Automatic generation method for moving path of robot manipulator
3 claims: 2 independent, 1 dependent
- 1【特許請求の範囲】 【請求項1】 三次元空間内で作業をするロボットマニピュレータの要素を,障害物を回避しつつ,初期位置から目標位置まで移動させる経路を自動生成する方法において,上記要素の初期位置と目標位置とを含む複数の平面のうちの所望の平面上の障害物の断面形状を,ロボットマニピュレータの要素が障害物との干渉を回避できる程度に拡大した侵入禁止領域を作成し,上記所望の平面上の上記要素の初期位置と目標位置とを直線で結んで上記経路を初期生成し,上記経路が上記侵入禁止領域と干渉するときは,上記経路を,上記侵入禁止領域との干渉を回避するような侵入禁止領域の頂点を通る経路に補正してなることを特徴とするロボットマニピュレータの移動経路の自動生成方法。
- 2【請求項2】 上記補正された経路が複数ある場合には,所定の評価を行って1つの経路を選択する請求項1記載のロボットマニピュレータの移動経路の自動生成方法。
- 3【請求項3】 上記所望の平面が複数ある場合には,該所望の平面ごとに得られた上記経路の中から,所定の評価を行って1つの経路を選択する請求項1又は2記載のロボットマニピュレータの移動経路の自動生成方法。
Independent claims3
54 paragraphs in 1 section, as filed
Description: TECHNICAL FIELD [Detailed description of the invention]
【0001】
[Technical field to which the invention belongs]
The present invention relates to a method for automatically generating a movement path of a robot manipulator. Specifically, the present invention automatically moves a path for moving an element of a robot manipulator working in a three-dimensional space from an initial position to a target position while avoiding obstacles. It is about how to generate.
【0002】
[Conventional technology]
There are many disclosure examples of a method of automatically generating a path for moving an element of a robot manipulator working in a three-dimensional space from an initial position to a target position while avoiding obstacles. For example, Japanese Patent Application Laid-Open No. 4-235606 and Japanese Patent Application Laid-Open No. 4-280304 both use the configuration space method. However, since the amount of calculation is enormous if the configuration space method is used as it is, in the former, the amount of calculation is reduced by sequentially searching the free space near the boundary between the obstacle and the free space in the configuration space. .. Moreover, in the latter, the amount of calculation is reduced and the speed is increased by providing a table in which the joint angle space, which is a kind of configuration space, and the Cartesian coordinate space are associated with each other. Further, in Japanese Patent Application Laid-Open No. 5-250023, the potential method is used, and the neural network is applied to generate the potential field. Here, the neural network is trained using the risk calculated from the distance between each link of the articulated manipulator and the obstacle, facilitating the evaluation calculation regarding the risk of collision. The descent method is used to generate the orbit. Furthermore, Japanese Patent Application Laid-Open No. 5-198155 and Japanese Patent Application Laid-Open No. 5-204436 use both the configuration space method and the potential method. That is, in the former, the evaluation function that changes depending on the distance from the obstacle is reflected in the configuration space to avoid the tendency that the generated route is set in the immediate vicinity of the obstacle.<sup>* </sup>It uses an algorithm. In the latter, the chaotic steepest descent method is used as a search method for the potential space to ensure escape from the local minimum point (local minimum). In this way, all of the disclosure examples deal with the three-dimensional interference avoidance problem in which the robot manipulator moves from the initial posture to the target posture while avoiding the obstacle in the environment where the obstacle exists.
【0003】
[Problems to be Solved by the Invention]
The conventional automatic generation method of the movement path of the robot manipulator as described above has the following problems. (1) A large amount of storage area is required. In the configuration space method, it is necessary to memorize the free space where there are no obstacles. In Japanese Patent Application Laid-Open No. 4-235606, Japanese Patent Application Laid-Open No. 4-280304, and Japanese Patent Application Laid-Open No. 5-250023, the space is divided into cells to compress the amount of handling information, but a large amount of storage area is still required. is there. (2) It takes time to calculate. In the potential method, it takes time to calculate the potential field when an obstacle with a complicated shape exists. Japanese Patent Application Laid-Open No. 5-119815 claims high-speed collision avoidance processing that makes use of the parallel processing property of the neural network, but it is not necessarily high-speed processing when the learning time of the neural network is taken into consideration. Moreover, in the configuration space method, a large amount of calculation is required to calculate the mapping between the Cartesian coordinate space and the configuration space. Therefore, when the dimension of space becomes high, it is difficult to efficiently calculate the free space in the configuration space. In Japanese Patent Application Laid-Open No. 4-280304, the amount of calculation is reduced by using a table in which the orthogonal coordinate space and the configuration space are associated with each other, but a large amount of storage area is required accordingly.
