difference between the random coordinates and the target coordinates is verified. Whether the corresponding random joint variable group can be used as an inverse solution simplifies the process of inverting the solution of the tandem robot.

Embodiment 3

FIG. 3 is a flowchart of a method for determining a reverse solution result of a serial robot according to Embodiment 3 of the present disclosure, and Embodiment 3 is based on the foregoing embodiment.

The method for determining the reverse solution result of the serial robot provided by this embodiment includes steps 310 to 360. Steps 310-330 and step 360 are the same as steps 110-130 and 150 in the first embodiment, respectively.

In step 310, target coordinates of the end joints of the tandem robot are acquired, wherein at least one front end joint is connected in series above the end joints.

In step 320, a plurality of random joint variable groups are randomly generated, wherein each random joint variable group includes joint variables respectively corresponding to each of the front end joints.

In step 330, random coordinates of the end joint are calculated based on each of the random joint variable groups.

In step 340, an error value between the target coordinates of the end joint and the random coordinates of the end joint is calculated.

There are multiple random coordinates of the end joints, and the difference between them and the target coordinates of the end joints can be represented by an error value. The coordinates of the end joint contain the position coordinates and attitude coordinates of the end joint, so the error values include position error and attitude error. Since the coordinates of the end joint are a function of the joint variable and the target coordinates of the end joint are known quantities, the error value is a function of the random joint variable.

Optionally, calculating an error value between a target coordinate of the end joint and a random coordinate of the end joint, including:

According to the formula ΔE=ΔP+ΔR and

An error value ΔE between the target coordinates of the end joint and the random coordinates of the end joint is calculated.

Where ΔP represents the position difference, ΔR represents the attitude difference, ω _ρ and ω _γ are the weights of the position difference ΔP and the attitude difference ΔR, P _s is the target position of the end joint, and p _sj is the target position component of the end joint , d _sj is the target pose of the end joint, P _s , p _sj and d _sj can be obtained from the target coordinates of the end joint, P _r is the random position of the end joint, P _rj is the random position component of the end joint, d _rj For the random pose of the end joint, P _r , p _rj and d _rj can be obtained from the random coordinates of the end joint, and ||P _s -P _r || is the calculation of the target position of the end joint and the random position of the end joint. The distance between them.

Among them, the appropriate ω _γ can be selected to make the attitude error and the position error converge synchronously. For example, ω _γ can be greater than 10 ⁴ . If the motion state of the robot requires higher accuracy for the end position, the value of ω _γ can be 0; if the position and attitude need to satisfy a certain precision, the value of ω _γ can be 10 ⁶ .

In step 350, the random coordinates of the end joint whose error value is less than the error threshold are obtained as approximate coordinates of the end joint.

The error value of the random coordinates of the end joint and the target coordinate is smaller than the error threshold, indicating that the random coordinates of the end joint are very similar to the target coordinates, and the random coordinates of the end joint can be used as the approximate coordinates of the end joint.

In step 360, an inverse solution result of the target coordinates is determined based on a random joint variable group corresponding to the approximate coordinates.

In this embodiment, by calculating the error value between the target coordinates of the end joint and the random coordinates of the end joint, it can be known whether the random joint variable group corresponding to the random coordinates of the end joint can be used as an inverse solution, and the inverse solution result can be quickly screened out.

Embodiment 4

FIG. 4 is a flowchart of a method for determining a reverse solution result of a serial robot according to Embodiment 4 of the present disclosure, and Embodiment 4 is based on the foregoing embodiment.

The method for determining the inverse solution result of the serial robot provided by this embodiment includes step 410 to step 4110. The steps 410-440 are the same as the steps 110-140 in the first embodiment.

In step 410, the target coordinates of the end joints of the tandem robot are acquired, wherein at least one front end joint is connected in series above the end joint.

In step 420, a plurality of random joint variable groups are randomly generated, wherein each random joint variable group includes joint variables respectively corresponding to each of the front end joints.

