Abstract: The invention relates to a technique for improving confidence estimates associated with neural networks. The technique involves computing neuron activation statistics during training, evaluating neuron activations during inferencing and determining how the activations compare with the previously computed statistics (e.g. whether prediction activations are within the bounds of the training activation statistics). The comparison may be used to compute a confidence value for the neural network.

Abstract: The present invention relates to systems and methods for accounting for edge cases (i.e. tail data) in automated decision making systems, for example automated robotic picking systems. The systems and methods provide for retraining machine learning (ML) models so that the edge cases can be handled in a manner that requires less (or no) human intervention. The disclosed systems and methods create updated ML models, replacement ML models, and/or supplementary ML models that can provide better performance (e.g. improved automated robotic picking) when edge cases are encountered. Furthermore, the present inventions disclose systems and methods for obtaining training data faster and in a more cost effective manner, which enables the systems and methods disclosed herein to update models at a faster rate, thereby enabling broader, system-wide handling of edge cases in a more effective and efficient manner.

Abstract: The present disclosure is for systems and methods for adjusting operational configurations of robots in real-time. The invention pertains to overriding or replacing one operational configuration of a robot with another when appropriate circumstances arise and certain conditions have been met. In one aspect, the invention is applicable to robotic picking operations and serves to allow for unique robotic picking operations outside of the normal or standard limitations typically imposed on a robotic picking system. The invention provides the ability to remotely adjust robotic operational configurations in real-time, on-demand, in order to address various circumstances that may arise without requiring interruption of a picking session or requiring on-site human intervention.

Abstract: The present invention relates to systems and methods for accounting for edge cases (i.e. tail data) in automated decision making systems, for example automated robotic picking systems. The systems and methods provide for retraining machine learning (ML) models so that the edge cases can be handled in a manner that requires less (or no) human intervention. The disclosed systems and methods create updated ML models, replacement ML models, and/or supplementary ML models that can provide better performance (e.g. improved automated robotic picking) when edge cases are encountered. Furthermore, the present inventions disclose systems and methods for obtaining training data faster and in a more cost effective manner, which enables the systems and methods disclosed herein to update models at a faster rate, thereby enabling broader, system-wide handling of edge cases in a more effective and efficient manner.

Abstract: The present disclosure is for a system and a method for computer vision based object detection. The invention uses images of objects from multiple perspectives and for each image identifies planes belonging to different objects. The planes are then analyzed to determine planes belonging to the same physical object. This is accomplished by comparing characteristics of the identified planes with each other and/or expected criteria. Planes identified as belonging to the same object can be grouped and used to provide pick instructions to a robot.

Abstract: The present disclosure is for systems and methods for providing observational support. The invention comprises an observational support system which generally operates as a supplemental system to an existing automated decision support system, such as a primary vision system for robotic picking operations. The observational support system provides an auxiliary sensor module which is operable to obtain data associated with a pick scene independently of the primary vision system and provide the data to an intervention system for further review and processing. The observational support system is generally called upon in situations where the primary vision system fails or encounters circumstances it cannot handle in a timely manner. The observational support system in combination with the intervention system provides supplemental assistance in these circumstances so that robotic picking operations can continue more readily.

Abstract: The present disclosure is for systems and methods for adjusting operational configurations of robots in real-time. The invention pertains to overriding or replacing one operational configuration of a robot with another when appropriate circumstances arise and certain conditions have been met. In one aspect, the invention is applicable to robotic picking operations and serves to allow for unique robotic picking operations outside of the normal or standard limitations typically imposed on a robotic picking system. The invention provides the ability to remotely adjust robotic operational configurations in real-time, on-demand, in order to address various circumstances that may arise without requiring interruption of a picking session or requiring on-site human intervention.

Abstract: The present invention relates to systems and methods for accounting for edge cases (i.e. tail data) in automated decision making systems, for example automated robotic picking systems. The systems and methods provide for retraining machine learning (ML) models so that the edge cases can be handled in a manner that requires less (or no) human intervention. The disclosed systems and methods create updated ML models, replacement ML models, and/or supplementary ML models that can provide better performance (e.g. improved automated robotic picking) when edge cases are encountered. Furthermore, the present inventions disclose systems and methods for obtaining training data faster and in a more cost effective manner, which enables the systems and methods disclosed herein to update models at a faster rate, thereby enabling broader, system-wide handling of edge cases in a more effective and efficient manner.