【0004】
(3) The generated route may be set in the immediate vicinity of an obstacle. In the configuration space method, it is difficult to consider the position in the Cartesian coordinate space because the route is planned in the configuration space instead of the Cartesian coordinate space. Therefore, the generated path may be set in the immediate vicinity of an obstacle in the Cartesian coordinate space. (4) The generated route may be set by detouring from obstacles more than necessary. For the same reason as in (3) above, in the configuration space method, the generated route may be set by detouring from the obstacle more than necessary. Moreover, in the potential method, it is possible to prevent the route from being set in the immediate vicinity of the obstacle by adjusting the parameters in the generation of the potential field, but it is difficult to prevent the route from being detoured more than necessary. Moreover, since parameter adjustment is generally a trial and error process, it is difficult to make accurate adjustment. Therefore, depending on the route setting, a local minimum point may occur in the potential field and stop in the middle of the search. For this reason, Japanese Patent Application Laid-Open No. 5-204436 presents a method of getting out of the local minimum point by the vibration component generated from the time evolution equation of chaos, but the path passing through the local minimum point takes an extra detour. There may be. In order to solve the problems in the conventional technology, the present invention improves the automatic generation method of the movement path of the robot manipulator, and performs a high-speed route generation process with a relatively small storage capacity and a practical calculation time. It is an object of the present invention to provide a method for automatically generating a movement path of a robot manipulator that can be realized.
【0005】
[Means for solving problems]
In order to achieve the above object, the present invention is in a method of automatically generating a path for moving an element of a robot manipulator working in a three-dimensional space from an initial position to a target position while avoiding obstacles. Create an intrusion prohibition area in which the cross-sectional shape of an obstacle on a desired plane among multiple planes including the initial position and the target position of the robot manipulator is expanded to the extent that the elements of the robot manipulator can avoid interference with the obstacle. When the initial position of the element on the desired plane and the target position are connected by a straight line to initially generate the route, and the route interferes with the intrusion prohibited area, the route is referred to as the intrusion prohibited area. It is configured as a method for automatically generating a movement path of a robot manipulator, which is characterized in that the path passes through the apex of the intrusion prohibited area so as to avoid the interference of the robot manipulator. Furthermore, it is a method of automatically generating a movement route of a robot manipulator that performs a predetermined evaluation and selects one route when there are a plurality of the corrected routes. Further, when there are a plurality of the desired planes, a method for automatically generating a movement path of a robot manipulator that performs a predetermined evaluation and selects one path from the paths obtained for each desired plane. Is.
【0006】
According to the present invention, when automatically generating a path for moving an element of a robot manipulator working in a three-dimensional space from an initial position to a target position while avoiding obstacles, the initial position and the target position of the above element A desired plane is selected from a plurality of planes including and. Then, an intrusion prohibition area is created in which the cross-sectional shape of the obstacle on the plane is expanded to the extent that the elements of the robot manipulator can avoid interference with the obstacle. Expanding the elements of the robot manipulator to the extent that interference with obstacles can be avoided means, for example, expanding a predetermined distance in anticipation of safety, further expanding the distance according to the acceleration of the elements, and the like. Next, the path is initially generated by connecting the initial position of the element on the desired plane and the target position with a straight line. When the route interferes with the intrusion prohibited area, the route is corrected to a route passing through the apex of the intrusion prohibited area so as to avoid interference with the intrusion prohibited area. For example, among the routes that pass through the vertices, the routes that pass through the intrusion prohibited area are invalid. In this way, obstacle avoidance in the three-dimensional space is reduced to the two-dimensional plane obstacle avoidance problem in the present invention. As a result, high-speed route generation processing with practical calculation time can be realized in a relatively small storage area. In addition, routes are generated that do not interfere with obstacles and do not detour more than necessary. Further, when there are a plurality of the corrected routes, a predetermined evaluation may be performed and one route may be selected. Further, when considering a plurality of the desired planes in order to consider interference avoidance three-dimensionally, it is sufficient to perform a predetermined evaluation from the above-mentioned paths obtained for each desired plane and select one path. .. In this way, when a plurality of solutions are obtained, the narrowing down can be automatically performed.
