KR102228937B1 - Planning method for target retrieval using a robotic manipulator in cluttered and occluded environments - Google Patents

Abstract

This specification discloses an optimized planning method for bringing targets using a robot manipulator in a crowded and closed environment. A method for establishing a robot movement plan according to the present specification includes the steps of: (a) recognizing, by a processor, sizes and positions of objects disposed in the work space from an image photographed of the work space; (b) when the processor is capable of moving between any two of the objects without collision, constructing a graph (hereinafter referred to as'path graph') connecting the two arbitrary objects with an edge; And (c) calculating, by the processor, a path in which the number of times of movement of the obstacle is the least until bringing the target from the path graph to the shortest path.

Description

How to establish a robot manipulator's movement plan in a crowded environment{PLANNING METHOD FOR TARGET RETRIEVAL USING A ROBOTIC MANIPULATOR IN CLUTTERED AND OCCLUDED ENVIRONMENTS}

The present invention relates to a method of establishing a plan to shorten the time to move a robot, and more particularly, to a method of establishing a plan to bring a target in the shortest time using a robot manipulator in a crowded environment. will be.

1 is an exemplary diagram for a working environment of a robot manipulator.

Referring to FIG. 1, it can be seen that there are a number of items in a space surrounded by walls, such as a refrigerator, a shelf, and a warehouse. In an environment in which a number of obstacles are concentrated around a target, such as the environment shown in FIG. 1, it is difficult to bring only the target using a robot manipulator without collision with the obstacle. Therefore, some obstacles need to be rearranged.

However, planning to relocate obstacles is not easy even in a static environment where all information about the obstacle (eg location, size, weight, etc.) is known. In addition, chaotic and constrained environments can often create uncertainty due to closures caused by obstructions in the front. Various methods have been proposed to search for targets in a crowded environment, but the focus has not been on optimizing the relocation plan of obstacles.

2 is a reference diagram for the necessity of an optimized obstacle relocation plan.

Referring to FIG. 2, it can be seen that a target (green stripe) and an obstacle (pink) are placed in a work space surrounded by a wall. Since the work space is a limited space surrounded by a wall, the robot manipulator can be accessed along the floor surface of the space. As shown in the figure on the left side of FIG. 2, when the manipulator selects the shortest path to approach the target, four obstacles (indicated by a red thick outline) on the shortest path must be removed. On the other hand, as shown in the figure on the right side of FIG. 2, when a path that minimizes the number of relocation of obstacles is selected, the number of obstacles to be relocated decreases to 2 (indicated by a thick red outline). Most of the total execution time to bring the target is devoted to catching and releasing the obstacle, that is, relocating the obstacle, and the time it takes to transport the target is relatively small. Therefore, in the latter case, the moving path is longer than the shortest path, but the overall execution time is shortened.

Dogar, M. Koval, A. Tallavajhula, and S. Srinivasa, "Object search by manipulation," Autonomous Robots, vol. 36, pp. 153-167, 2014.

An object of the present specification is to provide an optimized planning method for bringing a target using a robot manipulator in a crowded and closed environment.

The present specification is not limited to the above-mentioned tasks, and other tasks that are not mentioned will be clearly understood by those skilled in the art from the following description.

A method for establishing a robot movement plan according to the present specification for solving the above-described problem includes the steps of: (a) recognizing, by a processor, sizes and positions of objects disposed in the work space from an image photographed of the work space; (b) when the processor is capable of moving between any two of the objects without collision, constructing a graph (hereinafter referred to as'path graph') connecting the two arbitrary objects with an edge; And (c) calculating, by the processor, a path in which the number of times of movement of the obstacle is the least until bringing the target from the path graph to the shortest path.

The step (b) may be a step of determining whether the processor can move between the arbitrary two objects without collision based on the width of the object having the largest width among the objects.

