Method For Parallelized Simulation Of The Movement Of A Machine Component In Or Near A Work Object

Abstract

The present invention relates to a method for parallelized simulation of the movement of a machine component in or near a work object, wherein: a planned trajectory of the machine component is divided into several trajectory portions; a simulation of at least a first trajectory portion and a second trajectory portion is performed at least partly in parallel yielding simulation results, wherein the simulation comprises determining incidents, preferably collisions, along the trajectory portions; at least the simulation results of the first trajectory portion and the second trajectory portion are merged yielding a merged simulation result; and the merged simulation result is outputted.

Claims

1. A method for parallelized simulation of the movement of a machine component in or near a work object, wherein: a planned trajectory of the machine component is divided into several trajectory portions; a simulation of at least a first trajectory portion and a second trajectory portion is performed at least partly in parallel yielding simulation results, wherein the simulation comprises determining incidents along the trajectory portions; at least the simulation results of the first trajectory portion and the second trajectory portion are merged yielding a merged simulation result, wherein merging comprises that the simulation results of the first trajectory portion and the second trajectory portion are combined and incidents determined along the second trajectory portion are reassessed based on the simulation results of the first trajectory portion, wherein the incident is removed from or not taken over into the merged simulation results if the reassessment indicates that the incident cannot occur; and the merged simulation result is outputted.

2. The method of claim 1, wherein the determined incidents are collisions along the trajectory portions.

3. The method of claim 1, wherein the planned trajectory is divided into trajectory portions that follow one another in time.

4. The method of claim 1, wherein the merging of the simulation results only takes place in a predefined space surrounding the trajectory portions.

5. The method of claim 1, wherein the space of at least one of the work object and the planned trajectory is represented using voxels.

6. The method of claim 5, wherein the space of the planned trajectory and the space within a predetermined distance from the planned trajectory is represented using smaller voxels than the remaining space.

7. The method of claim 5, wherein each voxel is assigned to at least one state, wherein the state comprises information if the voxel contains material of the work object.

8. The method of claim 5, wherein the merged simulation result comprises a result value for each or at least some of the voxels.

9. The method of claim 8, wherein the result value is the state of a respective voxel after comparing the simulation results of the first and second trajectory portions.

10. The method of claim 1, wherein an incident is a collision of the machine component with the work object.

11. The method of claim 10, wherein in case a collision is determined, then a spatial orientation of the machine component is modified in order to avoid the collision.

12. The method of claim 1, wherein a first merged simulation result and a second merged simulation result are merged to create a new merged simulation result, wherein: merging comprises that the first merged simulation result and the second merged simulation result are combined and incidents stored in the second merged simulation result are reassessed based on the first merged simulation result, wherein the incident is removed from or not taken over into the new merged simulation results if the reassessment indicates that the incident cannot occur; and the new merged simulation result is outputted.

13. The method of claim 1, wherein the simulation of the first trajectory portion and the second trajectory portion is performed on at least two different CPUs and/or GPUs.

14. The method of claim 1, wherein the simulation of the first trajectory portion and the second trajectory portion is performed on different processor cores.

15. The method of claim 1, wherein the machine component is a milling head comprising a milling tool.

16. The method of claim 1, wherein prior to the start of the simulation of the first trajectory portion and the second trajectory portion, dimensions of the work object and/or the machine component are measured.

17. The method of claim 1, wherein the merged simulation result is outputted to an industrial machine, wherein the industrial machine performs work on the work object in accordance with the simulation results.

18. An industrial machine, wherein the industrial machine executes a method for parallelized simulation of the movement of a machine component in or near a work object, wherein: a planned trajectory of the machine component is divided into several trajectory portions; a simulation of at least a first trajectory portion and a second trajectory portion is performed at least partly in parallel yielding simulation results, wherein the simulation comprises determining incidents along the trajectory portions; at least the simulation results of the first trajectory portion and the second trajectory portion are merged yielding a merged simulation result, wherein merging comprises that the simulation results of the first trajectory portion and the second trajectory portion are combined and incidents determined along the second trajectory portion are reassessed based on the simulation results of the first trajectory portion, wherein the incident is removed from or not taken over into the merged simulation results if the reassessment indicates that the incident cannot occur; and the merged simulation result is outputted.

