Method for optimizing geometric path of robotic device bypassing obstacle
By automatically optimizing the mixing zone parameters of the robotic equipment and generating collision-free and efficient paths, the problem of inefficient path optimization in existing technologies is solved, achieving shorter, smoother robotic motion and longer equipment life.
Patent Information
- Application Number
- CN202380093671.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-02-21
- Publication Date
- 2025-09-19
AI Technical Summary
In the existing technology, users need to manually adjust the mixing zone parameters through trial and error to avoid collisions between the robot equipment and obstacles, resulting in inefficient path optimization and performance limitations.
A method is provided to automatically find collision-free and efficient paths according to an optimal cost function by automatically optimizing blend zone parameters on a predefined geometric path of a robotic device using nonlinear functions to generate curves in joint or Cartesian space.
Significantly reduce robot motion time and mechanical stress, improve movement smoothness and energy efficiency, extend robot equipment life, and reduce user effort in optimizing parameters.
Smart Images
Figure CN120677036A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a method for optimizing a geometric path of a robotic device to circumvent obstacles. Background Art
[0002] In non-productive movements of a robotic device, the non-productive movement geometric path generally does not need to follow a predefined geometry. Such a path can consist of a number of so-called through-points, which are usually defined by the user, so that the robotic device moves between or along the productive parts of the task without colliding with itself or the rest of the cell. By not forcing the robotic device to stop at through-points, but taking shortcuts, the movements can be made shorter and smoother, resulting in reduced movement time and less stress on the robot. Such shortcuts are defined by a blending zone parameter for a given target, which specifies, for example, the maximum distance that the robot's TCP (= Tool Center Point) can deviate from the target. With a larger blending zone parameter, the path of the robotic device can become smoother and shorter.
[0003] However, current state-of-the-art techniques often require users to perform a trial-and-error process for all via points to find the optimal blend zone parameters that are large but do not cause collisions. This is typically done by moving a robotic device along potential geometric paths in simulation or the real world to check whether the resulting geometric paths will cause collisions.
[0004] This existing approach often results in the user selecting a small region for all through-points of the geometry or motion path, thereby unnecessarily limiting the performance of the robotic device and the smoothness of the motion.
[0005] It is necessary to resolve these issues. Summary of the Invention
[0006] Therefore, it would be advantageous to provide an improved concept for automatically generating optimal collision-free blending zone parameters for a predefined or given geometry or motion path of a robotic device.
[0007] The objects of the invention are solved by the subject-matter of the independent claims, wherein further embodiments are incorporated in the dependent claims.
[0008] In a first aspect of the present invention, there is provided a method for optimizing a geometric path of a robotic device to circumvent obstacles, the method comprising the following steps:
[0009] providing a geometric path for the robotic device, wherein the geometric path includes at least one target location point of the robotic device, the target location point being defined as an intersection of a first segment and a second segment forming the geometric path;
[0010] defining at least a first blending zone around at least one target location point, wherein the at least first blending zone is defined as a curve of the geometric path that blends the first segment into the second segment, and wherein a size of the curve is defined by a starting position of a starting point on the first segment of the geometric path and an ending position of the curve of an ending point on the second segment of the geometric path;
[0011] The size of at least the first mixing zone is optimized by shifting the starting point along the first segment and the ending point along the second segment to find an optimal mixing zone for at least the first mixing zone, the optimal mixing zone corresponding to an optimal cost function according to a defined criterion between the curve starting point of the first segment and the curve ending point of the second segment on the geometric path, wherein the optimal mixing zone for at least the first mixing zone satisfies at least one predefined condition.
[0012] In other words, the core idea behind the present invention is to automatically find the optimal zone parameters for a given set of through-hole points on a predefined geometric path of a robotic device, so as to ensure collision-free and optimal motion of the robotic device along the nominal path of the robotic device between the modeled robotic links or accessories and obstacles. These through-hole points can be derived from a user program or from an algorithm such as collision-free path planning. The solution requires a collision volume representation of the robotic links, robotic accessories, and obstacles. The collision geometry can be provided, for example, as a CAD model in a simulation software tool or via a point cloud from a perception system.
[0013] In the context of the present invention, a mixing zone is defined by a first segment of a geometric path mixing into a second segment of the geometric path.
[0014] The main advantage achieved through this approach is that the optimal collision-free zone parameters for a given path of the robotic device can be automatically found. This can significantly reduce the robot's movement time and mechanical stress, without leaving the user with the effort of manipulating the optimization zone parameters during unproductive movements. By providing an optimal zone for the robotic device's movement, the robotic device's movements can be smoother and consume less energy. This, in turn, places less mechanical stress on the robotic device's mechanical structure, resulting in an extended lifespan and a reduction in service intervals for the robotic device.