Preferably, the step (b) comprises the step of determining whether the processor can move between the arbitrary two objects without collision based on the width including the radius of the object having the largest width among the objects and the thickness of the robot hand. Can be

In the step (c), when there are a plurality of initially accessible obstacles, the processor calculates the minimum number of movements before bringing the target to each of the first accessible obstacles, and the minimum number of movements of each of the first accessible obstacles It may be a step of calculating the path with the least number of times as the shortest path.

When there are two or more paths having the least number of times among the minimum number of movements of each of the first accessible obstacles, step (c) is performed by the processor calculating the Euclidean distance of each of the paths having the least number of times, and the calculated It may be a step of calculating the path with the least Euclidean distance as the shortest path.

A method of establishing a robot movement plan according to an embodiment of the present specification includes: (d) calculating whether a collision caused by the robot arm occurs when the processor controls the robot manipulator to catch the next obstacle according to the shortest path. step; And (e) when the processor is expected to collide with the robot arm, setting the predicted obstacle to be collided as a temporary target, and re-executing step (c).

A method of establishing a robot movement plan according to another embodiment of the present specification includes: (d) the size and position of objects placed in the working space from an image photographed of the working space while the processor controls the robot according to the shortest path. Updating to; And (e) re-executing steps (a) to (c) when there is an object newly recognized in step (d) by the processor.

A method of establishing a robot movement plan according to another embodiment of the present specification is, between steps (a) and (b), (a-1) in which the processor determines whether or not a target is recognized in the work space. Step; may include.

According to the first example, after the step (a-1), (a-2) when the target is not recognized in the working space, the processor removes the bulky obstacle among the first accessible obstacles, and the (a) re-executing step; may further include.

According to a second example, after the step (a-1), (a-2), when the target is not recognized in the work space, the processor removes an obstacle closest to the robot among the first accessible obstacles, and the (a) re-executing step; may further include.

According to a third example, after the step (a-1), (a-2), when the target is not recognized in the work space, the processor removes an obstacle located far away from the robot among the first accessible obstacles. It may include; re-executing step (a).

Other specific details of the present invention are included in the detailed description and drawings.

According to the present specification, it is possible to provide an optimized planning method for bringing a target using a robot manipulator in a crowded and closed environment.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is an exemplary diagram for a working environment of a robot manipulator.
2 is an exemplary diagram of an obstacle relocation plan optimization.
3 is a reference diagram of a work space grasped by a camera.
4 is a flowchart of a method of establishing a robot movement plan according to the present specification.
5 is a reference diagram of three cases in which a target and an obstacle are arranged.
6 is a reference diagram for understanding a path that can be moved without collision.
7 is a reference diagram when a robot hand grasps an object.
8 is an exemplary diagram of configuring a path graph.
9 is an exemplary diagram for calculating the shortest path.
10 is a flowchart of a method of establishing a robot movement plan according to an embodiment of the present specification.
11 is a flowchart of a method of establishing a robot movement plan according to another embodiment of the present specification.
12 is a reference diagram for a case in which an invisible obstacle is later seen according to another embodiment of the present specification.
13 is a flowchart of a method of establishing a robot movement plan according to another embodiment of the present specification.
14 is a flowchart of a method of establishing a robot movement plan according to another embodiment of the present specification.

Advantages and features of the invention disclosed in the present specification, and a method of achieving them will become apparent with reference to the embodiments described later in detail together with the accompanying drawings. However, the present specification is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms. It is provided to completely inform the scope of the present specification to a technician (hereinafter referred to as'the person in charge'), and the scope of the rights of the present specification is only defined by the scope of the claims. Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Retrieving targets in a congested environment requires several processes such as perception, relocation planning, grasping, path planning, and navigation. This specification focuses on a method of establishing a path in which a robot manipulator does not collide with an obstacle to bring a target. Before describing the method of establishing a robot movement plan according to the present specification, the prerequisites will be described.

