GB2533676B - System and method for picking validation - Google Patents

Description

System and Method for Picking Validation

Cross-Reference to Related Application [0001] This application claims the benefit of U.S. Patent Application Ser. No. 62/062,175 for Systems and Methods for Dimensioning, (filed October 10, 2014), which is hereby incorporated by reference in its entirety.

Field of the Invention [0002] The present invention relates to computer-assisted picking and more specifically to systems/methods to validate a picking operation using a picked item's physical characteristics .

Background [0003] Order preparation (i.e., picking) is one operation in logistics. Picking consists of gathering a specified quantity of items from a location (e.g., a bin, a shelf in a warehouse, etc.) for some purpose (e.g., shipment, order fulfillment, assembly, etc.). Picking may be accomplished by providing a worker with a list of items and quantities, (i.e., a pick list). The worker then moves about the location gathering the items according to the pick list.

[0004] Various tools exist to improve the basic picking process. For example, a worker may use a mobile computing device (i.e., MOD) that is wirelessly connected to a warehouse management system (i.e., WMS) to receive and respond to picking instructions generated by the WMS. The worker's response may include information regarding the picking status (e.g., ready for next item), or may include validation information to help prevent picking errors (e.g., a check digit to confirm location). Relevant disclosures can be found in US 5730252 A, US 2014/350710 A1, DE 3335760 A1, JP 2014210646 A, JP 04129902 A, and US 2011301994A1. US 5730252 A discloses a picking system in which the weight of articles is used to validate the action. US 2014/350710 A1 discloses radiofrequency identification of articles in a picking system to validate actions. DE 3335760 A1 discloses weighing articles to validate a picking system.

[0005] Automated picking validation may be used in place of (or in addition to) user-enabled picking validation. For example, the worker may place items on a scale as part of a picking process. The weight of the picked item (or items) may be used to validate (i) that the correct item has been picked and/or (ii) that the correct quantity of the item has been picked. Despite these measures, picking errors still occur. Therefore, a need exists for a more accurate picking validation scheme that uses new and/or more sensors to verify one or more of an item's physical attributes automatically.

Summary [0006] The present invention in its various aspects is as set out in the appended claims. The present invention relates to a method for validating a picking operation. The method begins with the picking of an item. The picked item is then placed on a platform in view of sensors. The sensors sense sensed-physical attributes of the picked item. A computing device receives the sensed-physical attributes and compares these attributes to one or more expected-physical attributes stored on the computing device. If the one or more sensed-physical attributes do not match the one or more expected-physical attributes, then a picking-error alert is created.

[0007] In the method for validating a picking operation, sensors include a dimensioning system, and the one or more sensed-physical attributes include a one-dimensional (ID), two-dimensional (2D), and/or three-dimensional (3D) measurement corresponding to the item's shape/size.

[0008] The one or more sensed-physical attributes include the item's weight.

[0009] The sensors include a dimensioning system and a digital camera, the one or more sensed-physical attributes include the item's visual appearance.

[0010] In another exemplary embodiment of the method for validating a picking operation, the sensors include a dimensioning system, a scale, and a digital camera, and the sensed-physical attributes include (i) a ID, 2D, and/or 3D measurement of the item's shape/size, (ii) the item's weight, and (iii) the item's visual appearance.

[0011] In another exemplary embodiment of the method for validating a picking operation, the picking-error alert is an audio alert, a visual alert, a haptic alert, or some combination thereof.

[0012] In another exemplary embodiment of the method for validating a picking operation, the method includes updating the one or more expected-physical attributes using the one or more sensed-physical attributes if a match is found between the sensed-physical attributes and the expected-physical attributes. In some embodiments, the method may further include storing the updated expected-physical attributes on the computing device.

[0013] In another exemplary embodiment of the method for validating a picking operation, the method includes creating a picking-confirmation alert if the one or more sensed-physical attributes matches the one or more expected-physical attributes, and in this case, the picking-confirmation alert can be an audio alert, a visual alert, a haptic alert, or some combination thereof.

[0014] In another aspect, the present invention embraces a powered-industrial vehicle. The powered-industrial vehicle includes a user interface, at least one sensor, and a computing device, wherein the computing device is communicatively coupled to the user interface and sensors.

The powered-industrial vehicle's sensors are configured to (i) sense the physical attributes of an item placed on the power-industrial vehicle and (ii) output physical data. The powered-industrial-vehicle's computing-device receives the physical data from the sensors and generates a set of sensed-physical attributes for the item. The computing device then retrieves a set of expected-physical attributes for the item from the computing device's memory. The set of sensed-physical attributes is compared to the set of expected-physical attributes, and based on the comparison, the computing device transmits an alert message to the user interface .