[0015] According to an example, at least the first blending zone is a non-linear function generating a curve in the joint space or in the Cartesian space.The advantage achieved is an efficient performance of the robotic device.
[0016] According to one example, the optimization step is performed in an iterative manner. The advantage achieved is that the optimal mixing region can be found in an efficient manner.
[0017] According to one example, the optimal cost function according to the defined criterion is the shortest path between a starting point and an end point of a curve. The advantage achieved is that an optimal path for the robotic device can be generated in an efficient manner.
[0018] According to one example, the at least one predefined condition is at least one of the following: a collision-free path around an obstacle while maintaining a minimum distance from the robotic device to the obstacle, a collision-free movement of the robotic device with itself, a minimum distance from a curve to an obstacle, a minimum length of a geometric path, a defined cycle time of the robotic device, or a predefined energy consumption. The advantage achieved is that the performance of the robotic device can be optimized in an efficient manner and adapted to changing technical environments or changing requirements placed on the robotic device.
[0019] According to one example, the optimization step is performed by a discrete search or a continuous search. The advantage achieved is that the optimal mixing zone for the robotic device is found in an efficient manner.
[0020] According to one example, the optimization step is performed by using a machine learning model. The advantage achieved is that the optimal mixing zone is found for the robotic device in an efficient manner.
[0021] In a second aspect of the present invention, a computer is provided. The computer includes a processor configured to execute the method of the aforementioned aspect.
[0022] In a third aspect of the present invention, a computer program product is provided. The computer program product comprises instructions. When the program is executed by a processor of a computer, the instructions cause the computer to perform the method of any one of the first and second aspects.
[0023] In a fourth aspect of the invention, a machine-readable data medium and / or download product comprises the computer program of the third aspect. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Exemplary embodiments will be described below with reference to the following drawings:
[0025] Figure 1 A schematic flow chart illustrating the method of the present invention; and
[0026] Figures 2a to 2d A schematic example of finding an optimal mixing zone for a robotic device along a given path in an iterative manner according to the method of the present invention is shown. DETAILED DESCRIPTION
[0027] Figure 1 2 , a schematic flow chart of a method 100 for optimizing a geometric path 10 for a robotic device to circumvent an obstacle 60 is illustrated, which includes the following steps of the present invention.
[0028] In a first step 102 , a geometric path 10 of the robotic device is provided, wherein the geometric path 10 comprises at least one target position point 2 of the robotic device, which is defined as an intersection of a first segment 11 and a second segment 12 forming the geometric path 10 .
[0029] In a second step 104, at least a first mixing zone 20 is defined around at least one target position point 2, wherein the at least first mixing zone 20 is defined as a curve 21 of the geometric path 10, which mixes the first segment 11 into the second segment 12, and wherein the size of the curve 21 is defined by a starting position of a starting point 23 on the first segment 11 of the geometric path 10 and an end position of the curve of an end point 25 of the second segment 12 of the geometric path 10.
[0030] In this context, it should be noted that the shifting of the starting point 23 and the end point 25 also changes the shape of the curve 21. The curve 21 is not necessarily a circular curve. At least the first mixing zone 20 is a nonlinear function that generates a curve in joint space or Cartesian space.
[0031] In a third step 106, the size of at least the first mixing zone 20 is optimized by shifting the starting point 23 along the first segment 11 and the end point 25 along the second segment 12 to find an optimal mixing zone of at least the first mixing zone 20, which corresponds to an optimal cost function according to a defined criterion between the curve starting point 23 of the first segment 11 and the curve end point 25 of the second segment 12 on the geometric path 10, wherein the optimal mixing zone of at least the first mixing zone 20 satisfies at least one predefined condition.
[0032] In the context of step 106, it should also be noted that since the robotic device moves continuously through the mixing zone 20 and calculating costs and constraints in addition to distance in only at least one segment of the mixing zone 20 is a challenge, the present invention focuses on automatically finding the optimal zone that corresponds to the minimum motion cost and satisfies at least one predefined constraint in the motion or movement of the robotic device from the starting point 23 of the segment 11 to the ending point 25 of the segment 12.
[0033] Preferably, the step 106 of optimizing and finding the optimal mixing zone of the robotic device is performed in an iterative manner.
[0034] Preferably, the optimization step 106 is performed by discrete search or continuous search.
