An embodiment includes a robotic welding system for generating a motion program, having a programmable robot controller of a robot having a computer processor and a computer memory. The programmable robot controller is configured to digitally record, in the computer memory, a plurality of spatial points along an operator path in a 3D space taken by a calibrated tool center point (TCP) of the robot as an operator manually moves a robot arm of the robot along the operator path from a start point to a destination point within the 3D space. The programmable robot controller is also configured to identify and eliminate extraneous spatial points from the plurality of spatial points as digitally recorded, leaving a subset of the plurality of spatial points as digitally recorded, where the extraneous spatial points are a result of extraneous movements of the robot arm by the operator.

Method and apparatus for optimizing control of robotic systems. Failures of a robotic device performing a lifting operation for one or more items within a fulfillment center over a window of time are monitored. Responsive to the failures exceeding a predefined threshold number of failures for the window of time, one or more control operations for optimizing performance of the robotic device are determined, by processing environmental metrics, failure type, mode of operation and item type information as inputs to a trained machine learning model. A control system for the robotic device is configured based on the determined one or more control operations, and movement of the robotic device is controlled using the configured control system, to perform the lifting operation for one or more additional items within the fulfillment center.