In step 430, random coordinates of the end joint are calculated based on each of the random joint variable sets.

In step 440, the end joint is selected in the random coordinates according to the target coordinates Approximate coordinates.

In step 450, a random joint variable group corresponding to at least two of the approximate coordinates is acquired as a candidate random joint variable group.

There are multiple approximate joint coordinates of the end joint satisfying the error threshold, and there are also multiple random joint variable groups corresponding to them. These random joint variable groups can be used as the inverse solution results of the target coordinates, but they are not necessarily the optimal inverse solution results. The optimal inverse solution result can be selected from the group of random joint variables to be selected.

Step 460 to step 4120 are processes for selecting an optimal inverse solution result from the group of random joint variables to be selected. The smaller the amount of movement of the joint is, the less energy is consumed. Therefore, the group of random joints to be selected with a small amount of joint movement is used as the optimal inverse solution. The tandem robot has many joints. When the joint movement amount is the same, the joint away from the end joint consumes more energy than the joint near the end joint, so it is possible to give priority to the distal joint when searching for the optimal inverse solution result. The amount of joint movement is small. In summary, the minimum amount of joint movement can be found one by one from the joint farthest from the end joint and toward the end joint.

In step 460, a front end joint that is furthest from the end joint is selected as the current alignment joint.

A method for selecting the optimal inverse solution result from the selected random joint variable group is to select the candidate random joint variable group with the smallest joint movement amount as the optimal inverse solution result. Since the tandem robot has multiple joints, the minimum amount of joint movement can be found one by one. When the amount of joint movement is the same, the energy of the joint away from the end is much more consumed than the joint near the end, and it is possible to preferentially ensure that the amount of joint movement away from the end joint is small. So at the beginning you can select the front joint that is furthest from the end joint as the current comparison joint.

In step 470, joint variables corresponding to the current comparison joint are acquired in each of the selected random joint variable groups.

The joint variable of each joint is included in the group of random joint variables to be selected. Each candidate random joint variable group corresponds to a set of joint variables. Because it is a joint-by-joint comparison, the joint variable of the current alignment joint can be obtained for each comparison.

In step 480, a joint movement amount of the current comparison joint corresponding to each of the candidate random joint variable groups is determined according to the joint variable.

The smaller the amount of joint movement, the less energy is consumed, and the joint of the current joint is determined according to the joint variable. The amount of movement.

In step 490, the joint movement amount of at least two of the current comparison joints is compared, and the candidate random joint variable group corresponding to the minimum joint movement amount of the current comparison joint is taken as the preferred random joint variable group.

According to the current comparison of the joint movement amount of the joint, the random joint variable group is selected for screening, and the minimum joint movement amount of the current joint is found, and the candidate random joint corresponding to the joint movement amount of the smallest current comparison joint is selected. The variable group is the preferred random joint variable group. When the preferred random joint variable group is a group, the preferred random joint variable group is the current optimal inverse solution result. Optimization can continue when there are more than one set of preferred random joint variable groups.

In step 4100, it is determined whether the number of preferred random joint variable groups is greater than 1. If the number of preferred random joint variable groups is greater than 1, step 4110 is performed, and if the number of preferred random joint variable groups is 1, then step 4120 is performed.

In step 4110, the front end joint adjacent to the current alignment joint is selected as the current alignment joint toward the end joint, and the flow returns to step 470.

If the number of preferred random joint variable groups is greater than 1, indicating that the optimal inverse solution result is not found, then the front end joint adjacent to the current alignment joint is selected as the current comparison joint toward the end joint, and the smallest current ratio is selected. The candidate random joint variable group corresponding to the joint movement amount of the joint is selected as the preferred random joint variable group. When the number of joints of the tandem robot is large, the new current alignment joint can also select the second or third front joint adjacent to the current alignment joint near the end joint.

In step 4120, the preferred set of random joint variables is taken as the inverse of the target coordinates.