In the present specification, the robot manipulator includes a robot hand that holds an object and a robot arm that is connected to the robot hand and transmits a driving force. All configurations sometimes referred to only as'robots' in this specification mean'robot manipulators'.

All objects (including targets and obstacles) are concentrated in the working space of the robot manipulator. At this time, the manipulator path without obstacles and collisions does not exist, so repositioning of the obstacles is indispensable. Also, for simplicity, it is assumed that all objects can be held by the robot hand in any direction. In addition, it is not considered to hold the top of the object for motion in a closed space, but consider that the robot hand takes a 2D path (at a fixed height in 3D space) along the side of the object. It is assumed that the camera fixed at the front-top of the working space is sufficient to recognize the working space of the robot manipulator (see Fig. 3).

It can be seen that the work space is occluded by an obstacle. In order for the robot hand to bring the target, the obstacle placed in front of it must be removed. At this time, the obstacle accessible by the robot hand without rearrangement is referred to as the'first accessible obstacle'. Referring to the drawing shown at the far right of FIG. 3, there are four first accessible obstacles (indicated by thick outlines), and the remaining obstacles cannot be approached due to lack of space to which the robot hand can access.

The robot movement plan establishment method according to the present specification may be executed by a processor. The processor is a microprocessor known in the art, an application-specific integrated circuit (ASIC), another chipset, a logic circuit, a register, a communication modem, and data to execute each step of the method for establishing a robot movement plan to be described below. It may include a processing device and the like. In addition, when the method to be described below is implemented in software, each step may be implemented as a set of program modules. In this case, the program module may be stored in a memory and executed by a processor.

4 is a flowchart of a method of establishing a robot movement plan according to the present specification.

Referring to FIG. 4, first, in step S10, the processor may recognize the sizes and positions of objects disposed in the working space from an image photographed of the working space. At this time, it can be divided into three cases according to the arrangement state of the target and the obstacle.

5 is a reference diagram of three cases in which a target and an obstacle are arranged.

Referring to the left side of FIG. 5, the arrangement of all obstacles is identified, and the target is visible (Case 1). In this case, the robot movement plan will all be established before relocating the first obstacle. Referring to the center of FIG. 5, only some of the obstacles are located and the target is visible (Case 2). In this case, some of the obstacles will be seen in the process of rearranging some of the obstacles according to the robot movement plan. Referring to the right side of FIG. 5, only some of the obstacles are located and the target is not visible (Case 3). In this case, some obstacles and targets will be visible after the obstacles placed in front are rearranged. Among the three cases, the arrangement of all obstacles is identified, and the case where the target is visible (Case 1) will be explained first. The remaining cases (Cases 2 and 3) will be explained next.

Referring to FIG. 4 again, in step S20, when the processor is able to move between any two of the objects without collision, a graph connecting the two arbitrary objects with an edge (hereinafter,'path graph') is created. Configurable. Such a path graph is a line representing a relationship between two objects, and the relationship means the existence of a collision-free path, and the path may or may not be a straight line.

Step S20 will be described in more detail with reference to FIGS. 6 to 8.

6 is a reference diagram for understanding a path that can be moved without collision.

Referring to FIG. 6A, five objects O ₁ to O ₅ having various sizes are shown. At this time, as shown in (b) of FIG. 6, it will be assumed that there are no _{O 2} and O _4. If so, the largest object, O ₃ , can move through a path connecting O ₂ and O ₄ (area indicated by a gray diagonal line). Therefore, as shown in (c) of Figure 6, the, if a large object, the O ₃ movement is possible through the path connecting the O ₂ and O _4, travels through a path to all of these other objects connected to the O ₂ and O ₄ This is possible. As shown in (d) of FIG. 6, the same method may be applied between _{O 1} and O _2. And as shown in (e) of Figure 6, when O ₄ , O ₂ and O ₁ are removed, O ₃ can be brought. On the other hand, _{even if only O 5} is removed, the robot can pull _{O 3.} To determine whether there is a collision, the modified VFH ⁺ can be applied (see: I. Ulrich and J. Borenstein, "VFH+: Reliable obstacle avoidance for fast mobile robots," in Proc. of IEEE Int. Conf. on Robotics and Automation. (ICRA), 1998, pp. 1572-1577.)

