A dynamic vision system includes a variable focus liquid lens optical assembly. The dynamic vision system includes a control system configured to adjust one or more optical parameters and data collected from the variable focus liquid lens optical assembly in real time. The dynamic vision system includes a processing system that dynamically learns on a training set of outcomes, parameters, and data collected from the variable focus liquid lens optical assembly to train one or more machine learning models to recognize an object.

A dynamic vision system includes a variable focus liquid lens optical assembly. The dynamic vision system includes a variable lighting assembly. The dynamic vision system includes a control system configured to adjust one or more optical parameters and data collected from the variable focus liquid lens optical assembly in real time. The dynamic vision system includes a control system configured to adjust the variable lighting assembly. The dynamic vision system includes a processing system that dynamically learns on a training set of outcomes, parameters, and data collected from the variable focus liquid lens optical assembly to train a set of machine learning models to control the variable focus liquid lens optical assembly to optimize collection of data for processing by the set of machine learning models.

An autonomous additive manufacturing platform includes sensors positioned in, on, and/or near a part and configured to collect sensor data related to the part. An adaptive intelligence system is configured to receive the sensor data from the sensors. The adaptive intelligence system includes a machine learning system configured to input the sensor data as training data into one or more machine learning models. The machine learning models are configured to transform the sensor data into simulation data. A digital twin system is configured to create a part twin based on the simulation data. The part twin provides for representation of the part and simulation of a possible future state of the part via the simulation data. An artificial intelligence system is configured to execute simulations on the digital twin system. The machine learning models are utilized to make classifications, predictions, and other decisions relating to the part.

An information technology system for a distributed manufacturing network includes an additive manufacturing management platform with an artificial intelligence system configured to learn on a training set of outcomes, parameters, and data collected from a set of distributed manufacturing network entities and execute simulations on digital twins of the set of distributed manufacturing network entities to make classifications, predictions, and optimization-related decisions for the set of distributed manufacturing network entities. The information technology system includes a distributed ledger system integrated with a digital thread and configured to provide unified views of workflow and transaction information to the set of distributed manufacturing network entities.

Systems, methods, devices, and other techniques for planning and re-planning robot motions. The techniques can include obtaining a schedule that defines an execution order for a plurality of plans, each plan defining a sequence of tasks; providing instructions to the robotic system to execute plans from the plurality of plans according to the schedule; obtaining measurements for one or more parameters that represent a state of the robotic system or its environment that results from execution of at least one plan from the plurality of plans; determining, based on the measurements, that a particular plan from the plurality of plans, which has not completed execution, is to be revised; generating, using the measurements, a revised version of the particular plan; and adjusting the schedule so as to resolve a conflict between the revised version of the particular plan and at least one other plan in the plurality of plans.

A method of generating a robot operation command includes extracting a work command from an overall work flow; extracting, from a command definition database, a command definition corresponding to the extracted work command, the work command having been read; generating, by referring to the read work command, a set of unit jobs in which at least one work command is arranged in order, based on the extracted command definition; generating, for a plurality of unit jobs, a connecting job for causing a robot to be moved from an end position of a previous unit job to a start position of a subsequent unit job; and generating a robot operation command in which the unit jobs and the connecting job are continuous.

A method of generating a robot operation command includes extracting a work command from an overall work flow; extracting, from a command definition database, a command definition corresponding to the extracted work command, the work command having been read; generating, by referring to the read work command, a set of unit jobs in which at least one work command is arranged in order, based on the extracted command definition; generating, for a plurality of unit jobs, a connecting job for causing a robot to be moved from an end position of a previous unit job to a start position of a subsequent unit job; and generating a robot operation command in which the unit jobs and the connecting job are continuous.

This disclosure provides systems and methods for robotic task planning when a complex task instruction is provided in natural language. Conventionally robotic task planning relies on a single task or multiple independent or serialized tasks in the task instruction. Alternatively, constraints on space of linguistic variations, ambiguity and complexity of the language may be imposed. In the present disclosure, firstly dependencies between multiple tasks are identified. The tasks are then ordered such that a dependent task is always scheduled for planning after a task it is dependent upon. Moreover, repeated tasks are masked. Thus, resolving task dependencies and ordering dependencies, a complex instruction with multiple interdependent tasks in natural language facilitates generation of a viable task execution plan. Systems and methods of the present disclosure finds application in human-robot interactions.

A process plan optimization method for manufacturing a workpiece by adding a material in a plurality of layers is provided. The method includes: building a predicting model, the predicting model configured to predict a temperature variation of at least a portion of the workpiece; predicting an expected temperature variation of the portion of the workpiece to be manufactured during a given time period based on the predicting model and the process plan; and adjusting the process plan in response to the expected temperature variation of the portion failing to meet a preset condition, to make the expected temperature variation of the portion meet the preset condition.

To reliably execute processing to be prioritized for each processing state for an industrial machine. A control device executes processing of processing requests for a machine tool from a client, the numerical control device including: a processing switching unit configured to switch, in a case in which a plurality of the processing requests is received from the client, an order of processing for each of the plurality of processing requests based on a priority according to a processing state for the machine tool.