Preferably, the error between the random coordinates of the end joint corresponding to the random joint variable group and the target coordinate is within the error threshold range, and preferably the energy consumption of the joint movement of the random joint variable group is the smallest, which can be used as the inverse solution result of the target coordinate.

In this embodiment, starting from the front end joint farthest from the end joint, toward the end joint, the group of random joint variables to be selected corresponding to the minimum joint movement amount is the inverse of the least energy consumed when the tandem robot moves the joint. Solve the result.

Embodiment 5

FIG. 5 is a schematic structural diagram of an apparatus for determining a reverse solution result of a serial robot according to Embodiment 5 of the present disclosure. It is intended that the apparatus can perform the method of determining the inverse solution result of the tandem robot in the above embodiment. The apparatus includes: a target coordinate acquisition module 501, a random joint variable group generation module 502, a random coordinate calculation module 503, an approximate coordinate selection module 504, and an inverse solution result determination module 505.

The target coordinate acquisition module 501 is configured to acquire target coordinates of the end joints of the tandem robot, wherein at least one front end joint is connected in series above the end joint.

The random joint variable group generating module 502 is connected to the target coordinate acquiring module 501 and configured to randomly generate a plurality of random joint variable groups, wherein each random joint variable group includes a joint corresponding to each of the front end joints respectively variable.

The random coordinate calculation module 503 is connected to the random joint variable group generation module 502, and is configured to calculate the random coordinates of the end joint according to each of the random joint variable groups.

The approximate coordinate selection module 504 is connected to the random coordinate calculation module 503, and is configured to select an approximate coordinate of the end joint in the random coordinate according to the target coordinate.

The inverse solution result determining module 505 is connected to the approximate coordinate selecting module 504, and is configured to determine an inverse solution result of the target coordinate according to the random joint variable group corresponding to the approximate coordinate.

Optionally, the random coordinate calculation module includes a coordinate transformation matrix unit configured to substitute each random joint variable group into a preset coordinate transformation matrix to obtain a plurality of random coordinates corresponding to the end joint.

Optionally, the random coordinate calculation module further includes a coordinate system establishing unit, a transformation matrix determining unit, and a coordinate transformation matrix calculating unit.

a coordinate system establishing unit is configured to establish a base coordinate system O ₀ (x ₀ , y ₀ , z ₀ ) of the tandem robot, and a joint coordinate system O _i (x _i , y _i , z _i ) of each joint, wherein i ∈ [1, n], n is the total number of joints included in the tandem robot.

a transformation matrix determining unit, connected to the coordinate system establishing unit, configured to determine a transformation matrix between two adjacent coordinate systems according to the base coordinate system and the joint coordinate system

a coordinate transformation matrix calculation unit respectively connected to the transformation matrix determination unit and the coordinate transformation matrix unit, and configured to be according to a formula

Calculating a transformation matrix of the end joint in the base coordinate system O ₀ (x ₀ , y ₀ , z ₀ )

Will be said

As the preset coordinate transformation matrix.

Wherein, d _1x , d _1y , d _1z , d _2x , d _2y , d _2z , d _3x , d _{3y ,} and d _{3z are} used to represent the posture of the end joint, and p _x , p _{y ,} and p _{z are} used to represent the end joint. a position, the values of d _1x , d _1y , d _1z , d _2x , d _2y , d _2z , d _3x , d _3y , d _3z , p _x , p _{y ,} and p _z are determined by each of the front ends The joint variables corresponding to the joints are determined.

In this embodiment, by substituting each random joint variable group into a preset coordinate transformation matrix, a plurality of random coordinates corresponding to the end joint are obtained, and the corresponding random joint can be determined by checking the difference between the random coordinates and the target coordinates. Whether the variable group can be used as an inverse solution simplifies the process of inverting the solution of the tandem robot.

Optionally, the approximate coordinate selection module includes an error value calculation unit and an error value determination unit.

The error value calculation unit is configured to calculate an error value between the target coordinates of the end joint and the random coordinates of the end joint.