【0007】
BEST MODE FOR CARRYING OUT THE INVENTION
as well as [Example]
Hereinafter, embodiments and examples of the present invention will be described with reference to the accompanying drawings for the purpose of understanding the present invention. It should be noted that the following embodiments and examples are examples that embody the present invention and do not limit the technical scope of the present invention. Here, FIG. 1 is a flow chart showing a schematic configuration of an embodiment of the present invention and a method for automatically generating a movement path of a robot manipulator according to an embodiment (first embodiment), and FIG. 2 is another embodiment of the present invention. A flow chart showing a schematic configuration of an automatic generation method of a movement path of a robot manipulator according to an example (second embodiment), FIG. 3 is an explanatory diagram and a diagram showing a specific example of a plane set in the first embodiment method. 4 is an explanatory diagram showing the state after the execution of step S02 with respect to the plane, FIG. 5 is an explanatory diagram showing the state after the execution of step S04 with respect to the plane, and FIG. 6 is an explanatory diagram showing the final state of the movement path generation with respect to the plane. is there.
【0008】
In the method of automatically generating the movement path of the robot manipulator according to the embodiment and the embodiment (first embodiment) of the present invention, the elements of the robot manipulator working in the three-dimensional space are avoided while avoiding obstacles. It is the same as the conventional example in that a route for moving from the initial position to the target position is automatically generated. However, in the first embodiment, as shown in FIG. 1, the robot manipulator element determines the cross-sectional shape of an obstacle on a desired plane among a plurality of planes including the initial position and the target position of the element. An intrusion prohibition area expanded to the extent that interference with obstacles can be avoided is created (S01, S02), and the above path is initially generated by connecting the initial position and the target position of the above element on the desired plane with a straight line. When the route interferes with the intrusion prohibited area, the route is corrected to a route passing through the apex of the intrusion prohibited area so as to avoid interference with the intrusion prohibited area (S03, S04). It differs from the conventional example in that it is used. Further, when there are a plurality of the corrected routes, a predetermined evaluation may be performed to select one route (S05). Expanding the elements of the robot manipulator to the extent that interference with obstacles can be avoided means, for example, expanding a predetermined distance in anticipation of safety, further expanding the distance according to the acceleration of the elements, and the like.
【0009】
Hereinafter, the manual route planning of the robot manipulator by the method of the first embodiment will be described in more detail. In FIG. 1, first, the cross-sectional shape of an obstacle is shown on a representative plane (corresponding to a desired plane) selected by the operator from among innumerable planes including two points, the initial position and the target position of the robot manipulator hand. Create (S01). On this plane, as shown in FIG. 3, there are an initial position of the hand, a target position, and a cross section of an obstacle. In Fig. 3, the initial position of the manipulator's hand (corresponding to one of the elements) is Start, the target position is Goal, and the three hatched closed regions Obs1, Obs2, and Obs3 all show the cross section of the obstacle. FIG. 4 shows the state after executing steps S02 and S03 on such a plane. In FIG. 4, in step S02, an intrusion prohibition area is set by expanding the cross section of the obstacle by the distance r based on the avoidance distance r for avoiding the interference between the hand of the given robot manipulator and the obstacle. This distance r may be a uniform value specified by the user, or the size of a certain ratio may be automatically set according to the size of the cross section of the obstacle. Further, the distance r may be expanded or contracted according to the acceleration of the robot. In Fig. 4, the obstacle cross sections Obs1 and Obs2 are close to each other, so the intrusion prohibited areas overlap each other. Therefore, these two areas are integrated and displayed as one intrusion prohibited area. That is, after the execution of step S02 is completed, the intrusion prohibited areas A1 and A2 are generated.
【0010】
Subsequently, in step S03, the interference check between the intrusion prohibited area generated in step S02 and the virtual route is performed. In the first execution of step S03, an evasion point for avoiding the interference has not been generated yet, so the interference check is performed on the virtual path Tr01 connecting the initial position Start and the target position Goal with a straight line. In FIG. 4, since the intrusion prohibited area A1 and the virtual path Tr01 interfere with each other, step S04 is executed for the virtual path Tr01 after that. Figure 5 shows the state after executing steps S04 and S03. In step S04 in FIG. 5, the avoidance point is set based on the interference information of step S03 executed immediately before that step S04. In step S03 executed immediately before the above, the intrusion prohibited area A1 and the virtual path Tr01 interfered with each other. Therefore, using the vertex information of the intrusion prohibited area A1, an avoidance point for avoiding interference with the intrusion prohibited area A1 is set. In this case, avoidance points P11 and P12 are set.