[0015] The sensors include a dimensioning system.

[0016] In another exemplary embodiment of the powered-industrial vehicle, the alert message includes (i) an indication that the wrong item has been placed on the powered-industrial vehicle, or (ii) an indication that the correct item has been placed on the powered-industrial vehicle.

[0017] In another aspect, the present invention embraces a picking system. The picking system includes a host computer that is communicatively coupled to a mobile computer. The host computer creates and sends a pick list to the mobile computer. The pick list includes one or more items to be picked and a set of expected physical attributes corresponding to the one or more items to be picked. The mobile computer is communicatively coupled to sensors for sensing the physical attributes of a picked item and for outputting physical data. The mobile computer also includes a user interface, a memory for storing the pick list, and a processor communicatively coupled to the memory, the user interface, and the at least one sensor. The processor is configured by software to validate the picked item based on a set sensed-physical attributes generated using the physical data. Validation includes retrieving a set of expected-physical attributes for the picked item from memory and comparing the set of expected-physical attributes to the set of sensed-physical attributes. Based on this comparison, an alert message is transmitted to the user interface.

[0018] The sensors include a dimensioning system.

[0019] The sensors include (i) a dimensioning system and (ii) a scale and a digital camera.

[0020] In another exemplary embodiment of the picking system, the processor is further configured to update the set of expected-physical attributes using the set of sensed-physical attributes .

[0021] The foregoing illustrative summary, as well as other exemplary objectives and/or advantages of the invention, and the manner in which the same are accomplished, are further explained within the following detailed description and its accompanying drawings .

Brief Description of the Drawings [0022] Figure 1 is a flow diagram illustrating a method for validating a picking operation according to an embodiment of the present invention.

[0023] Figure 2 graphically depicts the implementation of a powered-industrial vehicle according to a first exemplary embodiment of the present invention.

[0024] Figure 3 graphically depicts the implementation of a powered-industrial vehicle according a second exemplary embodiment of the present invention.

[0025] Figure 4 is a block diagram illustrating a picking system according to an exemplary embodiment of the present invention.

Detailed Description [0026] The present invention embraces an automatic picking validation system/method based on sensing physical aspects of a picked item and providing feedback (i.e., messages, alerts, etc.) to a user based on a comparison between the sensed-physical aspects and expected-physical aspects stored in memory.

[0027] Picking includes gathering particular quantities of items in an item-storage location (e.g., a warehouse). Typically, picking is performed as part of a logistics (e.g., supply-chain) process. Picking may be accomplished in a variety of ways (e.g., picker-to-part, part-to-picker, etc.), and different tools have been created to facilitate aspects of this work. These tools may help (i) picking management (e.g., warehouse management system (WMS)), (ii) directing a worker (e.g., mobile computers, voice headsets, etc.), (iii) lifting/carrying items (e.g., powered industrial vehicles), and (iv) sorting/counting items (e.g., robotics). Despite, this automation/mechanization picking is still a labor-intensive activity and prone to human error.

[0028] Picking errors are costly and should be avoided. To this end, workflows are often designed to eliminate picking errors by requiring workers to interact with the WMS (e.g., to confirm a location, to confirm number of items picked, etc.). Even with these measures, picking errors may still occur.

[0029] Automated picking validation (e.g., picking error detection, picking confirmation, etc.) may support or replace worker-supported (i.e., user-enabled) picking validation. Automation relies on the fact that picked items have distinguishing physical attributes (e.g., weight, size/shape, color, markings, etc.). By sensing a picked item's physical attributes and comparing these sensed-physical attributes to physical characteristics that are expected (i.e., expected-physical attributes), picking errors may be detected or correct picking may be confirmed.

[0030] Scales have been used to determine the weight of picked items as they are gathered. Validating items by weight, while useful, has not completely replaced user-enabled validation.

One reason for this is that the measured weight is not always accurate. Scales designed for heavy items may not accurately weigh lighter items. In addition, when weighing many items, each item having a weight that slightly deviates from an expected value, errors can result when the weight deviations for each item accumulate.

[0031] Sensors may be combined with (or may replace) the scale in order to improve picking validation and/or add functionality to the picking process. The set of sensors may perform a variety of functions (e.g., sense the physical attributes of items, count items, weigh/measure items, read markings on items, etc.). A computing device, with memory and software algorithms running on a processor, may aggregate the data from the sensors to provide a more accurate representation of the picked item and further automate the picking process.

