A method and a device for simulating a machining process of a workpiece on an NC-controlled machine tool by means of a self-learning artificial neural network. Process parameters both from a machining process on a real machine tool located in a manufacturing section and a digital machine model implemented in a simulation section are provided to the artificial neural network to learn the behavior of the machine tool including the tools and workpieces used and are reformatted into input parameters by means of mathematical transformation. By learning the behavior of the machining process, the artificial neural network ca, send output files back to the simulation software of the simulation section and optimally adapt the behavior of the digital machine model to the conditions of the real machine tool by adapting the simulation parameters and make it more efficient in order to optimize the machining process on the machine tool.

A relationship between an input, a set-point of a plurality of processes and an output of a corresponding process is learned using machine learning. A regression function is derived for each process based upon historical data. An autoencoder is trained for each process based upon the historical data to form a regularizer and the regression functions and regularizers are merged together into a unified optimization problem. System level optimization is performed using the regression functions and regularizers and a set of optimal set-points of a global optimal solution for operating the processes is determined. An industrial system is operated based on the set of optimal set-points.

A method (1000) of diagnosis of operation of a machine tool (10, 100) that includes one or more axes (X, Y, Z) moved by one or more actuators (101, 102, 104) and at least one sensor (30) coupled to the machine tool (10, 100), the method (1000) comprising operations of: generating (1200) a programming sequence of movement of the axes (X, Y, Z) of the machine tool (10, 100); controlling (1210) the movement of the axes (X, Y, Z) of the machine tool (10, 100) according to the programming sequence; receiving (1220) a read-out signal (S) of the at least one sensor (30) coupled to the machine tool (10, 100); and processing (1230) the read-out signal (S) of the at least one sensor (30) coupled to the machine tool (10, 100). The programming sequence comprises instructions that are such as to apply (T) at least one single impulsive variation of a kinematic quantity that regards one or more actuators (101, 102, 104). The operation (1230) of processing the read-out signal (S) comprises processing a response of the machine tool (10, 100) to at least one single impulsive variation. The operation (1230) of processing the read-out signal (S) comprises artificial-neural-network processing (206) via one or more artificial neural networks (206, 2060) configured for analysing operating profiles in particular, one or more signals indicative of the status of the machine tool (W) in the read-out signal (S).

Methods, apparatus, systems, and articles of manufacture are disclosed that optimize layers of a machine learning model for a target hardware platform. An example apparatus includes a communication processor to obtain information specific to the target hardware platform (THP) on which to execute the machine learning model; a layer generation controller to generate layers of the machine learning model based on the information specific to the THP; and a deployment controller to, in response to the machine learning model satisfying a threshold error metric, deploy the machine learning model to the THP.

Disclosed are a collision avoidance method for a moving device, a collision avoidance apparatus for a moving device, and a computer-readable storage medium. This application relates to the field of artificial intelligence technologies. According to the method, a parking direction of a moving device in an avoidance area is adjusted, so that a startup time used by the moving device after avoidance completes may be reduced. The method includes: determining a target path direction of a moving device; determining a first candidate parking direction and a second candidate parking direction; determining, based on the target path direction, a target parking direction of the moving device from the first candidate parking direction and the second candidate parking direction; and controlling, based on the target parking direction, the moving device to be parked in the avoidance area.

Robots and robotic systems and methods can employ artificial neural networks (ANNs) to significantly improve performance. The ANNs can operate alternatingly in forward and backward directions in interleaved fashion. The ANNs can employ visible units and hidden units. Various objective functions can be optimized. Robots and robotic systems and methods can execute applications including a plurality of agents in a distributed system, for instance with a number of hosts executing respective agents, at least some of the agents in communications with one another. The hosts can execute agents in response to occurrence of defined events or trigger expressions, and can operate with a maximum latency guarantee and/or data quality guarantee.

Embodiments disclosed herein include a method of training a DNN. A processor initializes an element of an A matrix. The element may include a resistive processing unit. A processor determines incremental weight updates by updating the element with activation values and error values from a weight matrix multiplied by a chopper value. A processor reads an update voltage from the element. A processor determines a chopper product by multiplying the update voltage by the chopper value. A processor directs storage of an element of a hidden matrix. The element of the hidden matrix may include a summation of continuous iterations of the chopper product. A processor updates a corresponding element of a weight matrix based on the element of the hidden matrix reaching a threshold state.

An acquisition unit (11) acquires an explanatory variable that is to be input to a model (37) configured to output an objective variable for the explanatory variable, a specification unit (12) associates a frequency at which an explanatory variable included in each of a plurality of areas, which are obtained by dividing an explanatory variable space, is acquired by the acquisition unit (11) with each of the plurality of areas, and specifies an area to which an explanatory variable included in learning data used to learn the model (37) belongs and in which a frequency of an explanatory variable acquired by the acquisition unit (11) is a predetermined value or less, and an update unit (14) updates the model (37) in such a manner that learning data including an explanatory variable belonging to an area specified by the specification unit (12) is forgotten.

A processor controller includes: a deep neutral network, for extracting, based upon feature information of process control data, from a process control data storage device, process control data available to a production device to be controlled, the feature information of the process control data including at least production device feature parameters and a production device load; and an enhanced neural network, for performing, based upon a process control prediction model, process control prediction by using real-time process control data of said production device. In an embodiment, the process control prediction model is trained by using the extracted available process control data. The process controller further includes a process control decision unit, for determining an operation control instruction for the production device based upon the result of process control prediction. As such, prediction accuracy and training efficiency of the process control prediction model of the process controller can be improved.

It is disclosed a method and a predicting system for automatic prediction of porosity appearance generated during Laser Powder Bed Fusion (L-PBF), performed by an additive manufacturing system from at least one material. The method comprises steps for training a neural network comprising: generating labels of pore in every pixel using a porosity simulator; pre-training, comprising a first sub-step and a second sub-step, the second sub-step comprises using the data set created from the first sub-step to generate a pre-trained ML model; and training, comprising a first sub-step and a second sub-step, the second sub-step comprises using the data set created from the first sub-step to train the pre-trained ML model to generate a trained ML model.