【0011】
Subsequently, in step S03, the interference check between the intrusion prohibited area and the virtual route is performed. In the second and subsequent executions of step S03, workarounds have already been generated. Therefore, the virtual path Tr11 connecting the initial position Start and the avoidance point P11 with a straight line, the virtual path Tr12 connecting the avoidance point P11 and the avoidance point P12 with a straight line, and the avoidance point P12 and the target position Goal Interference check is performed for each of the virtual paths Tr13 connecting the lines with a straight line. Then, step S04 is executed for each of the interfering virtual paths. In Fig. 5, the virtual path Tr11 and the virtual path Tr12 do not interfere with the intrusion prohibited area, but the virtual path Tr13 interferes with the intrusion prohibited area A2. Therefore, after that, step S04 is executed for the virtual path Tr13. Figure 6 shows the final state of path generation in this plane. Here, the avoidance point P21 according to step S04 is set between the avoidance point P12 that caused the interference in FIG. 5 and the target position Goal. Since no interference has occurred in the new virtual routes Tr21 and Tr22, this completes the route generation process. As a result, Tr11, Tr12, Tr21 and Tr22 could be obtained as the hand paths of the robot manipulator from the initial position Start to the target position Goal.
【0012】
In the above, an example of route generation in the upper part of the plane of FIG. 3 has been described, but other routes can be generated in the upper part or the lower part of the plane by the same processing procedure. For example, it is possible to generate another virtual route that passes through Start P13 P14 Goal. A route of Start P15 P16 Goal is also conceivable, but in this case, the route passes through the intrusion prohibited area, so it is not adopted. That is, the present invention does not limit the range of path generation in a plane. Therefore, when a plurality of routes are generated in this way, the interference avoidance locus is selected based on the evaluation function, and one route is selected (S05). Specifically, first, the trajectory that causes interference is removed, and the one that optimizes the evaluation function is selected from the remaining ones. As this evaluation function, the distance of the trajectory is used, the distance between the manipulator arm and the obstacle, the number of avoidance points is used, or the evaluation function used is switched depending on the application to flexibly respond to various requirements. The trajectory can be selected.
【0013】
By the way, in order to think about three-dimensional interference avoidance, it may be necessary to think on multiple planes. When there are a plurality of the desired planes in this way, as shown in FIG. 2, a predetermined evaluation is performed and one route is selected from the routes obtained for each of the desired planes (S11 ~. S13) may be done (another embodiment of the present invention (second embodiment)). Hereinafter, the manual route planning method of the robot manipulator according to the second embodiment will be described in more detail. In Fig. 2, first, some typical planes are selected according to the application from the innumerable planes including the initial position and the target position of the hand of the robot manipulator (S11). This selection may be an arbitrary choice of the operator, or may be automatically selected according to a predetermined rule. The predetermined rule is, for example, horizontal + vertical + selecting a plane at a constant inclination interval with respect to a robot or an obstacle.
【0014】
Next, the interference avoidance locus on each plane is generated for some of the selected planes (S12). The specific content of step S12 is the same as that of steps S01 to S05 in the first embodiment method. Then, the interference avoidance locus is selected and determined based on the evaluation function from the interference avoidance loci generated on each plane (S13). The more planes are selected in step S11, the more candidates for the interference avoidance locus will be generated. As for this selection method, the same method as in step S05 in the first embodiment may be adopted. However, one may be selected from a plurality of locus groups at a time in step S13 without executing step S05. As described above, in any of the examples, the interference avoidance problem in three dimensions can be reduced to the path generation of the manipulator hand in the plane and solved. Therefore, in the present invention, the storage area used can be relatively reduced. This is because the configuration space in the conventional example is not handled, so an area for storing the free space is not required. In addition, the calculation time can be reduced. This is because there is no need for mapping calculations between the configuration space and the Cartesian coordinate space. Furthermore, while the amount of information handled is enormous when searching for a route in a three-dimensional space, the need for it is small. By limiting the search space to a plane, the amount of information handled can be reduced, and the amount of calculation can be reduced accordingly.