The error value judging unit is connected to the error value calculating unit and configured to acquire the random coordinates of the end joint whose error value is smaller than the error threshold as the approximate coordinates of the end joint.

Optionally, the inverse solution result determining module includes a random joint variable group acquiring unit to be selected, a current matching joint selecting unit, a joint variable acquiring unit, a movement amount determining unit, a screening random joint variable group unit, and a reduced random joint variable group quantity value. Unit and inverse solution result acquisition unit.

The candidate random joint variable group obtaining unit is configured to acquire a random joint variable group respectively corresponding to at least two of the approximate coordinates as a candidate random joint variable group.

The current comparison joint selection unit is connected to the candidate random joint variable group acquisition unit, and is configured to select a front end joint farthest from the end joint as the current comparison joint.

The joint variable acquiring unit is connected to the current matching joint selecting unit, and is configured to acquire a joint variable corresponding to the current matching joint in each of the selected random joint variable groups.

The movement amount determining unit is connected to the joint variable acquiring unit, and is configured to determine, according to the joint variable, a joint movement amount of the current comparison joint corresponding to each of the candidate random joint variable groups.

The random joint variable group unit is screened, and is connected to the movement amount determining unit, and the candidate random joint variable group corresponding to the minimum joint movement amount of the current comparison joint is used as the preferred random joint variable group.

Reducing the random joint variable group number unit, and connecting to the screening random joint variable group unit, and setting, if the number of preferred random joint variable groups is greater than 1, selecting the current comparison with the direction toward the end joint The front end joint adjacent to the joint serves as a current alignment joint, and returns an operation of acquiring a joint variable corresponding to the current comparison joint in each of the selected random joint variable groups.

Retaining a random joint variable group unit, and connecting the reduced random joint variable group number value unit, and setting the preferred random joint variable group as the inverse of the target coordinate if the number of preferred random joint variable groups is 1. result.

In this embodiment, starting from the front end joint farthest from the end joint, toward the direction of the end joint, the group of random joint variables to be selected corresponding to the minimum amount of joint movement is taken as the opposite of the minimum energy consumed when the tandem robot moves the joint. Solve the result.

Optionally, the error value calculation unit is set according to the formula ΔE=ΔP+ΔR and

The apparatus for determining the inverse solution result of the serial robot provided by the embodiment of the present disclosure may be used to perform the method for determining the inverse solution result of the serial robot provided by any embodiment of the present disclosure, and has the corresponding functions and beneficial effects of performing the method.

Embodiment 6

6 is a schematic diagram of a closed-loop control system of a master-slave robot for implementing a method for determining a result of a reverse solution of a tandem robot according to a sixth embodiment of the present disclosure, and FIG. 7 is a method for determining an inverse solution result of a tandem robot according to a sixth embodiment of the present disclosure. Schematic diagram of the joint coordinate system.

A system for determining the inverse solution result of the tandem robot includes: an operator 601, a master robot 602, a PC 603, a controller 604, a slave robot 605, and an optical tracker 606, as shown in FIG. The master robot 602 communicates with the PC 603 via a Transmission Control Protocol/Internet Protocol (TCP/IP) protocol, and the controller 604 communicates with the PC 603 via Ethernet.

Figure 7 shows three joints of the slave robot, including a first joint 701, a second joint 702, and an end joint 703, and a base coordinate system is established on the first joint, and an intermediate coordinate system is established on the second joint 702. An end coordinate system is established on the end joint 703. It will be understood by those skilled in the art that the number of joints from the robot can be any other value.

The operator 601 operates the master robot 602, and transmits the motion command of the master robot 601 to the slave robot 605 via the PC 603 and the controller 604. The optical tracker 606 feeds the operator 601 back to the current motion state and position state of the slave robot 605.

The motion command received by the PC 603 from the master robot 602 is the target coordinate of the end joint 703 from the robot 605, and the PC 603 applies the method of determining the inverse solution result of the tandem robot provided by the above-described embodiment of the present disclosure to obtain the first slave robot 605. The joint variables of the joint 701 and the second joint 702, the controller 604 controls the movement of each joint from the robot 605 according to the joint variable such that the end joint is at the target coordinates.

