Autonomous Robot Packaging of Arbitrary Objects

A method for generating a three-dimensional model comprising:

• scanning a plurality of objects with at least one object detection sensor of a plurality of object detection sensors;
• segmenting a point cloud for each object in the plurality of objects;
• projecting the point cloud for each object in the plurality of objects onto an image obtained by at least one object detection sensor of the plurality of object detection sensors for each object in the plurality of objects to obtain a region of interest;
• based on the region of interest, generating a three-dimensional model for each object in the plurality of objects;
• assigning at least one parameter to the three-dimensional model for each object;
• and
• training a machine learning model based on the assigned at least one parameter.

A method for packing objects in a container, comprising:

• sensing a plurality of objects with at least one object detection sensor of a plurality of object detection sensors;
• obtaining a three-dimensional model for each object in the plurality of objects;
• determining a packing plan for the plurality of objects based on the three-dimensional model for each object;
• loading, by a robotic arm, at least a portion of the plurality of objects into a container according to the packing plan for the plurality of objects;
• sensing, by a plurality of object detection sensors, a content placement within the container;
• based on the content placement within the container, determining a deviation from the packing plan;
• and
• based on the deviation from the packing plan, performing at least one of:
• moving the object thereby correcting the deviation, or adjusting a remainder of the packing plan.