Methods for control of post-design free form deposition processes or joining processes are described that utilize machine learning algorithms to improve fabrication outcomes. The machine learning algorithms use real-time object property data from one or more sensors as input, and are trained using training data sets that comprise: i) past process simulation data, past process characterization data, past in-process physical inspection data, or past post-build physical inspection data, for a plurality of objects that comprise at least one object that is different from the object to be fabricated; and ii) training data generated through a repetitive process of randomly choosing values for each of one or more input process control parameters and scoring adjustments to process control parameters as leading to either undesirable or desirable outcomes, the outcomes based respectively on the presence or absence of defects detected in a fabricated object arising from the process control parameter adjustments.

In a track determination device, a CAD data acquisition unit acquires shape data that represents a shape of a three-dimensional formed object. A deposition direction setting unit generates control information for controlling a lamination device that laminates the molten metal in order to form a formed object, based on the shape data acquired by the CAD data acquisition unit. The control information is information indicating at least a specific deposition direction of the molten metal such that an error between a first deposition position set in advance and a second deposition position in accordance with an actual laminated state is reduced. A control program output unit outputs the control information generated by the deposition direction setting unit.

A card manufacturing system includes a locating device and a separation device. A laminate sheet comprising a plurality of cards is received by the locating device and is imaged using first and second imaging modalities. The first imaging modality identifies a location of each of the plurality of information carrying cards within the laminate sheet and the second imaging modality images at least one graphic formed on a surface of the laminate sheet. A position of the at least one graphic with respect to at least one information carrying card is determined and the plurality of cards are removed from the laminate sheet using information corresponding to the location of each of the plurality of information carrying cards when the position of the at least one graphic with respect to the information carrying cards is within a predetermined range.

The invention provides a sliced image processing method including following steps: analyzing a sliced object in a sliced image to determine whether the sliced object has a first contour line segment and a nearest second contour line segment, where the second contour line segment is located within a region encircled by the first contour line segment; determining whether vector directions of the first contour line segment and the second contour line segment are opposite when the sliced object has the first contour line segment and the second contour line segment, and correcting the vector direction of at least one of the first contour line segment and the second contour line segment when the vector directions of the first contour line segment and the second contour line segment are not opposite.

Methods, systems, and apparatus, including medium-encoded computer program products, for designing three dimensional lattice structures include, in one aspect, a method including: creating nodes in a plane normal to an axis in accordance with a spiral, wherein proper subsets of the nodes occur at successive radii positions away from the axis in the plane normal to the axis; repositioning every other one of the proper subsets, from at least a portion of the nodes, in a direction in 3D space along the axis; creating a three dimensional (3D) structure in the 3D space, the 3D structure comprising beams placed between the repositioned and non-repositioned proper subsets; duplicating the 3D structure one or more times to form a lattice in the 3D space; and selecting at least a portion of the lattice for inclusion in a 3D model.

Disclosed herein are machine learning-based methods and systems for automated object defect classification and adaptive, real-time control of additive manufacturing and/or welding processes.

An information processing apparatus for generating a toolpath of a three-dimensional fabrication apparatus includes circuitry to slice three-dimensional data of a three-dimensional object into a plurality of parallel planes to generate a first image slice and a second image slice, extract a first contour in the first image slice and a second contour in the second image slice, the first contour dividing the first image slice into an area inside the three-dimensional object and an area outside the three-dimensional object, the second contour dividing the second image slice into the area inside the three-dimensional object and the area outside the three-dimensional object, and correct a third contour of a third image slice disposed between the first image slice and the second image slice based on the first contour of the first image slice and the second contour of the second image slice.

A card manufacturing system includes a locating device and a separation device. A laminate sheet comprising a plurality of cards is received by the locating device and is imaged using first and second imaging modalities. The first imaging modality identifies a location of each of the plurality of information carrying cards within the laminate sheet and the second imaging modality images at least one graphic formed on a surface of the laminate sheet. A position of the at least one graphic with respect to at least one information carrying card is determined and the plurality of cards are removed from the laminate sheet using information corresponding to the location of each of the plurality of information carrying cards when the position of the at least one graphic with respect to the information carrying cards is within a predetermined range.

A causal relationship model building system includes a computer which processes information for building a causal relationship model relating to a manufacturing flow of an object to be controlled. The computer builds the causal relationship model by using monitor data representing a state of each of a plurality of steps of the manufacturing flow, and quality data as a result of an inspection step, and specifies an allowable range of the monitor data so as to satisfy a target value of the quality data, by using the causal relationship model and the target value, from prediction based on a causal relationship between a plurality of pieces of the monitor data. The computer graphically displays information including the causal relationship model and the allowable range of the monitor data on a display screen.

Methods for control of post-design free form deposition processes or joining processes are described that utilize machine learning algorithms to improve fabrication outcomes. The machine learning algorithms use real-time object property data from one or more sensors as input, and are trained using training data sets that comprise: i) past process simulation data, past process characterization data, past in-process physical inspection data, or past post-build physical inspection data, for a plurality of objects that comprise at least one object that is different from the object to be fabricated; and ii) training data generated through a repetitive process of randomly choosing values for each of one or more input process control parameters and scoring adjustments to process control parameters as leading to either undesirable or desirable outcomes, the outcomes based respectively on the presence or absence of defects detected in a fabricated object arising from the process control parameter adjustments.