The invention generally relates to programmable motion systems and relates in particular to end effectors for programmable motion devices (e.g., robotic systems) for use in object processing such as object sortation.
End effectors for robotic systems may be employed, for example, in certain applications to select and grasp an object, and then move the acquired object very quickly to a new location. End effectors should be designed to quickly and easily select and grasp an object from a jumble of dissimilar objects, and should be designed to securely grasp an object during movement. Certain end effectors, when used on different objects of different physical sizes, weights and materials, may have limitations regarding how securely they may grasp an acquired object during rapid movement, particularly rapid acceleration and deceleration (both angular and linear).
Many end effectors employ vacuum pressure for acquiring and securing objects for transport and/or subsequent operations by articulated arms. Other techniques for acquiring and securing objects involve electrostatic attraction, magnetic attraction, needles for penetrating objects such as fabrics, fingers that squeeze an object, hooks that engage and lift a protruding feature of an object, and collets that expand in an opening of an object, among other techniques. Typically, end effectors are designed as a single tool, such as for example, a gripper, a welder, or a paint spray head, and the tool is typically designed for a specific set of needs.
There remains a need however, for an end effector system in a programmable motion system that may select and grasp any of a wide variety of objects, and then move the acquired object very quickly to a new location.
In accordance with an embodiment, the invention provides a programmable motion system that includes a dynamic end effector system. The dynamic end effector system includes a plurality of acquisition units that are provided at an exchange station within an area accessible by the programmable motion device, and a coupling system for coupling any of the plurality of acquisition units to an end effector of the programmable motion device such that any of the acquisition units may be automatically selected from the exchange station and used by the programmable motion device without requiring any activation or actuation by the exchange station and without requiring any intervention by a human.
In accordance with another embodiment, the invention provides a programmable motion system that includes a dynamic end effector system. The dynamic end effector system includes a plurality of vacuum cups through which a vacuum may be provided, and each of which may be attached to an end effector of the end effector system, wherein the programmable motion system is capable of accessing any of the plurality of vacuum cups, and a coupling system for coupling any of the plurality of vacuum cups to the end effector of the end effector system of the programmable motion device.
In accordance with a further embodiment, the invention provides a programmable motion system that includes a dynamic end effector system. The dynamic end effector system includes a plurality of acquisition units that are provided within access to the programmable motion device on an acquisition unit rack, the acquisition unit rack being movable in at least two mutually orthogonal directions.
In accordance with yet a further embodiment, the invention provides a method of providing the processing of objects using a programmable motion system. The method includes the steps of providing a plurality of vacuum units, each of which may be attached to an end effector of the end effector system, and each of which may provide a vacuum therethrough, accessing any of the plurality of vacuum units, coupling any of the plurality of vacuum units to the end effector of the end effector system of the programmable motion device, and using the coupled vacuum unit to grasp and move an object by the programmable motion device.
The following description may be further understood with reference to the accompanying drawings in which:
The drawings are shown for illustrative purposes only.
In accordance with an embodiment the invention provides a programmable motion system for moving objects for processing, such as sortation and order fulfillment. In applications such as order fulfillment, objects are collected into heterogeneous sets and need to be processed into appropriate groupings. In particular, individual objects need to be identified (e.g., by perception systems) and then routed to object-specific locations. The described system reliably automates the grasping and movement of such objects by employing both a robotic arm and versatile gripping systems. In accordance with various embodiments, the perception units (e.g., cameras or scanners) may look for a variety of codes such as indicia, e.g., barcodes, radio frequency identification (RFID) tags, Stock Keeping Unit (SKU) codes, Universal Parcel Codes (UPC), low wavelength IR (LWIR), as well as invisible barcodes and digital watermarks such as Digimarc DWCode, etc.
Sorting for order fulfillment is one application for automatically identifying objects from a heterogeneous object stream. Barcode scanners have a wide variety of uses including identifying the Stock Keeping Unit of an article, or tracking parcels. The system described herein may have many uses in the automatic identification and processing, e.g., sortation, of objects.
Such a system automates part of the sorting process in conjunction with a robotic pick and place system, and in particular, the step of grasping and carrying objects. A robotic arm, for example, picks an object from a bin, places the object in front of (or drops an object into) a scanner, and then, having obtained identification information for the object (such as a barcode, QR codes, UPC codes, other identification codes, information read from a label on the object, or assessed size, weight and/or shape information), routes the object toward the appropriate bin or shelf location by either moving the object itself, or placing the object in or on a conveyance system. Since certain scanners employ cameras or lasers to scan 1D or 2D symbologies printed on labels affixed to objects, the barcodes must be visible to the scanner's sensors for successful scanning in order to automatically identify items in a heterogeneous stream of arbitrary objects, as in a jumbled set of objects found in a bin.