Accordingly, in step S20, the processor may determine whether it is possible to move between the arbitrary two objects without collision based on the width of the object having the largest width among the objects. Meanwhile, the processor may determine whether the arbitrary two objects can be moved without collision based on a width including a radius of an object having the largest width among the objects and a thickness of a robot hand.

7 is a reference diagram when a robot hand grasps an object.

Referring to FIG. 7, the radius of the object _{is expressed as "r i} " and the thickness of the robot hand is expressed as _{"r r ".} Therefore, when the robot hand grasps the object, the radius becomes r _g = r _i + r _r as shown in FIG. 7. The "r _g " may be basic information of a size for determining whether or not it is possible to move between two objects without collision. In (c) and (d) of FIG. 6, _{the size (gray ring) when the largest object O 3} is held by the robot's hand is used as a reference, and it is an example of determining whether it is possible to move between two objects without collision.

Preferably, the processor is based on a width obtained by adding a safety margin for preventing collision with another object during movement to the radius of the object having the largest width among the objects and the thickness of the robot hand. It can be determined whether it is possible to move between two objects without collision.

8 is an exemplary diagram of configuring a path graph.

8A to 8H, there are three obstacles (O ₁ to O ₃ ) and one target (O _t ), and four nodes (V ₁ to V ₃ , V _{t) corresponding thereto.} ) Can be checked. Although the wall around the object is omitted in FIGS. 8A to 8H, it is assumed that the robot is still surrounded by the wall and cannot access from any direction other than the front side.

In (a) of FIG. 8, O _{2, which} is the largest object, is selected as an object for determining whether a path can be moved without collision. The gray ring around the object O ₂ is the width taking into account the safety distance from the robot hand. In (b) of FIG. 8, a collision may occur between the _{object O 1} and the object O ₂ , and thus a path is not connected between the _{node V 1} and the node V _2. In (c) of FIG. 8, since the object O ₁ and the object O ₃ can move without collision, a path is connected between the _{node V 1} and the node V _3. In the same way, paths are connected between nodes in (d) to (f) of FIG. 8. In (g) of FIG. 8, a collision may occur between the _{object O t} and the object O ₃ , so that a path is not connected between the _{node V t} and the node V _3. In FIG. 8(h), it is determined whether a collision exists with respect to the number of all possible combinations between the object and the object, and the paths are connected accordingly.

Referring back to FIG. 4, in step S30, the path in which the number of times of movement of the obstacle is the least until the processor brings the target from the path graph may be calculated as the shortest path.

The processor calculates the number of movements from the first accessible obstacle in the path graph. When there are a plurality of initially accessible obstacles, the processor calculates the minimum number of movements to bring the target for each first accessible obstacle, and the path with the least number of the minimum number of movements of each of the first accessible obstacles Can be calculated as the shortest path. To calculate the minimum number of movements, the processor may use a BFS (Breadth First search) algorithm (see T. Cormen, C. Leiserson, R. Rivest, and C. Stein, Introduction to algorithms. MIT press, 2009). .).

On the other hand, when two or more paths with the least number of times among the minimum number of movements of each of the first accessible obstacles, that is, two or more paths with a tie, the processor determines the Euclidean distance of each path with the least number of times ( Euclidean distances) may be calculated, and a path having the smallest Euclidean distance may be calculated as the shortest path.

9 is an exemplary diagram for calculating the shortest path.