[0035] At least the first mixing zone 20 according to FIG. 2 is a nonlinear function which generates a curve in the joint space or in the Cartesian space.
[0036] According to the defined criteria, the optimal cost function is the shortest path 28 between the starting point 23 and the end point 25. Non-limiting examples of such defined criteria may be the shortest path for the robotic device to bypass an obstacle, the fastest path for the robotic device to bypass an obstacle, the least electromechanical energy consumed by the robotic device to bypass an obstacle, the longest robot life of the robotic device, the least robot tool speed change for the robotic device to bypass an obstacle.
[0037] Non-limiting examples of at least predefined conditions are a collision-free path around the obstacle 60 maintaining a minimum distance of the robotic device to the obstacle 60, a collision-free movement of the robotic device with itself, a minimum distance of the curve 21 to the obstacle 60, a minimum length of the geometric path 10, a defined cycle time or a predefined energy consumption of the robotic device, an optimal robot speed, an optimal robot acceleration, a definition of the robotic device in joint space or Cartesian space, and at least one of an optimal force or torque.
[0038] Figures 2a to 2d A schematic example of finding an optimal mixing zone for a robotic device along a given path in an iterative manner according to the method of the present invention is shown.
[0039] To avoid repetition, refer to the previous section for method steps that are performed in an iterative manner.
[0040] Figure 2a A geometric path 10 is shown which is predefined with through-points T1, T2, T3 and T4 along an obstacle 60. In through-point T2, target position point 2 and in through-point T3, target position point 3 are indicated as examples. Figure 2a In the example, the optimal mixing zone has not yet been found, as the process is about to begin. Geometric path 10 consists of first segments 11 and 14 and second segments 12 and 15. First segment 11 and second segment 12 meet or intersect at through-hole point T2. First segment 14 and second segment 15 meet or intersect at through-hole point T3.
[0041] Figure 2b The first iterative step of the method 100 is shown: defining a first blending zone 20 around the target location point 2. The first blending zone 20 is defined as a curve 21 of the geometric path 10 that blends the first segment 11 into the second segment 12. The size of the curve 21 is defined by the starting position of a starting point 23 on the first segment 11 of the geometric path 10 and the ending position of the curve at an ending point 25 of the second segment 12 of the geometric path 10.
[0042] For the through hole point T3 with the target position point 3, a second mixed region 30 is generated (see Figure 2d). The second blending zone 30 is defined as a curve of the geometric path 10 that blends the first segment 14 of the geometric path 10 into the second segment 15. The size of the curve is defined by the starting position of the starting point 31 on the first segment 14 of the geometric path 10 and the ending position of the curve at the ending point 33 of the second segment 15 of the geometric path 10 (see Figure 2d ).
[0043] Figure 2c A second iterative step of the method 100 regarding the first mixing zone 20 is shown: optimizing the size of the first mixing zone 20 by shifting the starting point 23 along the first segment 11 of the geometric path 10 and the ending point 25 along the second segment 12 to find an optimal mixing zone for at least the first mixing zone 20. Figure 2c The optimal mixing zone of the embodiment is the shortcut path 27.
[0044] Figure 2d The third iteration of method 100 is shown:
[0045] For the first mixing zone 20 , shifting the start point 23 and the end point 25 along the corresponding segment of the path 10 results in an optimal shortcut path 28 or optimal mixing zone for the first mixing zone 20 .
[0046] The optimal mixing zone corresponds to the optimal cost function between the curve starting point 23 of the first segment 11 and the curve ending point 25 of the second segment 12 on the geometric path 10 according to a defined criterion, wherein the optimal mixing zone of at least the first mixing zone 20 satisfies at least one predefined condition as described in detail above. One predefined condition can be a collision-free movement of the robotic device around or along the obstacle 60, wherein the distance between the robotic device and the obstacle 60 is minimized.
[0047] For the second hybrid zone 30, shifting the start point 31 and end point 33 along the corresponding segment of path 10 yields another optimal shortcut path to approach obstacle 60. In this scenario, when the predefined condition is collision-free movement of the robotic device around the obstacle, further optimization is not possible. Therefore, the optimal zone parameters for the second hybrid zone 30 are ultimately found.
[0048] In the context of the present invention, the following are stated to explain the present invention in more detail:
[0049] The method of the present invention is based on the following assumptions:
[0050] First, the robot links, robot accessories (e.g., tools), and the robot's surroundings (e.g., obstacles) are modeled in a way that allows checking the robot's collision state in a given configuration. Such models can be generated, for example, from CAD models in a RobotStudio simulation or from point clouds from a perception system.