Further applications for grasping systems of the invention include sortation for a wide variety of applications, including order fulfillment, collection of objects for shipping, and collection of objects for inventory purposes etc. Further, such grasp planning systems of the invention may be used for loading break-packs (organized packages for later breaking apart at a different location), palletization (loading pallets), de-palletization, truck loading, truck unloading etc. As used herein, the term “destination locations” involves locations at which objects are placed for any purpose, not necessarily a final destination, and not necessarily for sortation for order fulfillment.
In accordance with various embodiments, therefore, the invention provides a method for determining the identity of an object from a collection of objects, as well as a method for perceiving the information regarding an object employing one or more perception units (cameras or scanners), and a robotic arm with an end-effector for holding the object. The invention further provides a method for determining a sequence of placements of a robot end-effector so as to minimize the time it takes a configuration of one or more cameras or scanners to successfully scan an object, and a method for scanning the identification information (such as a barcode, QR codes, UPC codes, other identification codes, information read from a label on the object, or assessed size, weight and/or shape information) of an object by employing a scanner as an end-effector on a robotic arm.
An important aspect is the ability to identify identification or mailing information for the object (such as a barcode, QR codes, UPC codes, other identification codes, information read from a label on the object, or size, weight and/or shape information) of objects by employing a programmable motion device such as a robot arm, to pick up individual objects and place them in front of one or more scanners or to drop or place the object into a scanner. In accordance with other embodiments, the programmable motion device may include a parallel arm robot (Delta-type arm) or a linear indexing pick and place system. Certain convention scanning systems, for example, may be unable to see labels or barcodes on objects that are presented in a way that this information is not exposed or visible.
Important components of an automated processing system in accordance with an embodiment of the present invention are shown in
Images taken by the perception unit 26 may be displayed on a touch input screen 28 so that persons in the environment may interact with the system 10, e.g., by confirming, rejecting or proposing, possible grasp locations on objects, based on which the system 10 may undergo machine learning (e.g., in processor 34) with regard to the objects.
The system 10 provides that when an object has been grasped by the end effector 14 of the articulated arm 12, the end effector 14 may present the object to a perception station 29. As further shown in
Having identified a code or indicia, the processing system 34 then pulls routing information regarding the object, and the articulated arm 12 then moves the object using the end effector 14 to an appropriate location 32 of a bank of sortation locations 30. The end effector 14 further includes an acquisition unit 72 (such as a vacuum unit, e.g., a vacuum cup) for contacting and grasping the objects. In an embodiment, the acquisition unit 72 may be in the form of a flexible bellows, and may include a vacuum line 80 attached to the end effector 14 for providing a vacuum source at the mouth of the flexible bellows.
As further shown in
With reference to
In accordance with certain embodiments, therefore, the invention provides a programmable motion system including a dynamic end effector system. The dynamic end effector system includes a plurality of acquisition units and coupling means. The plurality of acquisition units are provided at an exchange station within an area accessible by the programmable motion device. The coupling means is for coupling any of the plurality of acquisition units to an end effector of the programmable motion device such that any of the acquisition units may be automatically selected from the exchange station and used by the programmable motion device without requiring any activation or actuation by the exchange station and without requiring any intervention by a human. While a human may place a new actuation unit onto the rack, and may then inform the system as to which actuation unit is placed on the rack and where it is placed, this is not required. The system may originally be set up such that the system is programmed to know which actuation units began in which positions (and thereafter track any movement via exchanges). In other embodiments, the vacuum cups may have unique physical or magnetic features that may be detected by the end effector.
The exchange station (such as the rack 78) is therefore inactive in that the system does not require the exchange station to actively open or close any latches, or move any carousels, etc. The programmable motion system knows where the rack is positioned, and knows and monitors which actuation units (e.g., vacuum cups) are on the rack and where each is positioned. The programmable motion system may therefore process the objects while switching vacuum cups automatically depending on input from the perception unit(s) 26 and/or 29.