[0032] Machine vision may be used for picking validation. Machine vision is the automatic visual inspection and analysis of items placed in front of an imaging device (e.g., digital camera). Images from the imaging device may be processed to detect an item by its appearance (e.g., color, reflectivity, shape, markings, etc.). The image processing may include variety of algorithms (e.g., image-stitching, filter, thresholding, pixel counting, segmentation, edge detection, color analysis, pattern recognition, optical-character-recognition, etc.). The processor for image processing (e.g., image processing unit) may be separate from the imaging device or combined with the imaging device (e.g., smart camera). The imaging device may capture images of visible or invisible light and in some machine vision systems utilize special illumination (e.g., ultraviolet (UV), infrared (IR), etc.) to make visible certain aspects of an item (e.g., security marks).

[0033] Systems to measure (without contact) an item in three dimensions (i.e., dimensioning systems, dimensioners) may be used for picking validation. Dimensioning systems typically emit radiation (e.g., light, ultrasonic, x-ray, etc.) towards an item and detect the reflected radiation from the item in order to determine the size/shape of the item. Dimensioning systems may use various methods to probe the item (e.g., time-of-flight, triangulation, etc.).

[0034] In one dimensioning system embodiment, an IR pattern (i.e., point cloud) is projected onto an item. The reflected IR pattern is detected on an imaging device (e.g., range camera) and the position of the pattern elements in the point cloud image are measured and compared to a known pattern.

Offsets in the imaged pattern correspond to changes in range (i.e., depth). Measured ranges may then be used to compute the dimensions of the item or to create a 3D model of the item. This dimensioning system is typically characterized by a projector and a range camera in a stereoscopic arrangement. The processing of the point cloud image may be performed by the dimensioning system or by a computing device (e.g., mobile computer, host computer, etc.) that is in communication with the dimensioner.

[0035] By combining the outputs from a plurality of sensors, the picked item may be validated more accurately. An exemplary method for validating a picking operation is shown in Figure (Fig.) 1. First, an item is picked 10. The item may be picked automatically and brought to the user (e.g., sorting system and/or conveyor belt) or the user may move to the item and pick it manually. The item is then placed on a platform 20. The platform of a powered industrial vehicle.

The platform supports the item so that it may be measured by at least one sensor. In one possible embodiment, the platform may be part of a scale to weigh the item. The platform is in the field of view of a digital camera (e.g., machine vision system). The platform is in the field of view of a dimensioning system. In still another possible embodiment, the platform may be part of a scale and in the field of view of the digital camera and the dimensioner. Various combinations of the sensors are within the scope of the invention.

[0036] After placing the item on the platform, the item is measured using othe sensors 30 to collect sensed-physical attributes 40. The sensed physical attributes are then compared to a set of expected-physical attributes 45 stored in a computer readable memory. The comparison is performed by a computing device (e.g., mobile computer, host computer, etc.) that is communicatively coupled (e.g., wireless or wired) to the one or more sensors and the computer readable memory. The computing device may be body worn, handheld, desktop, laptop, or may be integrated with a powered-industrial vehicle. Algorithms running on the computer's processor then compare the sensed and expected physical attributes 50. If the attributes do not match, then a picking-error alert is created 60. Alternatively, if the comparison results in a match, then a picking confirmation may be created 70. The picking-error alert and the picking-confirmation alert may be an audio (e.g., voice), visual (e.g., text/graphics on a screen, signal light, printed message, etc.), or haptic (e.g., vibration) messages. In addition, the alerts may use some combination of these messages (e.g., a flashing light and a vibration).

[0037] The expected-physical attributes for an item may be stored in the computing device's memory. In one possible embodiment, the expected attributes for all possible picked items are installed in the memory during a setup procedure.

In another possible embodiment, the expected attributes are downloaded as needed from a host computer (e.g., warehouse management system) and stored locally in the computing device's memory during the picking process.

[0038] Expected-physical attributes may be acquired through preliminary training (i.e., supervised learning) and used until it is necessary to retrain (e.g., until a new item is added, until an item's attributes change, etc.). In many cases, however, it is desirable for the picking validation to be adaptable, and in these cases, updating the expected-physical attributes using the sensed-physical attributes (i.e., machine learning) is required. For example, if the one or more sensed-physical attributes does not match the expected-physical attributes (e.g., an item's package has changed) then the expected-physical attributes may be updated 80. In one possible embodiment, the sensed-physical attributes replace or add to the expected-physical attributes stored in memory. In another possible embodiment, the sensed-physical attributes are uploaded to a host computing system along with other sensed-physical attributes (e.g., collected by another user, collected at a different time, etc.) to help create new expected-physical attributes (i.e., unsupervised learning).

