Systems and methods for identifying package properties in an automated industrial robotics system

A computer-implemented method, comprising:

• obtaining an interrupt notification, the interrupt notification indicating that an item is positioned within a detection region of an interrupt sensor;
• obtaining two-dimensional (2D) data from a 2D camera of a plurality of cameras configured to capture 2D data in a package conveyance system, individual cameras being arranged at different positions with respect to markers within the package conveyance system, the 2D data including a representation of the item;
• obtaining three-dimensional (3D) data from a depth detection camera of a plurality of depth detection cameras configured to capture depth data in the package conveyance system, individual depth detection cameras being arranged at different positions with respect to the markers, the 2D data and the depth data captured substantially simultaneously, the 3D data including depth data associated with the item;
• transforming the 2D data and 3D data to a reference 3D coordinate space based on positions of the plurality of cameras and positions of the plurality of depth detection cameras to associate 3D coordinates with geometric features represented in the 2D data relative to the reference 3D coordinate space;
• and
• determining item properties associated with the item based on the geometric features.

A non-transitory computer readable storage medium storing one or more sequences of instructions executable by one or more processors to perform a set of operations comprising:

• obtaining an interrupt notification, the interrupt notification indicating that an item is positioned within a detection region of an interrupt sensor;
• obtaining two-dimensional (2D) data from a 2D camera of a plurality of cameras configured to capture 2D data in a package conveyance system, individual cameras being arranged at different positions with respect to markers within the package conveyance system, the 2D data including a representation of the item;
• obtaining three-dimensional (3D) data from a depth detection camera of a plurality of depth detection cameras configured to capture depth data in the package conveyance system, individual depth detection cameras being arranged at different positions with respect to the markers, the 2D data and the depth data captured substantially simultaneously, the 3D data including depth data associated with the item;
• transforming the 2D data and 3D data to a reference 3D coordinate space based on positions of the plurality of cameras and positions of the plurality of depth detection cameras to associate 3D coordinates with geometric features represented in the 2D data relative to the reference 3D coordinate space;
• and
• determining item properties associated with the item based on the geometric features.