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 laser machining apparatus, a scanner is configured to scan a laser beam emitted from a laser beam emission device. A first part of a workpiece is exposed through an opening formed in the workpiece. A controller is configured to perform: acquiring shape data indicative of a shape of the workpiece; acquiring machining pattern data indicative of a machining pattern to be machined on the first part; acquiring a length of the machining pattern based on the machining pattern data; calculating an unmachinable position on a setting surface using the length and the shape data, the unmachinable position resulting from a second part of the workpiece hindering the laser beam reaching the first part, at least a part of the machining pattern being unmachinable on the first part in a state where the workpiece is set on the unmachinable position; and displaying the unmachinable position on a display.

This robot system includes a robot, a tool which is attached to the robot, an imaging device, and a controller which controls the robot and the tool, and the controller is configured so that a distal end portion of the robot is placed at a plurality of calibration positions, the tool is operated at each of the calibration positions, images of light irradiated to an object or a trace left on the object due to tool operation processing are captured by the imaging device, and calibration is performed based on comparison between irradiation positions of the light or positions of the traces, which are in an image captured by the imaging device, and predetermined reference positions.

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.

A method for simulating laser machining of a material by a laser machining system comprising the following steps: providing a central unit with: information relating to the material to be machined: delta , threshold fluence, incubation coefficient S, complex refractive index n+ik, information about the laser processing system: information relating to a polarization, pulse energy E.sub.P, diameter of said machining laser beam at a focal point w, order of a Gaussian p, pulse repetition rate PRR n, wavelength; determining with said central unit on the basis of the information relating to said material to be machined and the laser machining system, a machining profile in two dimensions corresponding to the simulation of a machining of said material to be machined with said laser machining system.

A laser processing machine includes a processing head having a sensor, and a control unit to which a signal from the sensor is inputted. The control unit determines, by the signal, whether a drive shaft for the processing head needs to be adjusted. When the drive shaft is a state of needing adjustment, the control unit informs a display unit that the drive shaft is in a state of needing adjustment, and adjusts the drive shaft for the processing head by comparing data of a signal from the sensor with prestored data obtained during normal operation, in a different process than a laser processing process.

A time difference setting device is provided with: a speed change acquisition unit that acquires a change in speed of a driving mechanism with respect to a positioning command based on a machining program; a mechanism delay time calculation unit that calculates a mechanism delay time; a laser output calculation unit that uses an oscillator model for simulating the operation of a laser oscillator and calculates a change in output of the laser oscillator with respect to a laser command based on the machining program; a laser delay time calculation unit that calculates a laser delay time; and a command time difference setting unit that, on the basis of the mechanism delay time and the laser delay time, sets a command time difference for an interval in which an emission of laser light is designated in the machining program.

The present invention relates to a method for generating control data for the secondary machining of a solid body (1), in particular wafer, which is modified by means of laser beams (10). The interior of said solid body (1) has multiple modifications (12), said modifications (12) having been produced by means of laser beams (10). The method comprises the following steps: defining a criterion of analysis for analyzing the modifications (12) produced in the interior of the solid body (1); defining a threshold value with respect to the criterion of analysis, an analytical value on one side of the threshold value triggering a secondary machining registration; analyzing the wafer by means of an analytical unit (4), said analytical unit (4) analyzing the modifications (12) with respect to the criterion of analysis and outputting analytical values regarding the analyzed modifications, said analytical values lying above or below the threshold value; outputting location data with respect to the analyzed modifications, said location data containing information regarding in which region(s) of the solid body (1) the analytical value lie above or below the threshold value; and generating control data for controlling a laser treatment device (11) for the secondary machining of the solid body (1), said control data comprising at least the location data of the modifications (12) registered for secondary machining.

The present invention relates to a machining device (100), a control apparatus (200) and a method for the continuous machining of workpieces (W.sub.1-W.sub.3), preferably plate-shaped workpieces (W.sub.1-W.sub.3), which preferably consist at least in sections of wood, wood material and/or synthetic material, a synchronization of a movement of a machining aggregate (130) with a feed movement of the workpiece (W.sub.1-W3) being carried out by a first control (201) and a positioning of the machining aggregate (130) according to a preset machining movement being carried out by a second control (202) with electronic cam disc.

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.