Abstract: Aspects of the disclosure are directed towards artificial intelligence-based modeling of target objects, such as aircraft parts. In an example, a system initially trains a machine learning (ML) model based on synthetic images generated based on multi-dimensional representation of target objects. The same system or a different system subsequently further trains the ML model based on actual images generated by cameras positioned by robots relative to target objects. The ML model can be used to process an image generated by a camera positioned by a robot relative to a target object based on a multi-dimensional representation of the target object. The output of the ML model can indicate, for a detected target, position data, a target type, and/or a visual inspection property. This output can then be used to update the multi-dimensional representation, which is then used to perform robotics operations on the target object.

Abstract: Techniques for auto-locating and positioning relative to an aircraft are disclosed. An example method can include a robot receiving a multi-dimensional representation of an enclosure that includes a candidate target aircraft. The robot can extract a geometric feature from the multi-dimensional representation associated with the candidate target aircraft. The robot can compare the geometric feature of the candidate target aircraft with a second geometric feature from a reference model of a target aircraft. The robot can determine whether the candidate target aircraft is the target aircraft based on the comparison. The robot can calculate a path from a location of the robot to the target aircraft based on the determination. The robot can traverse the path from the location to the target aircraft based on the calculation.

Abstract: The present disclosure relates to a multi-tiered computing environment for balancing compute resources in support of robot operations. In an example, a robot is tasked with performing an operation associated with an airplane having an airplane model. To do so, the robot may need another operation that is computationally complex to be performed. An on-premises server can execute a process that corresponds to this computationally-complex operation based on sensor data of the robot and can output the resulting data to the robot. Next, the robot can use the resulting data to execute another process corresponding to its operation and can indicate performance of this operation to the on-premises network. The on-premises network can send the indication about the operation performance to a top-tier server that is also associated with the airplane model.

Abstract: Techniques for auto-locating and autonomous operation data storage are disclosed. An example method can include storing a representation of an aircraft in a data storage of the computing system. The representation and instructions for performing a first operation on the aircraft can be transmitted to a first robot. The first robot can be configured to identify the aircraft based on a first feature of the representation. The method can further include receiving sensor data. The method can further include transmitting the sensor data to an analysis service to identify a state of a part of the aircraft. The method can further include storing the analysis in the data storage. The method can further include generating second instructions for performing a second operation on the aircraft based on the analysis and the representation stored in the data storage.