Looking at the left side of FIG. 9, targets marked in green and obstacles marked in red can be seen in a work space surrounded by a wall. Looking at the right side of FIG. 9, nodes and paths are displayed corresponding to targets and obstacles. The robot's working direction is the downward direction in the drawing, and the first accessible obstacles are 4, 5, 6, and 8. The processor may calculate the minimum number of times that the target number 7 can be fetched, starting from 4, 5, 6, and 8. In addition, it is assumed that the path calculated by the processor using the fourth obstacle as the first accessible obstacle is calculated as the smallest number of times as two times. It is a path that is removed in the order of number 4 and number 9 according to the path indicated in bold red on the right side of FIG. 9.

In the process of constructing the path graph described so far, the kinematic constraints of the robot were not considered. Although the robotic hand is the path to grab objects without collision, the robotic arm may collide with other objects. Therefore, it may be necessary to determine whether the robot arm collides and to reset the path.

10 is a flowchart of a method of establishing a robot movement plan according to an embodiment of the present specification.

Referring to FIG. 10, it can be seen that steps S10 to S30 of the flowchart shown in FIG. 4 are the same. Therefore, a repetitive description of steps S10 to S30 will be omitted. In step S40 after step S30, the processor may control the robot manipulator to catch the next obstacle along the shortest path. The first obstacle to be removed will be the first accessible obstacle. As explained earlier, there is no mechanical limitation of the robot for the robot to remove the first accessible obstacle. The processor performs step S40 and then proceeds to step S50.

In step S50, the processor may calculate whether a collision occurs due to the robot arm. The processor determines whether a robotic arm is collided with or without "J. Pan, S. Chitta, and D. Manocha, "Fcl: A general purpose library for collision and proximity queries," in Proc. of IEEE Int. Conf. on Robotics and Automation (ICRA), 2012, pp. 3859-3866." If there is no collision by the robot arm (NO in step S50), the processor proceeds to step S40 to remove the next obstacle along the shortest path. Therefore, if there is no collision by the robot arm in all paths, the processor will repeatedly execute branch steps S40 and S50 when bringing the target. On the other hand, if a collision by the robot arm is expected (YES in step S50), the processor proceeds to step S51.

In step S51, when a collision by the robot arm is predicted by the processor, the predicted collision target obstacle is set as a temporary target, and the process proceeds to step S30. In addition, the processor preferentially removes the expected collision target obstacle by recalculating the shortest path through which the temporary target can be brought. Accordingly, until the temporary target is removed, the processor repeatedly performs steps S30 to S50. When the temporary target is removed, the processor may remove the obstacle to bring the original target according to the shortest path initially calculated.

Meanwhile, out of the three cases in which the target and the obstacle shown in FIG. 5 are arranged so far, the arrangement of all the obstacles in FIG. 5 is recognized, and the case where the target is visible (Case 1) has been described. Hereinafter, as shown in the center of FIG. 5, only some of the obstacles are identified, and a case where the target is visible (Case 2) will be described.

11 is a flowchart of a method of establishing a robot movement plan according to another embodiment of the present specification.

Referring to FIG. 11, it can be seen that steps S10 to S30 of the flowchart shown in FIG. 4 are the same. Therefore, a repetitive description of steps S10 to S30 will be omitted. In step S40 after step S30, the processor may control the robot manipulator to catch the next obstacle along the shortest path. The first obstacle to be removed will be the first accessible obstacle. As explained earlier, there is no mechanical limitation of the robot for the robot to remove the first accessible obstacle. The processor performs step S40 and then proceeds to step S50.

In step S50, while the processor controls the robot according to the shortest path, the size and position of objects disposed in the work space may be updated from an image photographed of the work space. At this time, the image of the work space used by the processor is the image taken after step S40, that is, the image after one obstacle has been removed. Then, the processor proceeds to step S51.

If there is no newly recognized object (NO in step S51), the processor will proceed to step S40 and remove the next obstacle according to the shortest path. On the other hand, if there is no newly recognized object (YES in step S51), the processor proceeds to step S10 and executes steps S10 to S30 again. That is, the path graph for the newly recognized obstacle is reconstructed, and the shortest path is calculated again according to the reconstructed path graph. Thereafter, the processor will control to remove the obstacle according to the recalculated shortest path.