The present embodiment provides a master-slave robot system and a joint structure of the slave robot using a method for determining the inverse solution result of the tandem robot, so that the PC can quickly and accurately obtain the target coordinates of the end joint of the robot according to the master robot. The joint variable from each joint of the robot is obtained, and the controller controls the movement of the slave robot in real time according to the joint variable.

Example 7

The present disclosure also provides a non-transitory computer readable storage medium storing computer executable instructions arranged to perform the method of any of the above embodiments.

Example eight

The present disclosure also provides a hardware structure diagram of an electronic device. Referring to FIG. 8, the electronic device includes:

At least one processor 80, which is exemplified by a processor 80 in FIG. 8; and a memory 81, may further include a communication interface 82 and a bus 83. The processor 80, the communication interface 82, and the memory 81 can complete communication with each other through the bus 83. Communication interface 82 can transmit information. The processor 80 can call the logic instructions in the memory 81 to perform a method of determining the result of the tandem robot inverse solution.

In addition, the logic instructions in the memory 81 described above may be implemented in the form of a software functional unit and sold or used as a stand-alone product, and may be stored in a computer readable storage medium.

The memory 81 is a computer readable storage medium, and can store a software program, a computer executable program, a program instruction or a module corresponding to the method in the embodiment of the present disclosure. The processor 80 executes the functional application and the data processing by running a software program, an instruction or a module stored in the memory 81, that is, a method of determining the inverse solution result of the tandem robot.

The memory 81 may include a storage program area and an storage data area, wherein the storage program area may store an operating system, an application required for at least one function; the storage data area may store data created according to use of the terminal device, and the like. Further, the memory 81 may include a high speed random access memory, and may also include a nonvolatile memory.

The technical solution of the present disclosure may be embodied in the form of a software product stored in a storage medium, including one or more instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) Performing all or part of the steps of the method of the embodiments of the present disclosure. The foregoing storage medium may be a non-transitory storage medium, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. A medium that can store program code, or a transitory storage medium.

Industrial applicability

The method and device for determining the inverse solution result of the serial robot provided by the embodiment of the present disclosure can be quickly quasi-standard Calculate the result of the inverse solution.

Claims

A method for determining the inverse solution result of a tandem robot includes:

Obtaining target coordinates of the end joint of the tandem robot, wherein at least one front end joint is connected in series above the end joint;

Randomly generating a plurality of random joint variable groups, wherein each random joint variable group includes a joint variable corresponding to each of the front end joints;

Calculating random coordinates of the end joint according to each of the random joint variable groups;

Selecting approximate coordinates of the end joints in the random coordinates according to the target coordinates;

Determining an inverse solution result of the target coordinate according to a random joint variable group corresponding to the approximate coordinate;

The approximate coordinate is a random coordinate in which the error of the random coordinate and the target coordinate is within an error threshold range.
The method of claim 1, wherein calculating the random coordinates of the end joint according to each of the random joint variable sets comprises:

Each random joint variable group is substituted into a preset coordinate transformation matrix to obtain a plurality of random coordinates corresponding to the end joint.
The method according to claim 2, before substituting each set of random joint variables into a preset coordinate transformation matrix to obtain a plurality of random coordinates corresponding to the end joints, the method further comprises:

Establishing a base coordinate system O 0 (x 0 , y 0 , z 0 ) of the tandem robot and a joint coordinate system O i (x i , y i , z i ) of each joint, where i∈[1,n ], n is the total number of joints included in the tandem robot;

Determining a transformation matrix between two adjacent coordinate systems according to the base coordinate system and the joint coordinate system
as well as

According to the formula
Calculating a transformation matrix of the end joint in the base coordinate system O 0 (x 0 , y 0 , z 0 )
Will be said
As the preset coordinate transformation matrix;