19. The industrial machine of claim 18, wherein the industrial machine is a milling machine.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The invention will be explained in the following in detail by means of exemplary embodiments and with reference to the figures.

[0061] FIG. 1 shows a schematic view of an industrial machine and a work object;

[0062] FIG. 2 shows a voxelization of the work object; and

[0063] FIG. 3 schematically shows a simulation process.

DETAILED DESCRIPTION OF THE INVENTION

[0064] FIG. 1 shows an industrial machine, namely a milling machine 10. The milling machine 10 comprises a milling head 12 having a milling tool 14. The milling head 12 and the milling tool 14 can be considered as a machine component. The milling head 12 can be moved in 3D-space, which results in a corresponding movement of the milling tool 14.

[0065] The milling machine 10 comprises a laser scanner 16 which measures the dimensions of a work object 18. The laser scanner 16 also measures the dimensions of the milling head 12 and the milling tool 14. Alternatively, these dimensions may be known.

[0066] The milling machine 10 further comprises a control unit 20 having a GPU with a plurality of processor cores (not shown).

[0067] The control unit 20 is adapted to execute the method as described herein.

[0068] Firstly, based on the measurement data of the work object 18, a voxelization of the work object 18 is performed, as is shown in FIG. 2. During the voxelization the work object 18 is divided into cubic spaces, i.e. voxels 22. It is noted that in FIG. 2 the 3D character of the voxels is only slightly indicated for a better overview. During the voxelization some smaller voxels 22a are created along a trajectory (FIG. 3). The smaller voxels 22a allow for a better spatial resolution and thus a more accurate simulation result.

[0069] After the voxelization is performed, the control unit 20 starts a simulation, as is shown in FIG. 3. For the simulation a planned trajectory 24 is determined e.g. from a programming of the milling machine 10. The planned trajectory 24 indicates the path along which the end of the milling tool 14 is planned to be moved. The planned trajectory 24 is then divided into a first, second, third and fourth trajectory portion 26a, 26b, 26c, 26d. For each of the trajectory portions 26, a separate simulation 28 is performed on a different processor core of the GPU. During the simulation 28 it is determined if an incident, e.g. a collision or an overheating of the milling head 12 or the milling tool 14 will occur. Thereby, for each simulation 28, simulation results are produced. In a merging step 30 the simulation results are merged into a merged simulation result 32, wherein it is determined for each voxel 22 if an incident is present. If an incident is present, the incident is reassessed based on the corresponding simulation results of an earlier trajectory portion (e.g. simulation results for trajectory portion 26b are reassessed based on the simulation results of trajectory portion 26a). The incident is then removed or not even taken over into the merged simulation result 32, if the reassessment indicates that the incident cannot occur.

[0070] The merged simulation results 32 are then merged in a further merging step 30 into a final simulation result 34. If the final simulation result 34 does not show any incidents/collisions, the planned trajectory can be used to work on the real work object 18 using the real milling head 12 and the real milling tool 14. The real work can be controlled by control unit 20. If incidents persist, the planned trajectory 24 can be modified and the simulation can be repeated.

[0071] Due to the parallelization of the simulation of the different trajectory portions 26 (based on different GPU cores) it is possible to dramatically reduce the time required for the simulation of the whole planned trajectory 24. Therefore, the throughput of the milling machine 10 can be increased.

LIST OF REFERENCE NUMERALS

[0072] 10 milling machine [0073] 12 milling head [0074] 14 milling tool [0075] 16 laser scanner [0076] 18 work object [0077] 20 control unit [0078] 22, 22a voxel [0079] 24 planned trajectory [0080] 26a-d trajectory portion [0081] 28 simulation [0082] 30 merging step [0083] 32 merged simulation result [0084] 34 final simulation result