When an acquisition unit (e.g., 72) is returned to the rack 78, the unit is moved in a direction as generally shown at A in
With reference again to
The interaction of the brackets 77, 79 and the rim 100 is further shown in
In accordance with a further embodiment and with reference to
Further flexibility may be built into systems of the invention by providing that the bracket rack 78 may be mounted to a frame 110 that includes a first beam 112 that extends in a first direction along the rack 78, and is coupled to the rack by spring elements 114 as shown in
The systems of
The x-y position zeroing system of
Systems of certain embodiments of the present invention provide that an acquisition unit, such as a vacuum cup (e.g., a flexible bellows type vacuum cup), through which a high vacuum may be designed to flow, may be exchanged for another vacuum cup during use, by the programmable vacuum device. In particular, and with reference to
If an object is identified by the perception unit 282 that requires a different vacuum cup than is currently attached to the end effector, the end effector may place the object back into the bin so that the object may be again grasped, but by a newly attached acquisition device. In certain embodiments, the perception unit 268 may sufficiently identify a next object, and if the vacuum cup on the end effector needs to be changed, the system may exchange a current vacuum cup to a desired one that is known to be a better acquisition unit for grasping the identified object in bin 260.
The system may further seek to identify all objects in a bin 260, may associate each with an optimal vacuum cup, and may then seek to grasp, one at a time, each of the objects associated with a common vacuum cup prior to changing the vacuum cup on the end effector. In each of these embodiments, the system itself identifies the need to change acquisition units, and then changes acquisition units by itself in the normal course of operation.
Systems of certain embodiments of the invention may also employ machine learning to improve performance over time. The system provides the performance of picking as a function of item, pick station and handling parameters. Further, objects that have not yet been picked will periodically be encountered. It is likely, however, that new objects that are similar to previously picked items, will have similar performance characteristics. For example, object S may be a kind of shampoo in a twenty ounce bottle, and object C may be conditioner in a twenty ounce bottle. If distributed by the same company, then the shape of the bottles may be the same. Systems of embodiments of the invention include processes that use observations of past performance on similar items to predict future performance, and learn what characteristics of the items available to the system are reliable predictors of future performance.
In accordance with certain embodiments, the system provides a learning process that (a) extrapolates the performance of newly seen objects, and (b) is continually updating the data with which it learns to extrapolate so as to continually improve performance. The potential pick parameters are diverse. Several controllable pick parameters may govern the process, such as, which picking stations can pick a given item, which effectors (vacuum cup size or gripper type) are effective for that item, and what rules might be used to choose locations on an item to grasp etc. Because these process parameters can change on a per-SKU basis, and will determine the efficacy and speed of a picking station and further may be determined on a per-SKU basis, it is necessary to estimate these parameters correctly. In particular, the correct values of process parameters depend on the nature of the item, its weight and size, its packaging, its material properties such as whether it is deformable or clear, whether vacuum grippers are effective at holding it, where good grasp locations are on the object, and whether it is easily damaged.
In many operating conditions however, this can be challenging, as new SKUs may be present, which means that for a new object, there is no known set of parameters available. While these parameters will be learned from repeated interactions with the object, this can slow down handling time considerably. To speed up the time it takes to learn the appropriate parameters, using previously recorded data based on similar SKUs can be useful.
In accordance with various embodiments, the invention provides processes for an automated material handling system that routes bins to picking stations, and which provides the following. The system may predict object-specific parameters for new objects based on previously seen objects. For new objects similar to previously handled objects, the processes predict what are expected to be good routing and handling parameters. In this instance an object is readily recognized as being quite similar to objects with which the system has extensive experience. From the bar code or SKU number or product category or description text or from appearance or other features, the system might recognize the object and index information in the database, which might include process parameters, or will at least include information from which process parameters can be determined with high confidence.
Further, the system may explore the parameter space for completely unknown objects. For new objects that are not sufficiently similar to any previously handled objects, the system may propose multiple candidate routing and handling parameters with the aim of finding good routing and handling parameters. When an unfamiliar object is first introduced, process parameters must be determined.
The system may also update predictive models of object-specific handling performance from observed item handling performance. Processes refine the routing and handling parameters on an object basis, as experience with that object is gained. The predictive model is refined as experience is gained.
The system may further update predictive models of object-to-object similarity from observed object handling performance. The parameters affecting the schemes and processes for classifying and/or clustering objects are refined as experience with all available items is increased. Further, the system may recognize and correct for persistent discrepancies in actual versus predicted performance. Some objects, when replenished by the manufacturer, have different weights, packaging or other characteristics that impact the object's handling performance. Old routing and handling parameters that yielded good performance before may be inappropriate for the changed object. When the actual performance repeatedly exceeds the range of the predicted performance, the system favors exploration of the parameter space.