12 is a reference diagram for a case in which an invisible obstacle is later seen according to another embodiment of the present specification.

Referring to (a) of FIG. 12, targets marked in green and obstacles marked in red can be identified in a work space surrounded by a wall. At this time, two obstacles marked in pink behind the first accessible obstacle marked in gray are not recognized. The thick red line means the shortest path.

Referring to FIG. 12B, it is a situation in which an invisible obstacle 4 is recognized after removing the 0 obstacle. Accordingly, the processor may reconfigure the route graph including No. 4, and calculate the shortest route connecting Nos. 4 and 8 according to the reconfigured route graph.

Referring to (c) of FIG. 12, it is a situation in which an invisible obstacle 7 is recognized after removing the number 4 obstacle. Similarly, the processor may reconfigure the route graph including No. 7, and calculate the shortest route connecting Nos. 7 and 8 according to the reconfigured route graph.

Referring to (d) of FIG. 12, the final route from 0 to 8 is shown.

So far, the algorithm for finding the shortest path in the situation where the location of the target is recognized has been described. Hereinafter, as shown on the right side of FIG. 5, only a part of the obstacles are identified, and a case where the target is not visible (Case 3) will be described.

13 is a flowchart of a method of establishing a robot movement plan according to another embodiment of the present specification.

Referring to FIG. 13, it can be seen that steps S10 to S30 of the flowchart shown in FIG. 4 are the same. Therefore, a repetitive description of steps S10 to S30 will be omitted. The difference from FIG. 4 is that steps S11 and S12 are added between steps S10 and S20.

In step S11 after step S10, the processor may determine whether or not the target object is recognized in the work space. When the position of the target or the like is recognized (YES in step S11), the processor proceeds to step S20. Subsequent steps S20 and S30 are the same as those described above. On the other hand, if the position of the target or the like is not recognized (NO in step S11), the processor proceeds to step S12.

In step S12, the processor removes any one of the obstacles and proceeds to step S10. The processor may repeatedly execute steps S10 to S12 until the target is recognized. Meanwhile, various embodiments are possible in the process of selecting which one of the obstacles to be removed by the processor in step S12.

According to the first example, in step S12, the processor may remove an obstacle that is the largest among the first accessible obstacles, that is, the obstacle that most obstructs the view of the robot. According to the second example, in step S12, the processor may remove an obstacle closest to the robot among the first accessible obstacles. According to the third example, in step S12, the processor may remove the obstacle located furthest from the robot among the first accessible obstacles.

On the other hand, the robot movement plan establishment method according to the present specification described with reference to FIGS. 4, 10, 11, and 13 has been described as an embodiment separated from each other in order to more clearly describe the characteristics of the algorithm according to each situation. , Each algorithm described in this specification may be implemented as a single integrated algorithm.

14 is a flowchart of a method of establishing a robot movement plan according to another embodiment of the present specification.

Referring to FIG. 14, steps S10, S11, S12, S20, and S30 are the same as the embodiment shown in FIG. 13, and steps S40, 50, and 51 are the same as the embodiment shown in FIG. 11, and steps S60 and S61 Corresponds to steps S50 and S51 shown in FIG. 10.

Meanwhile, the method of establishing a robot movement plan according to the present specification may be implemented in the form of a computer program. The computer program includes C/C++, C#, JAVA, Python that can be read by the computer's processor (CPU) through the computer's device interface in order for the computer to read the program and execute the methods implemented as a program. , May include a code (Code) coded in a computer language such as machine language. Such code may include a functional code related to a function defining necessary functions for executing the methods, and a control code related to an execution procedure necessary for the processor of the computer to execute the functions according to a predetermined procedure. can do. In addition, these codes may further include additional information required for the processor of the computer to execute the functions or code related to a memory reference to which location (address address) of the internal or external memory of the computer should be referenced. have. In addition, when the processor of the computer needs to communicate with any other computer or server in a remote in order to execute the functions, the code is It may further include a communication-related code for whether to communicate or what information or media to transmit and receive during communication.