Wherein, d 1x , d 1y , d 1z , d 2x , d 2y , d 2z , d 3x , d 3y , and d 3z are used to represent the posture of the end joint, and p x , p y , and p z are used to represent the end joint. position;

The values of d 1x , d 1y , d 1z , d 2x , d 2y , d 2z , d 3x , d 3y , d 3z , p x , p y , and p z are determined by each of the front joints The corresponding joint variable is determined.
The method according to claim 3, wherein the approximate coordinates of the end joint are selected in the random coordinates according to the target coordinates, including:

Calculating an error value between a target coordinate of the end joint and a random coordinate of the end joint;

Obtaining a random coordinate of the end joint whose error value is smaller than the error threshold as an approximate coordinate of the end joint.
The method according to claim 4, wherein determining an inverse solution result of said target coordinates based on a random joint variable group corresponding to said approximate coordinates comprises:

Obtaining a random joint variable group respectively corresponding to at least two of the approximate coordinates as a candidate random joint variable group;

Selecting a front end joint that is furthest from the end joint as the current alignment joint;

Obtaining a joint variable corresponding to the current comparison joint in each of the selected random joint variable groups;

Determining, according to the joint variable, a joint movement amount of a current comparison joint corresponding to each of the candidate random joint variable groups;

a set of candidate random joint variables corresponding to the minimum amount of joint movement of the current comparison joint is taken as a preferred random joint variable group; if the number of preferred random joint variable groups is greater than 1, toward a direction close to the end joint, Selecting a front end joint adjacent to the current comparison joint as a current comparison joint, and returning to perform an operation of acquiring a joint variable corresponding to the current comparison joint in each of the to-be-selected random joint variable groups ;as well as

If the number of preferred random joint variable groups is 1, the preferred random joint variable group is taken as the inverse of the target coordinates.
The method of claim 4, wherein calculating an error value between a target coordinate of the end joint and a random coordinate of the end joint comprises:

According to the formula ΔE=ΔP+ΔR and

Calculating an error value ΔE between a target coordinate of the end joint and a random coordinate of the end joint;

Where ΔP represents the position difference, ΔR represents the attitude difference, ω ρ and ω γ are the weights of the position difference ΔP and the attitude difference ΔR, P s is the target position of the end joint, and p sj is the target position component of the end joint , d sj is the target pose of the end joint, P s , p sj and d sj can be obtained from the target coordinates of the end joint, P r is the random position of the end joint, p rj is the random position component of the end joint, d rj For the random pose of the end joint, P r , p rj and d rj can be obtained from the random coordinates of the end joint, and ||P s -P r || is the calculation of the target position of the end joint and the random position of the end joint. The distance between them.
A device for determining the inverse solution result of a tandem robot, comprising:

a target coordinate acquiring module, configured to acquire target coordinates of an end joint of the tandem robot, wherein at least one front end joint is connected in series above the end joint;

a random joint variable group generating module, connected to the target coordinate acquiring module, and set to random production Generating a plurality of random joint variable groups, wherein each random joint variable group includes a joint variable corresponding to each of the front end joints;

a random coordinate calculation module, connected to the random joint variable group generating module, configured to calculate a random coordinate of the end joint according to each of the random joint variable groups;

An approximate coordinate selection module, coupled to the random coordinate calculation module, configured to select an approximate coordinate of the end joint in the random coordinate according to the target coordinate;

The inverse solution result determining module is connected to the approximate coordinate selecting module, and is configured to determine an inverse solution result of the target coordinate according to the random joint variable group corresponding to the approximate coordinate;

The approximate coordinate is a random coordinate in which the error of the random coordinate and the target coordinate is within an error threshold range.
The apparatus of claim 7, wherein the random coordinate calculation module comprises:

