Back to results
US 12,761,628 B2Granted
Object packings with volume subsets
AssigneesPeridot Print LLC, Instituto Atlantico
InventorsJuan Carlos Catana Salazar, Alyne Gomes Soares Cantal, Andre Luiz Lima Pedro, Scott Alan White, Jun Zeng
Life of this patent
Estimating the term…
Abstract
Examples of methods for object packings are described herein. In some examples, a method includes loading a first object in a first subset of a volume. In some examples, the method includes determining, using a genetic procedure, an arrangement of second objects in a second subset of the volume to produce a packing that includes the first object and the second objects.
Claims14
A method comprising:
- performing, by a processor, a process for a plurality of iterations, wherein the process comprises:
- generating chromosomes in a genetic procedure,
- wherein the chromosomes respectively indicate different packings of objects in a build volume of an additive manufacturing device, the objects including one or more first objects and a plurality of second objects,
- and wherein the first objects have predefined static positions within the build volume that remain fixed over the different packings of the chromosomes during chromosome generation across the iterations, and the second objects have dynamic positions within the build volume that are modifiable during the chromosome generation in each iteration;
- and
- determining fitness measures corresponding to the chromosomes,
- wherein the fitness measures determined in each iteration other than a last iteration are used during the chromosome generation in a subsequent iteration;
- selecting, by the processor, one of the chromosomes based on at least the fitness measures;
- and
- causing the additive manufacturing device to additively manufacture the objects in the build volume in accordance with the different packing of the objects indicated by the selected chromosome,
- wherein the genetic procedure includes preserving elite chromosomes across the iterations, generating mutation chromosomes by mutating the elite chromosomes generated in a previous iteration, and generating crossover chromosomes by crossing over pairs of the elite chromosomes,
- and wherein each pair includes one of the elite chromosomes, and either another of the elite chromosomes or a non-elite chromosome generated the previous iteration.
9 more elements
A system comprising:
- a processor;
- and
- a memory storing instructions executable by the processor to perform processing comprising:
- performing a process for a plurality of iterations, wherein the process comprises:
- generating chromosomes in a genetic procedure,
- wherein the chromosomes respectively indicate different packings of objects in a build volume of an additive manufacturing device, the objects including one or more first objects and a plurality of second objects,
- and wherein the first objects have predefined static positions within the build volume that remain fixed over the different packings of the chromosomes during chromosome generation across the iterations, and the second objects have dynamic positions within the build volume that are modifiable during the chromosome generation in each iteration;
- and
- determining fitness measures corresponding to the chromosomes,
- wherein the fitness measures determined in each iteration other than a last iteration are used during the chromosome generation in a subsequent iteration;
- selecting one of the chromosomes based on at least the fitness measures;
- and
- causing the additive manufacturing device to additively manufacture the objects in the build volume in accordance with the different packing of the objects indicated by the selected chromosome,
- wherein the genetic procedure includes preserving elite chromosomes across the iterations, and either or both of:
- generating mutation chromosomes by mutating the elite chromosomes generated in a previous iteration;
- and
- generating crossover chromosomes by crossing over pairs of the elite chromosomes, each pair including one of the elite chromosomes, and either another of the elite chromosomes or a non-elite chromosome generated the previous iteration.
14 more elements
A non-transitory computer-readable storage medium storing instructions executable by a processor to perform processing comprising:
- performing a process for a plurality of iterations, wherein the process comprises:
- generating chromosomes in a genetic procedure,
- wherein the chromosomes respectively indicate different packings of objects in a build volume of an additive manufacturing device, the objects including one or more first objects and a plurality of second objects,
- and wherein the first objects have predefined static positions within the build volume that remain fixed over the different packings of the chromosomes during chromosome generation across the iterations, and the second objects have dynamic positions within the build volume that are modifiable during the chromosome generation in each iteration;
- and
- determining fitness measures corresponding to the chromosomes,
- wherein the fitness measures determined in each iteration other than a last iteration are used during the chromosome generation in a subsequent iteration;
- selecting, by the processor, one of the chromosomes based on at least the fitness measures;
- and
- causing the additive manufacturing device to additively manufacture the objects in the build volume in accordance with the different packing of the objects indicated by the selected chromosome,
- wherein, for each iteration after a first iteration, generating the chromosomes in the genetic procedure comprises:
- preserving at least some of the chromosomes generated in a previous iteration for which the fitness measures are greater than a threshold, as elite chromosomes:
- generating mutation chromosomes, by mutating the elite chromosomes;
- and
- generating crossover chromosomes by either or both of:
- crossing two of the elite chromosomes with one another:
- crossing one of the elite chromosomes with one of the chromosomes generated in the previous iteration for which the fitness measures are less than the threshold.
14 more elements
Dependencies are inferred from the claim text.