Machine vision-based method and system to facilitate the unloading of a pile of cartons in a carton handling system
A machine vision-based method and system to facilitate the unloading of a pile of cartons within a work cell are provided. The method includes the step of providing at least one 3-D or depth sensor having a field of view at the work cell. Each sensor has a set of radiation sensing elements which detect reflected, projected radiation to obtain 3-D sensor data. The 3-D sensor data including a plurality of pixels. For each possible pixel location and each possible carton orientation, the method includes generating a hypothesis that a carton with a known structure appears at that pixel location with that container orientation to obtain a plurality of hypotheses. The method further includes ranking the plurality of hypotheses. The step of ranking includes calculating a surprisal for each of the hypotheses to obtain a plurality of surprisals. The step of ranking is based on the surprisals of the hypotheses.
1. A machine vision-based method to facilitate the unloading of a pile of cartons within a work cell in an automated carton handling system, the method comprising the steps of:
providing at least one 3-D or depth sensor having a field of view at the work cell, the at least one sensor having a set of radiation sensing elements which detect reflected, projected radiation to obtain 3-D sensor data, the 3-D sensor data including a plurality of pixels;
for each possible pixel location and each possible carton orientation, generating a hypothesis that a carton with a known structure appears at that pixel location with that container orientation to obtain a plurality of hypotheses;
ranking the plurality of hypotheses wherein the step of ranking includes calculating a surprisal for each of the hypotheses to obtain a plurality of surprisals and wherein the step of ranking is based on the surprisals of the hypotheses; and
unloading at least one carton of interest from the pile based on the ranked hypotheses.
2. The method as claimed in claim 1 , further comprising utilizing an approximation algorithm to unload a plurality of cartons at a time from the pile in a minimum number of picks.
3. The method as claimed in claim 1 , wherein the work cell is a robot work cell.
4. The method as claimed in claim 1 , wherein the at least one sensor is a hybrid 2-D/3-D sensor.
5. The method as claimed in claim 1 , wherein the at least one sensor includes a pattern emitter for projecting a known pattern of radiation and a detector for detecting the known pattern of radiation reflected from a surface of the carton.
6. The method as claimed in claim 1 , wherein the at least one sensor is at least one volumetric sensor capable of capturing thousands of individual points in space.
7. The method as claimed in claim 1 , wherein at least one of the hypotheses is based on print on at least one of the cartons.
8. The method as claimed in claim 5 , wherein the pattern emitter emits a non-visible pattern of radiation and the detector detects the reflected non-visible pattern of radiation.
9. A machine vision-based system to facilitate the unloading of a pile of cartons within a work cell in an automated carton handling system, the system comprising:
at least one 3-D or depth sensor having a field of view at the work cell, the at least one sensor having a set of radiation sensing elements which detect reflected, projected radiation to obtain 3-D sensor data, the 3-D sensor data including a plurality of pixels;
at least one processor to process the 3-D sensor data and, for each possible pixel location and each possible carton orientation, generate a hypothesis that a carton with a known structure appears at that pixel location with that container orientation to obtain a plurality of hypotheses;
the at least one processor ranking the plurality of hypotheses wherein the ranking includes calculating a surprisal for each of the hypotheses to obtain a plurality of surprisals and wherein the ranking is based on the surprisals of the hypotheses; and
a vision-guided robot for unloading at least one carton of interest from the pile based on the ranked hypotheses.
10. The system as claimed in claim 9 , wherein the at least one processor utilizes an approximation algorithm so that the vision-guided robot unloads a plurality of cartons at a time from the pile in a minimum number of picks.
11. The system as claimed in claim 9 , wherein the work cell is a robot work cell.
12. The system as claimed in claim 9 , wherein the at least one sensor is a hybrid 2-D/3-D sensor.
13. The system as claimed in claim 9 , wherein the at least one sensor includes a pattern emitter for projecting a known pattern of radiation and a detector for detecting the known pattern of radiation reflected from a surface of the carton.
14. The system as claimed in claim 9 , wherein the at least one sensor is a volumetric sensor capable of capturing thousands of individual points in space.
15. The system as claimed in claim 9 , wherein at least one of the hypotheses is based on print on at least one of the cartons.
16. The system as claimed in claim 13 , wherein the pattern emitter emits a non-visible pattern of radiation and the detector detects the reflected non-visible pattern of radiation.