Item processing systems and methods employing parameter exploration
A control system is disclosed for an item processing system with a programmable motion device, said control system including a parameter estimator for estimating item handling parameters for items for which at least some parameters are known, a parameter explorer for identifying item handling parameters of items for which no parameters are known, and a parameter governor for determining whether to employ the parameter estimator or the parameter explorer in adjusting parameters for the item processing system.
1 . A control system for an item processing system with a programmable motion device, said control system comprising:
a parameter estimator for estimating adjustable item handling parameters of items for which item properties are known;
a parameter explorer for identifying through exploration adjustable item handling parameters of items for which no item properties are known; and
a parameter governor for determining whether to employ the adjustable item handling parameters from the parameter estimator or the parameter explorer to configure the programmable motion device to handle an item,
wherein the parameter governor determines to employ the adjustable item handling parameters from the parameter explorer responsive to at least one of a lack of history of handling with the item or a lack of information about item properties that affect handling of the item, and
wherein the parameter estimator, the parameter explorer, and the parameter governor are executed on a processing unit.
2 . The control system as claimed in claim 1 , wherein the adjustable item handling parameters include any of robot speed, choice of vacuum cup, vacuum pressure and vacuum flow.
3 . The control system as claimed in claim 1 , wherein the adjustable item handling parameters affect perception processes that generate grasp candidate locations.
4 . The control system as claimed in claim 1 , wherein the adjustable item handling parameters include threshold values that affect behavior during transport of an item.
5 . The control system as claimed in claim 1 , wherein the adjustable item handling parameters include a speed of movement of a shuttle carriage.
6 . The control system as claimed in claim 1 , wherein the adjustable item handling parameters include margins used in packing.
7 . The control system as claimed in claim 1 , wherein the adjustable item handling parameters include pose orientation.
8 . The control system as claimed in claim 1 , wherein the adjustable item handling parameters include case decanting parameters.
9 . The control system as claimed in claim 1 , wherein the adjustable item handling parameters are chosen by the parameter estimator or the parameter explorer using optimization-based modeling involving sequential hypothesis testing.
10 . The control system as claimed in claim 9 , wherein the adjustable item handling parameters are chosen by Bayesian optimization that recursively estimates a function being optimized.
11 . An item processing system comprising:
a programmable motion device, and
a control system including a parameter estimator, a parameter explorer, and a parameter governor for determining adjustable item handling parameters to configure the programmable motion device to handle an item,
wherein the parameter estimator is configured to estimate the adjustable item handling parameters for items for which item properties are known, and
wherein the parameter explorer is configured to identify through exploration the adjustable item handling parameters,
wherein the parameter governor provides the adjustable item handling parameters from the parameter explorer when the adjustable item handling parameters from the parameter estimator were unsuccessfully used in a predetermined number of past attempts to handle the item, and
wherein the parameter estimator, the parameter explorer, and the parameter governor are executed on a processing unit.
12 . The item processing system as claimed in claim 11 , wherein the item processing system include a plurality of perception systems for monitoring faults during item processing.
13 . The item processing system as claimed in claim 12 , wherein the faults include any of picking task faults, transport faults, packing faults, identification faults, and case decanting faults.
14 . The item processing system as claimed in claim 11 , wherein the adjustable item handling parameters include any of robot speed, choice of vacuum cup, vacuum pressure and vacuum flow.
15 . The item processing system as claimed in claim 11 , wherein the adjustable item handling parameters affect perception processes that generate grasp candidate locations.
16 . The item processing system as claimed in claim 11 , wherein the adjustable item handling parameters include threshold values that affect behavior during transport of an item.
17 . The item processing system as claimed in claim 11 , wherein the adjustable item handling parameters include a speed of movement of a shuttle carriage.
18 . The item processing system as claimed in claim 11 , wherein the adjustable item handling parameters include margins used in packing.
19 . The item processing system as claimed in claim 11 , wherein the adjustable item handling parameters include pose orientation.
20 . The item processing system as claimed in claim 11 , wherein the adjustable item handling parameters include case decanting parameters.
21 . The item processing system as claimed in claim 11 , wherein the adjustable item handling parameters are chosen by the parameter estimator or the parameter explorer using optimization-based modeling involving sequential hypothesis testing.
22 . The item processing system as claimed in claim 21 , wherein the adjustable parameters are chosen by Bayesian optimization that recursively estimates a function being optimized.
23 . A method of processing items with a programmable motion device, said method comprising:
providing a parameter estimator for estimating item handling parameters of items for which item properties are known;
providing a parameter explorer for identifying item handling parameters of items for which no item properties are known; and
employing a parameter governor for determining whether to handle an item using the item handling parameters from the parameter estimator or the parameter explorer,
wherein the parameter governor determines to employ the item handling parameters from the parameter explorer responsive to at least one of a lack of history of handling with the item, a lack of information about item properties that affect handling of the item, and wherein the parameter estimator, the parameter explorer, and the parameter governor are executed on a processing unit.
24 . The method as claimed in claim 23 , wherein the method further includes monitoring faults during item processing, the faults including any of picking task faults, transport faults, packing faults, identification faults, and case decanting faults.
25 . The method as claimed in claim 23 , wherein the adjustable item handling parameters for handling the item are chosen using optimization-based modeling involving sequential hypothesis testing.
26 . The method as claimed in claim 25 , wherein the adjustable item handling parameters for handling the item are chosen by Bayesian optimization involving recursively estimating a function being optimized.