The coupling of the different vacuum cups to an end effector via the magnets also presents fewer limitations on the lifting dynamics. In particular, and with reference to
In accordance with further embodiments of the invention, it may be desirable to design the vacuum cup such that any seal between the cup and an object will become compromised in the event that too large a weight is attempted to be lifted (protecting the magnetic coupling from being breached or protecting the articulated arm from overload). Such vacuum cups may, for example, permit some portions of the cup to open or otherwise break the seal between the cup and the object, thereby releasing the object from the vacuum cup.
Upon returning a vacuum cup 272 to a rack 278, the vacuum cup is placed on the rack as discussed above, and the coupling unit 204 and conduit 212 are pulled upward away from the rack. The rim 300 of the vacuum cup 272 is stopped by the underside of the upper bracket 279 as discussed above, and the vacuum cup 272 is separated from the coupling unit 204 and conduit 212 when the reactive force of the rack Fr overcomes the magnetic force fm as shown in
Those skilled in the art will appreciate that numerous modifications and variations may be made to the above disclosed embodiments without departing from the spirit and scope of the present invention.
The present application claims priority to U.S. Provisional Patent Application Ser. No. 62/618,184, filed Jan. 17, 2018, the disclosure of which is hereby incorporated by reference in its entirety.
Number | Name | Date | Kind |
---|---|---|---|
2853333 | Littell | Sep 1955 | A |
3005652 | Helm | Oct 1961 | A |
3181563 | Griffen | May 1965 | A |
3195941 | Morey | Jul 1965 | A |
4512709 | Hennekes | Apr 1985 | A |
4604787 | Silvers, Jr. | Aug 1986 | A |
4635985 | Rooke | Jan 1987 | A |
4653793 | Guinot et al. | Mar 1987 | A |
4722653 | Williams et al. | Feb 1988 | A |
4995493 | Cotsman | Feb 1991 | A |
5017084 | Lemelson | May 1991 | A |
5727832 | Holter | Mar 1998 | A |
5860900 | Dunning | Jan 1999 | A |
5865487 | Gore et al. | Feb 1999 | A |
5993365 | Stagnitto | Nov 1999 | A |
6015174 | Raes et al. | Jan 2000 | A |
6193291 | Morroney | Feb 2001 | B1 |
6213528 | Hufken et al. | Apr 2001 | B1 |
6244640 | Le Bricquer et al. | Jun 2001 | B1 |
6382692 | Schmalz et al. | May 2002 | B1 |
9266237 | Nomura | Feb 2016 | B2 |
9486926 | Kawano | Nov 2016 | B2 |
9492923 | Wellman et al. | Nov 2016 | B2 |
9687982 | Jules | Jun 2017 | B1 |
9981379 | Youmans et al. | May 2018 | B1 |
10007827 | Wagner et al. | Jun 2018 | B2 |
10576641 | Wagner et al. | Mar 2020 | B2 |
10618177 | Wagner et al. | Apr 2020 | B2 |
10625432 | Wagner et al. | Apr 2020 | B2 |
10639787 | Wagner et al. | May 2020 | B2 |
10668630 | Robinson et al. | Jun 2020 | B2 |
10723019 | Wagner et al. | Jul 2020 | B2 |
10850402 | Wagner et al. | Dec 2020 | B2 |
10857682 | Wagner et al. | Dec 2020 | B2 |
10913159 | Wagner et al. | Feb 2021 | B2 |
20020011735 | Nagai et al. | Jan 2002 | A1 |
20020185575 | Kalb | Dec 2002 | A1 |
20070006940 | Perlman et al. | Jan 2007 | A1 |
20070287884 | Schena | Dec 2007 | A1 |
20100078953 | Ban et al. | Apr 2010 | A1 |
20100109360 | Meisho | May 2010 | A1 |
20100175487 | Sato | Jul 2010 | A1 |
20120319416 | Ellis et al. | Dec 2012 | A1 |
20130147101 | Cho | Jun 2013 | A1 |
20130203572 | Denkmeier | Aug 2013 | A1 |
20150057793 | Kawano | Feb 2015 | A1 |
20150328779 | Bowman et al. | Nov 2015 | A1 |
20160089792 | Ojalehto | Mar 2016 | A1 |
20160167227 | Wellman et al. | Jun 2016 | A1 |
20160176043 | Mishra | Jun 2016 | A1 |
20160258473 | Koop et al. | Sep 2016 | A1 |
20170021499 | Wellman et al. | Jan 2017 | A1 |
20170050315 | Henry et al. | Feb 2017 | A1 |
20170057091 | Wagner et al. | Mar 2017 | A1 |
20170080571 | Wagner et al. | Mar 2017 | A1 |