[0039] The picking process may require a worker to use a powered-industrial vehicle (e.g., forklift, motorized trucks, stock-chasers, stackers, lifts, etc.) to gather items. Here, the sensors, the computing device, and a user interface may be combined or integrated with the vehicle to facilitate picking validation. One possible implementation of a powered-industrial vehicle according to an embodiment of the present invention vehicle is shown in Fig. 2. The powered-industrial vehicle 100 has a platform 110 onto which an item 120 may be placed. When placed on the platform 110, the item 120 is within the field of views 135,145 of a dimensioning system 130 and a digital camera 140. The platform 110 may be part of a scale (not shown) so that items placed on the platform may be measured for weight. The dimensioning system, the digital camera, and the scale are communicatively coupled (e.g., wired or wireless) to a computing device (not shown) that is integrated within the vehicle 100. The computing device is also communicatively coupled to a user interface 150. The user interface, shown in Fig. 2, is a display (e.g., touch display) for presenting and receiving information to/from a user 160. While shown as a touch display, interfacing with a user may be accomplished in a variety ways (e.g., voice headset, printer, etc.).

[0040] Another possible implementation of a powered-industrial vehicle according to an embodiment of the present invention is shown in Fig. 3. Here, the user interface is a voice headset 180 and the computing device is a body-worn mobile computing device (MCD) 170. A wireless link (e.g. BLUETOOTHtm) connects the headset 180 and the MCD 170. The MCD 170 is also wirelessly coupled to the dimensioning system 130, the digital camera 140, and the scale (not shown). The MCD 170 is also wirelessly coupled to a host computer 200 via a wireless network (e.g., WLAN, WIFItm, etc.) 190. In this configuration, the host computer 200 may perform the majority (or all) of the process steps necessary for picking validation. The host computer may also communicate with the user 160 via voice commands via the headset 180 (e.g., picking instructions, alert messages, etc.).

[0041] Multiple workers using multiple vehicles may operate simultaneously in a location (e.g., warehouse). In this case, the host computer (e.g., warehouse management system) may direct and validate picking for each user. The host computer may create a pick list for each user and transmit the user's pick list in its entirety or item-by-item to the user as needed. A picklist is a list of each item to be gathered along with the quantity to be picked. The present invention embraces a pick list that also includes the expected-physical attributes for one or more items in the pick list. The pick list (or portions of the pick list) may be downloaded to the computing device and stored locally in the computing device's memory.

[0042] Fig. 4 graphically depicts a block diagram of an exemplary picking system. The picking system includes a host computer 200 for creating a pick list 210 including at least one item to be picked and a set of expected-physical attributes associated for each item. The host computer transmits the pick list to a mobile computer 250. The mobile computer stores the pick list in its memory (e.g., RAM, HDD, SSD, etc.). The mobile computer includes, or is otherwise communicatively coupled to sensors (dimensioning system 130, scale 240, digital camera 140, etc.) to measure the attributes of a picked item 120. Physical data from the sensors 130,240, 140 are communicated to the mobile computer's processor 220 (e.g., CPU, microprocessor, multi-core processor, ASIC, FPGA, etc.). The processor is configured by software stored in the memory 230 to generate, from the physical data, a set of sensed-physical attributes for the picked item. The processor then retrieves the expected-physical attributes for the picked item from the memory 230 and compares the sensed-physical attributes to the expected-physical attributes. Based on the comparison, the processor creates an alert message (e.g., picking error, picking confirmation, etc.) and transmits this message to the user interface 225 (e.g., display, voice headset, printer, etc.). In some cases, the processor updates the expected-physical attributes measured for the particular picked item. The updated expected-physical attributes may then be stored in the memory 230 and/or communicated to the host computer 200.

[0043] In another embodiment of the picking system, the processing necessary for validation (i.e., comparing the sensed-physical attributes to expected-physical attributes, creating an alert message, etc.) may be performed on the host computer 200.

[0044] In some cases, automatic picking validation may require user interaction. The user 160, for example, may be required to input a picked-item identifier (e.g., serial number, part number, location, etc.) via the user interface 225 so that the correct expected-physical attributes may be retrieved from memory. In operation, a user may input this identifier via a keyboard/number pad, via voice, or by scanning a barcode on the picked item using a barcode reader communicatively coupled to the mobile computer 250.