The coordinate transformation matrix unit is arranged to substitute each random joint variable group into a preset coordinate transformation matrix to obtain a plurality of random coordinates corresponding to the end joint.
The apparatus according to claim 8, wherein the random coordinate calculation module further comprises:

a coordinate system establishing unit configured to establish a base coordinate system O 0 (x 0 , y 0 , z 0 ) of the tandem robot and a joint coordinate system O i (x i , y i , z i ) of each joint, wherein , i ∈ [1, n], n is the total number of joints included in the tandem robot;

a transformation matrix determining unit, connected to the coordinate system establishing unit, configured to determine a transformation matrix between two adjacent coordinate systems according to the base coordinate system and the joint coordinate system
as well as

a coordinate transformation matrix calculation unit respectively connected to the transformation matrix determination unit and the coordinate transformation matrix unit, and configured to be according to a formula
Calculating a transformation matrix of the end joint in the base coordinate system O 0 (x 0 , y 0 , z 0 )
Will be said
As the preset coordinate transformation matrix;

Wherein, d 1x , d 1y , d 1z , d 2x , d 2y , d 2z , d 3x , d 3y , and d 3z are used to represent the posture of the end joint, and p x , p y , and p z are used to represent the end joint. position;

The values of d 1x , d 1y , d 1z , d 2x , d 2y , d 2z , d 3x , d 3y , d 3z , p x , p y , and p z are determined by each of the front joints The corresponding joint variable is determined.
The apparatus of claim 9, wherein the approximate coordinate selection module comprises:

An error value calculation unit configured to calculate an error value between a target coordinate of the end joint and a random coordinate of the end joint;

The error value judging unit is connected to the error value calculating unit and configured to acquire the random coordinates of the end joint whose error value is smaller than the error threshold as the approximate coordinates of the end joint.
The apparatus according to claim 10, wherein the inverse solution result determining module comprises:

The candidate random joint variable group acquiring unit is configured to acquire a random joint variable group respectively corresponding to at least two of the approximate coordinates, as a candidate random joint variable group;

The current comparison joint selection unit is connected to the to-be-selected random joint variable group acquisition unit, and is configured to select a front end joint farthest from the end joint as the current comparison joint;

a joint variable acquiring unit is connected to the current matching joint selecting unit, and configured to acquire a joint variable corresponding to the current matching joint in each of the selected random joint variable groups;

a movement amount determining unit, connected to the joint variable acquiring unit, configured to determine, according to the joint variable, a joint movement amount of a current comparison joint corresponding to each of the candidate random joint variable groups;

Screening a random joint variable group unit, and connecting the movement amount determining unit, and setting the candidate random joint variable group corresponding to the minimum joint movement amount of the current comparison joint as a preferred random joint variable group;

Reducing the random joint variable group number unit, and connecting to the screening random joint variable group unit, and setting, if the number of preferred random joint variable groups is greater than 1, selecting the current comparison with the direction toward the end joint a front end joint adjacent to the joint as a current alignment joint, and returning to perform an operation of acquiring a joint variable corresponding to the current comparison joint in each of the selected random joint variable groups;

The inverse solution result obtaining unit is connected to the reduced random joint variable group number unit, and is set such that if the number of preferred random joint variable groups is 1, the preferred random joint variable group is used as the inverse solution result of the target coordinates.
The apparatus according to claim 10, wherein the error value calculation unit is configured to:

According to the formula ΔE=ΔP+ΔR and

Calculating an error value ΔE between a target coordinate of the end joint and a random coordinate of the end joint;

Where ΔP represents the position difference, ΔR represents the attitude difference, ω ρ and ω γ are the weights of the position difference ΔP and the attitude difference ΔR, P s is the target position of the end joint, and p sj is the target position component of the end joint , d sj is the target pose of the end joint, P s , p sj and d sj can be obtained from the target coordinates of the end joint, P r is the random position of the end joint, p rj is the random position component of the end joint, d rj For the random pose of the end joint, P r , p rj and d rj can be obtained from the random coordinates of the end joint, and ||P s -P r || is the calculation of the target position of the end joint and the random position of the end joint. The distance between them.
A non-transitory computer readable storage medium storing computer executable instructions arranged to perform the method of any of claims 1-6.