【0015】
Furthermore, it is possible to prevent the robot element that follows the generated path from interfering with the obstacle. This is because no route is generated in the intrusion prohibited area set around the obstacle based on the avoidance distance r between the robot manipulator hand and the obstacle. Furthermore, it is possible to prevent an unnecessarily detoured route from an obstacle. Route generation can be performed by setting some avoidance points, but since these avoidance points are set using the vertex information of the intrusion prohibited area that is interfering, the avoidance distance is larger than r and from obstacles. Because it never leaves. As a result, according to the present invention, in the problem of automatically generating a path in which the elements of the robot manipulator can move from the initial position to the target position while avoiding interference with surrounding interfering objects, it is practical in a relatively small storage area. It is possible to realize a high-speed route generation process with a large calculation time. In addition, a safe and efficient route can be obtained because the route does not interfere with obstacles and does not detour more than necessary. In the above embodiment, the generation of the hand path of the robot manipulator is described as an example of the applicable range, but the same applies to the elements of the robot manipulator, and the movement path of a moving object other than the robot is obtained. Is also applicable.
【0016】
[Effect of the invention]
Since the method for automatically generating the movement path of the robot manipulator according to the present invention is configured as described above, obstacle avoidance in the three-dimensional space is reduced to the two-dimensional plane obstacle avoidance problem in the present invention. As a result, high-speed route generation processing with practical calculation time can be realized in a relatively small storage area. In addition, routes are generated that do not interfere with obstacles and do not detour more than necessary. Further, when there are a plurality of the corrected routes, a predetermined evaluation can be performed to select the optimum one route. Further, when considering a plurality of the desired planes, it is safe and three-dimensionally considered by selecting one route by performing a predetermined evaluation from the routes obtained for each desired plane. An efficient route can be obtained. In this way, when a plurality of solutions are obtained, the narrowing down can be automatically performed, which can contribute to labor saving.
[Simple explanation of drawings]
[Figure 1]
The flow chart which shows the schematic structure of the automatic generation method of the movement path of the robot manipulator which concerns on embodiment and Example (1st Example) of this invention.
[Figure 2]
The flow chart which shows the schematic structure of the automatic generation method of the movement path of the robot manipulator which concerns on another Example (2nd Example) of this invention.
[Fig. 3]
Explanatory drawing which shows the specific example of the plane set in this 1st Example method.
[Fig. 4]
Explanatory drawing which shows the state after the execution of step S02 with respect to the said plane.
[Fig. 5]
Explanatory drawing which shows the state after the execution of step S04 with respect to the said plane.
[Fig. 6]
Explanatory drawing which shows the final state of movement path generation with respect to the said plane.
[Explanation of symbols]
S01 ... Cross-section creation process S02 ... Intrusion prohibited area creation / integration process S03 ... Interference check process between intrusion prohibited area and route S04 ... Subgoal setting process S05 ... Interference avoidance trajectory selection process
6 sheets
Sheet 1 Sheet 2 Sheet 3 Sheet 4 Sheet 5 Sheet 6
Every citation, both ways
| Document | Relation | Office | Cited during |
|---|---|---|---|
| US7110859B2 | Cited by | United States of America | Applicant |
| JP5127730A | Cites | Japan | – |
| JP4291403A | Cites | Japan | – |
| JP6324305A | Cites | Japan | – |
2 priority claims, no other members on record
Priority claims2
| Document | Office | Kind | Date |
|---|---|---|---|
| 18142995 | Japan | A | |
| JP19950181429 | – | – | – |
8 legal events, as the office reported them to INPADOC
Over the term
Point at a mark for the eventEvents
| Event | Code | |
|---|---|---|
| Cancellation because of completion of termEXPY | EXPY | |
| Renewal fee payment (event date is renewal date of database)FPAY | FPAY | |
| Renewal fee payment (event date is renewal date of database)FPAY | FPAY | |
| Renewal fee payment (event date is renewal date of database)FPAY | FPAY | |
| Renewal fee payment (event date is renewal date of database)FPAY | FPAY | |
| Renewal fee payment (event date is renewal date of database)FPAY | FPAY | |
| Renewal fee payment (event date is renewal date of database)FPAY | FPAY | |
| Renewal fee payment (event date is renewal date of database)FPAY | FPAY |
Numbers
- Publication
- 2875498
- Publication, DOCDB
- 2875498
- Publication, EPODOC
- JP2875498B
- Application
- 7181429
- Application, DOCDB
- 18142995
- Application, EPODOC
- JP19950181429
Titles2
- Japanese
- 【発明の名称】ロボットマニピュレータの移動経路の自動生成方法
- English
- PROBLEM TO BE SOLVED: To automatically generate a movement path of a robot manipulator.
Classification
- IPC, 4
- B25J9 10
- B25J9 16
- B25J19 06
- G05B19 4093