[0045] The results of the picking validation may be used for other purposes besides picking validation. For example, a warehouse management system may monitor the results of the picking validation to compute worker effectiveness, identify workflow problems, and/or compute volumetric weights of the picked items for shipping (e.g., compute shipping costs).

Claims (9)

Claims:

1. A method for validating a picking operation, comprising: picking an item; placing the picked item on a platform in a field of view of a dimensioning system, the platform having a set of sensors communicatively coupled to a computing device, wherein the platform is supported by a powered-industrial vehicle and wherein the set of sensors comprises: the dimensioning system, a weigh scale, and a camera; sensing, using the set of sensors, physical attributes of the picked item, while the item is placed on the platform and in the field of view of the dimensioning system, wherein the sensed physical attributes of the picked item comprises: (i) a dimensional measurement corresponding to the picked item's shape or size, (ii) the picked item's weight, and (iii) the picked item's visual appearance, wherein at least a portion of the dimensioning system is supported by the powered-industrial vehicle at a higher elevation than the platform so that the field of view of the dimensioning system is directed downwardly toward the platform and multiple sides of the picked item are in the field of view of the dimensioning system, and wherein at least a portion of the camera is supported by the powered-industrial vehicle at a higher elevation than the platform so that a field of view of the camera is directed downwardly towards the platform; receiving, by the computing device, the sensed physical attributes of the picked item; comparing, using the computing device, the sensed physical attributes received by the computing device, to expected-physical attributes stored on the computing device; and if the sensed-physical attributes do not match the expected-physical attributes, then creating a picking error alert; and if the sensed-physical attributes do not match the expected-physical attributes, then updating the expected-physical attributes using the sensed-physical attributes.

2. The method according to claim 1, wherein the picking error alert is (i) an audio alert, a visual alert, or a haptic alert or (ii) combinations thereof.

3. The method according to claim 1, further comprising: if the sensed-physical attributes match using the sensed-physical attributes, and storing the replaced or added physical attributes to the expected-physical attributes on the computing device.

4. The method according to claim 1, further comprising: if the sensed-physical attributes match the expected-physical attributes, then creating a picking-confirmation alert

5. The method according to claim 4, wherein the picking confirmation alert is (i) an audio alert, a visual alert, or a haptic alert or (ii) combinations thereof.

6. An apparatus, comprising: a user interface; a platform configured to support an item and communicatively coupled to a set of sensors, comprising: a dimensioning system, a weigh scale, and a camera, wherein the platform is supported by a powered-industrial vehicle and wherein the set of sensors is configured to: (i) sense physical attributes of the item when the item has been placed on the platform and in a field of view of the dimensioning system, wherein the sensed physical attributes comprise: dimensions of the item sensed by the dimensioning system, visual appearance of the item sensed by the camera, and weight of the item sensed by the weigh scale respectively and wherein at least a portion of the dimensioning system is supported by the powered-industrial vehicle at a higher elevation than the platform so that the field of view of the dimensioning system is directed downwardly toward the platform and multiple sides of the item are in the field of view of the dimensioning system, and wherein at least a portion of the camera is supported by the powered-industrial vehicle at a higher elevation than the platform so that a field of view of the camera is directed downwardly towards the platform; (ii) output the sensed physical attributes; and a computing device communicatively coupled to the user interface and the set of sensors, wherein the computing device is configured to: (i) store the sensed-physical attributes for the item into a set of sensed physical attributes, (ii) retrieve a set of expected-physical attributes for the item from a memory of the computing device, (iii) compare the sensed-physical attributes to the set of expected-physical attributes, (iv) based on the comparison, transmit an alert message to the user interface if the set of sensed physical attributes do not match with the expected physical attributes of the item, and (v) update the set of expected-physical attributes using the sensed- physical attributes, if the sensed-physical attributes does not match the set of expected-physical attributes.

7. The apparatus of claim 6, wherein the alert message is one of (i) an audio alert, a visual alert, or a haptic alert or (ii) combinations thereof.

8. The apparatus of claim 6, wherein the computing device is further configured to transmit an alert message to the user interface if the sensed physical attributes match the expected physical attributes and wherein the alert is (i) an audio alert, a visual alert, or a haptic alert or (ii) combinations thereof.

9. A picking system, comprising: a host computer for creating a pick list, the pick list comprising one or more items to be picked and a set of expected physical attributes corresponding to the one or more items to be picked; and the apparatus of claim 6, wherein the computing device is a mobile computer communicatively coupled to the host computer and the set of sensors, wherein the mobile computer comprises: the user interface, and a memory for storing the pick list