A system and method that automatically resolves conflicts among sensor information in a sensor fusion robot system. Such methods can accommodate converging ambiguous and divergent sensor information in a manner that can allow continued, and relatively accurate, robotic operations. The processes can include handling sensor conflict via sensor prioritization, including, but not limited, prioritization based on the particular stage or segment of the assembly operation when the conflict occurs, overriding sensor data that exceeds a threshold value, and/or prioritization based on evaluations of recent sensor performance, predictions, system configuration, and/or historical information. The processes can include responding to sensor conflicts through comparisons of the accuracy of workpiece location predictions from different sensors during different assembly stages in connection with arriving at a determination of which sensor(s) is providing accurate and reliable predictions.

Multi-sensor fusion is a technology which effectively utilizes the data from multiple sensors so as to portray a unified picture with improved information and offers significant advantages over existing single sensor-based techniques. This disclosure relates to a method and system for a multi-label classification using a two-stage autoencoder. Herein, the system employs autoencoder based architectures, where either raw sensor data or hand-crafted features extracted from each sensor are used to learn sensor-specific autoencoders. The corresponding latent representations from a plurality of sensors are combined to learn a fusing autoencoder. The latent representation of the fusing autoencoder is used to learn a label consistent classifier for multi-class classification. Further, a joint optimization technique is presented for learning the autoencoders and classifier weights together. Herein, the joint optimization allows discriminative features to be learnt from the plurality of sensors and hence it displays superior performance than the state-of-the-art methods with reduced complexity.

A system and method that automatically resolves conflicts among sensor information in a sensor fusion robot system. Such methods can accommodate converging ambiguous and divergent sensor information in a manner that can allow continued, and relatively accurate, robotic operations. The processes can include handling sensor conflict via sensor prioritization, including, but not limited, prioritization based on the particular stage or segment of the assembly operation when the conflict occurs, overriding sensor data that exceeds a threshold value, and/or prioritization based on evaluations of recent sensor performance, predictions, system configuration, and/or historical information. The processes can include responding to sensor conflicts through comparisons of the accuracy of workpiece location predictions from different sensors during different assembly stages in connection with arriving at a determination of which sensor(s) is providing accurate and reliable predictions.

Systems and methods for machine positioning are provided herein. Exemplary embodiments include systems and methods using external positional information of an object under observation to compare to a programmed position of the object under observation. The comparison may be used in different manners including, for example, course correction, future path planning, object avoidance, etc.