The stored medium is not a medium that stores data for a short moment, such as a register, cache, memory, etc., but a medium that stores data semi-permanently and can be read by a device. Specifically, examples of the storage medium include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. That is, the program may be stored in various recording media on various servers to which the computer can access, or on various recording media on the user's computer. In addition, the medium may be distributed over a computer system connected through a network, and computer-readable codes may be stored in a distributed manner.

As described above, embodiments of the present specification have been described with reference to the accompanying drawings, but those of ordinary skill in the art to which this specification pertains can be implemented in other specific forms without changing the technical spirit or essential features. You will be able to understand. Therefore, the embodiments described above are illustrative in all respects, and should be understood as non-limiting.

Claims (10)

(a) recognizing, by a processor, sizes and positions of objects arranged in the working space from an image photographed of the working space;
(b) When the processor is able to move between any two objects among all the recognized objects without collision, a graph connecting the two arbitrary objects with an edge (hereinafter referred to as'path graph') is used for all recognized objects. Configuring between them; And
(c) calculating, by the processor, a path in which the number of times of movement of the obstacle is the least until bringing the target from the path graph to the shortest path. The method according to claim 1,
The step (b),
The method of establishing a robot movement plan, wherein the processor determines whether it is possible to move between the arbitrary two objects without collision based on the width of the object having the largest width among the objects. The method according to claim 2,
The step (b),
The method of establishing a robot movement plan, wherein the processor determines whether or not the arbitrary two objects can be moved without collision based on a width including a radius of an object having the largest width among the objects and a thickness of a robot hand. The method according to claim 1,
The step (c),
When there are a plurality of initially accessible obstacles, the processor calculates the minimum number of movements before bringing the target for each first accessible obstacle, and the path with the least number of times among the minimum number of movements of each of the first accessible obstacles How to establish a robot movement plan, which is the step of calculating the shortest path. The method of claim 4,
In the step (c), when there are two or more paths having the least number of times among the minimum number of movements of each of the first accessible obstacles, the processor calculates the Euclidean distance of each path with the least number of times, and the calculation Robot movement planning method, which is the step of calculating the path with the smallest Euclidean distance as the shortest path. The method according to claim 1,
(d) calculating whether a collision by the robot arm occurs when the processor controls the robot manipulator to catch the next obstacle along the shortest path; And
(e) when the processor is expected to collide with the robot arm, setting the predicted obstacle to be collided as a temporary target and re-executing step (c). The method according to claim 1,
(d) updating the sizes and positions of objects placed in the working space from an image photographed of the working space while the processor controls the robot according to the shortest path; And
(e) re-executing steps (a) to (c) when the processor has an object newly recognized in step (d). The method according to claim 1,
Between the step (a) and the step (b),
(a-1) determining, by the processor, whether or not the target object is recognized in the work space; And
(a-2) when the processor does not recognize a target in the work space, removing the largest obstacle among the first accessible obstacles and re-executing the step (a); How to plan. The method according to claim 1,
Between the step (a) and the step (b),
(a-1) determining, by the processor, whether or not the target object is recognized in the work space; And
(a-2) when the processor does not recognize a target in the work space, removing an obstacle closest to the robot among the first accessible obstacles and re-executing the step (a); How to plan. The method according to claim 1,
Between the step (a) and the step (b),
(a-1) determining, by the processor, whether or not the target object is recognized in the work space; And
(a-2) when the processor does not recognize a target in the work space, removing the obstacle located furthest from the robot among the first accessible obstacles and re-executing the step (a); How to plan your travel.

Images