Abstract: A computing system including one or more processing devices configured to receive a first labeled image set including first images that each include one or more first regions of interest. The one or more processing devices are further configured to receive an unlabeled image set including a plurality of second images, and to identify second regions of interest included in the second images. The one or more processing devices are further configured to compute feature similarity values between the second and first regions of interest, identify a subset of the second regions of interest that have feature similarity values above a predetermined similarity threshold, and apply respective second labels to those second regions of interest. The one or more processing devices are further configured to construct a second labeled image set and train an image classification machine learning model with a training data set that includes the labeled image sets.

Abstract: A computing system including one or more processing devices configured to receive a first labeled image set including first images that each include one or more first regions of interest. The one or more processing devices are further configured to receive an unlabeled image set including a plurality of second images, and to identify second regions of interest included in the second images. The one or more processing devices are further configured to compute feature similarity values between the second and first regions of interest, identify a subset of the second regions of interest that have feature similarity values above a predetermined similarity threshold, and apply respective second labels to those second regions of interest. The one or more processing devices are further configured to construct a second labeled image set and train an image classification machine learning model with a training data set that includes the labeled image sets.

Abstract: The present application is directed for use with a printhead assembly that is mounted to a robot. One application provides a system for inkjet printing on large objects such as commercial aircraft surfaces. The application provides for calibrating a camera with relation to the motion of the robot. The calibration occurs is achieved using one or more calibration artifacts printed by the inkjet system.

Abstract: Methods for selecting a path of a multi-component end effector along a surface, robots that perform the methods, and storage media that directs robots to perform the methods. The multi-component end effector is attached to a robot, which is configured to move the multi-component end effector along the surface on a continuous tool centerpoint path (TCP). The multi-component end effector includes a plurality of end effector elements configured to move relative to one another. The method includes providing a discretized TCP that includes a plurality of spaced-apart waypoints along the continuous TCP. The method also includes determining a plurality of distance heuristics. The method further includes updating the plurality of distance heuristics to define a plurality of updated distance heuristics. The method also includes selecting the path of the multi-component end effector along the surface based upon the plurality of updated distance heuristics.

Abstract: Methods for selecting a path of a multi-component end effector along a surface, robots that perform the methods, and storage media that directs robots to perform the methods. The multi-component end effector is attached to a robot, which is configured to move the multi-component end effector along the surface on a continuous tool centerpoint path (TCP). The multi-component end effector includes a plurality of end effector elements configured to move relative to one another. The method includes providing a discretized TCP that includes a plurality of spaced-apart waypoints along the continuous TCP. The method also includes determining a plurality of distance heuristics. The method further includes updating the plurality of distance heuristics to define a plurality of updated distance heuristics. The method also includes selecting the path of the multi-component end effector along the surface based upon the plurality of updated distance heuristics.