Method and system for efficiently packing a transport container with items picked from a transport structure
A method and system for efficiently packing a transport container with a plurality of substantially identical target items picked from a transport structure with a picking tool are provided. The method includes the step of providing a plurality of equivalence classes which partition all possible sequences of legal picks. Each equivalence class containing a subset of items from a ranked configuration. The method also includes ranking the equivalence classes in order of expected time efficiency under the constraint that a minimum, predetermined level of space efficiency is maintained. Finally, the method includes selecting the best ranked equivalence class for picking.
1 . A method of efficiently packing a transport container with items picked from a transport structure with a picking tool, the method comprising the steps of:
illuminating top surfaces of the items to obtain reflected, backscattered illumination;
providing a 3-D or depth sensor having a field of view to sense the reflected, backscattered illumination to obtain depth and grayscale images of the top surfaces of the items in the field of view;
determining potential items to be picked in the field of view of the sensor based on an area of the depth image;
determining depth information which is based on changes in depth in the depth image;
determining grayscale information which is based on changes in brightness in the grayscale image;
for each possible pixel location in the depth and grayscale images and each possible item orientation, generating a hypothesis that the potential items in the field of view appear at that pixel location with that item orientation to obtain a plurality of hypotheses;
ranking the plurality of hypotheses, wherein the step of ranking includes calculating depth and grayscale surprisals from the depth and grayscale information, respectively, for each of the hypotheses to obtain a plurality of surprisals and wherein the step of ranking is based on the plurality of surprisals;
determining contiguous items with a subtended area less than a size of the transport container and that fit a shape of the picking tool based on the ranked hypotheses to obtain legal picks;
generating a plurality of possible sequences of moves of the picking tool to fill and pack the transport container with the contiguous items; and
ranking the possible sequences in order of optimal space usage in the transport container.
2 . The method as claimed in claim 1 , wherein the step of illuminating includes the step of uniformly illuminating the top surfaces of the items with light having an intensity within a relatively narrow range of wavelengths such that the light overwhelms the intensity of ambient light within the narrow range to obtain the reflected, backscattered illumination.
3 . The method as claimed in claim 1 further comprising providing an autonomous manipulator registered with the sensor and attached to the picking tool and controlling the autonomous manipulator and the picking tool to fill and pack the transport carrier with the contiguous items.
4 . The method as claimed in claim 1 , further comprising computing rotation and position invariant surface albedos of the items based on the images wherein the brightness is intrinsic brightness of the items.
5 . The method as claimed in claim 1 , wherein the sensor is a hybrid 2-D/3-D sensor.
6 . The method as claimed in claim 1 , further comprising determining printed pattern information which is based on printed patterns in the grayscale image and located on the top surfaces of the items.
7 . The method as claimed in claim 1 , wherein the items are box-like objects.
8 . The method as claimed in claim 1 , wherein the transport container is a decanting tote.
9 . The method as claimed in claim 3 , wherein the autonomous manipulator is a vision-guided robot configured to fill and pack the transport container with the contiguous items based on the ranked sequences.
10 . The method as claimed in claim 6 , wherein at least one of the hypotheses is based on the printed pattern and wherein the step of ranking includes calculating printed pattern surprisals from the printed pattern information.
11 . A system for efficiently packing a transport container with items picked from a transport structure with a picking tool, the system comprising:
a light source configured to illuminate top surfaces of the items to obtain reflected, backscattered illumination;
a volumetric sensor having a field of view and configured to sense the reflected, backscattered illumination to obtain depth and grayscale images of the top surfaces of the items in the field of view; and
at least one processor configured to determine potential items to be picked in the field of view based on an area of the depth image, configured to determine depth information which is based on changes in depth in the depth image and configured to determine grayscale information which is based on changes in brightness in the grayscale image wherein, for each possible pixel location in the depth and grayscale images and each possible item orientation, the at least one processor being configured to generate a hypothesis that the potential items in the field of view appear at that pixel location with that item orientation to obtain a plurality of hypotheses and to rank the plurality of hypotheses; the at least one processor being configured to calculate depth and grayscale surprisals from the depth and grayscale information, respectively, for each of the hypotheses to obtain a plurality of surprisals, the at least one processor being configured to rank the hypotheses based on the surprisals and to determine contiguous items with a subtended area less than a size of the transport container and that fit a shape of the picking tool based on the ranked hypotheses to obtain legal picks, the at least one processor being configured to generate a plurality of possible sequences of moves of the picking tool to fill and pack the transport container with the contiguous items wherein the at least one processor is configured to rank the possible sequences in order of optimal space usage in the transport container.
12 . The system as claimed in claim 11 , wherein the at least one processor is configured to determine printed pattern information which is based on printed patterns in the grayscale image and located on the top surfaces of the items.
13 . The system as claimed in claim 11 , wherein the items are box-like objects.
14 . The system as claimed in claim 11 , wherein the light source is configured to uniformly illuminate the top surfaces of the items with light having an intensity within a relatively narrow range of wavelengths such that the light overwhelms the intensity of ambient light within the narrow range to obtain the reflected, backscattered illumination.
15 . The system as claimed in claim 11 , further comprising an autonomous manipulator registered with the sensor and attached to the picking tool, the picking tool and the manipulator being configured to fill and pack the transport container with the contiguous items.
16 . The system as claimed in claim 11 , wherein the sensor is a hybrid 2-D/3-D sensor.
17 . The system as claimed in claim 11 , wherein the at least one processor is configured to compute rotation and position invariant albedos of the items based on the images wherein the brightness is intrinsic brightness of the items.
18 . The system as claimed in claim 11 , wherein the transport container is a decanting tote.
19 . The system as claimed in claim 12 , wherein at least one of the hypotheses is based on the printed patterns and wherein the at least one processor is configured to calculate printed pattern surprisals from the at least one hypothesis.
20 . The system as claimed in claim 15 , wherein the autonomous manipulator is a vision-guided robot configured to fill and pack the transport container based on the ranked sequences.