[0051] Second, the obstacles do not move when the robotic device performs motion.
[0052] Given the above assumptions, collisions can be checked along a geometric path. The path can be in joint space or Cartesian space.
[0053] The input path to the zone parameterization algorithm is a set of joint or Cartesian space targets and the desired interpolation method between these through-hole points. Given the input or defined geometric path, collision checking functionality and the availability of the robot's kinematic mapping and zone interpolation functions, the zones of all through-hole points are expanded until a collision is found in a through-hole point or the maximum zone parameter allowed by the target is reached (see Figure 2). Preferably, the search is performed in a breadth-first manner. If the search is performed in a depth-first manner, when the targets are too close, the earlier targets can reach a large mixed zone size, leaving little room for the expansion of subsequent targets. The algorithm also prevents the mixed zone size from causing overlap between subsequent targets. One implementation is that the algorithm aims to achieve a maximum zone size, which is equivalent to minimizing the path length.
[0054] However, any other criterion that may be affected by the mixing zone parameters (cycle time, energy, etc.) can be used as a "cost" in the search. The search can be implemented as a discrete search with fixed or variable step size or any other search method. The automatic parameterization of the collision-free zone provided by the method of the present invention makes it possible to utilize the full potential of the zone for non-productive motions with minimal effort on the part of the user in offline programming, where the robot cell has already been modeled. However, the method of the present invention can also be applied in online scenarios.
[0055] Reference numerals:
[0056] 100 methods
[0057] 102 offers
[0058] 104 Definition
[0059] 106 Optimization
[0060] 2, 3 Target location points
[0061] 10 Geometric Path
[0062] 11, 14, first paragraph
[0063] 12, 15, second paragraph
[0064] 20 First Mixed Zone
[0065] 21 Curve
[0066] 30 Second Mixed Zone
[0067] 23, 31 starting points
[0068] 25, 33 end points
[0069] 27, 28 shortcut paths
[0070] 60 obstacles
Claims
1. A method (100) for optimizing a geometric path (10) for a robotic device to navigate around an obstacle (60), comprising the following steps: providing (102) the geometric path (10) for the robotic device, wherein the geometric path (10) includes at least one target position point (2) of the robotic device, the target position point being defined as an intersection of a first segment (11) and a second segment (12) forming the geometric path (10); defining (104) at least a first blending zone (20) around the at least one target location point (2), wherein the at least first blending zone (20) is defined as a curve (21) of the geometric path (10), the curve blending the first segment (11) into the second segment (12), and wherein the size of the curve (21) is defined by a starting position of a starting point (23) on the first segment (11) of the geometric path (10) and an ending position of the curve of an ending point (25) of the second segment (12) of the geometric path (10); The size of the at least first mixing zone (20) is optimized (106) by shifting the starting point (23) along the first segment (11) and the ending point (25) along the second segment (12) to find an optimal mixing zone of the at least first mixing zone (20), the optimal mixing zone corresponding to an optimal cost function according to a defined criterion between the starting point (23) of the curve of the first segment (11) and the ending point (25) of the curve of the second segment (12) on the geometric path (10), wherein the optimal mixing zone of the at least first mixing zone (20) satisfies at least one predefined condition.
2. The method (100) of claim 1, wherein the at least first mixing zone (20) is a nonlinear function that generates a curve in joint space or Cartesian space.
3. The method (100) according to any one of the preceding claims, wherein the step of optimizing (106) is performed in an iterative manner.
4. The method (100) according to any one of the preceding claims, wherein the optimal cost function according to the defined criterion is the shortest path (28) between the curve starting point (23) and the end point (25).
5. The method (100) according to any one of the preceding claims, wherein the at least predefined condition is at least one of: a collision-free path around the obstacle (60) while maintaining a minimum distance from the robotic device to the obstacle (60), a collision-free movement of the robotic device with itself, a minimum distance from the curve (21) to the obstacle (60), a minimum length of the geometric path (10), a defined cycle time or a predefined energy consumption of the robotic device.
6. The method (100) according to any one of the preceding claims, wherein the step of optimizing (106) is performed by a discrete search or a continuous search.
7. The method (100) according to any one of the preceding claims, wherein the step of optimizing (106) is performed by using a machine learning model.
8. A computer comprising a processor configured to perform the method according to any one of the preceding claims 1 to 7.
9. A computer program product comprising instructions which, when executed by a processor of a computer, cause the computer to perform the method according to any one of claims 1 to 7.
10. A machine-readable data medium and / or download product comprising a computer program according to claim 9.