20170080579 | Wagner et al. | Mar 2017 | A1 |
20170087731 | Wagner et al. | Mar 2017 | A1 |
20170106532 | Wellman et al. | Apr 2017 | A1 |
20170120454 | Ferguson et al. | May 2017 | A1 |
20170120455 | Wagner et al. | May 2017 | A1 |
20170121113 | Wagner et al. | May 2017 | A1 |
20170136632 | Wagner et al. | May 2017 | A1 |
20170197316 | Wagner et al. | Jul 2017 | A1 |
20170225330 | Wagner et al. | Aug 2017 | A1 |
20180009113 | Lauder et al. | Jan 2018 | A1 |
20180222061 | Alber | Aug 2018 | A1 |
20180281202 | Brudniok et al. | Oct 2018 | A1 |
20190001505 | Wagner et al. | Jan 2019 | A1 |
20190030727 | Nagata et al. | Jan 2019 | A1 |
20190039240 | Wagner et al. | Feb 2019 | A1 |
20190061174 | Robinson et al. | Feb 2019 | A1 |
20190071260 | Laverdiere | Mar 2019 | A1 |
20190102965 | Greyshock | Apr 2019 | A1 |
20190216644 | Hershoff et al. | Jul 2019 | A1 |
20190344447 | Wicks et al. | Nov 2019 | A1 |
20190361672 | Odhner et al. | Nov 2019 | A1 |
20200030994 | Wagner et al. | Jan 2020 | A1 |
20210039268 | Anderson | Feb 2021 | A1 |
20210308874 | Gealy et al. | Oct 2021 | A1 |
20210308875 | Gealy et al. | Oct 2021 | A1 |
20220048717 | Sullivan et al. | Feb 2022 | A1 |
Number | Date | Country |
---|---|---|
102202837 | Sep 2011 | CN |
104415868 | Mar 2015 | CN |
104802177 | Jul 2015 | CN |
2708335 | Mar 2014 | EP |
WO-2014040843 | Mar 2014 | WO |
2015162390 | Oct 2015 | WO |
2016100235 | Jun 2016 | WO |
Entry |
---|
International Preliminary Report on Patentability issued by the International Bureau of WIPO in related International Application No. PCT/US2019/013833 dated Jul. 21, 2020, 12 pages. |
Communication pursuant to Article 94(3) EPC issued by the European Patent Office in related European Patent Application No. 19704111.4 dated Feb. 16, 2022, 9 pages. |
Communication pursuant to Rules 161(1) and 162 EPC issued by the European Patent Office in related European Patent Application No. 19704111.4 dated Aug. 25, 2020, 3 pages. |
International Search Report & Written Opinion issued by International Searching Authority in related International Patent Application No. PCT/US2019/013833 dated May 16, 2019, 17 pgs. |
Examiner's Report issued by the Innovation, Science and Economic Development Canada (Canadian Intellectual Property Office) in related Canadian Patent Application No. 3,088,655 dated Aug. 20, 2021, 4 pages. |
Examiner's Report issued by the Innovation, Science and Economic Development Canada (Canadian Intellectual Property Office) in related Canadian Patent Application No. 3,088,655 dated Feb. 25, 2022, 4 pages. |
Examiner's Report issued by the Innovation, Science and Economic Development Canada (Canadian Intellectual Property Office) in related Canadian Patent Application No. 3,088,655 dated Sep. 21, 2022, 5 pages. |
Notice on the First Office Action, along with its English translation, issued by the China Intellectual Property Administration in related Chinese Patent Application No. 201980008600.3 dated Oct. 11, 2022, 31 pages. |
Vittor et al., “A Flexible Robotic Gripper for Automation of Assembly Tasks,” Proceedings of the ASME Dynamic Systems and Control Division, vol. 2, 72-2, presented at the 2003 ASME International Mechanical Engineering Congress, Nov. 15-21, 2003, Washington, D.C., 7 pages. |
Notice on the Second Office Action, along with its English translation, issued by the China National Intellectual Property Administration in related Chinese Patent Application No. 201980008600.3 dated Jun. 5, 2023, 34 pages. |
Examiner's Report issued by the Innovation, Science and Economic Development Canada (Canadian Intellectual Property Office) in related Canadian Patent Application No. 3,088,655 dated Aug. 24, 2023, 5 pages. |
Number | Date | Country | |
---|---|---|---|
20190217471 A1 | Jul 2019 | US |
Number | Date | Country | |
---|---|---|---|
62618184 | Jan 2018 | US |