OPTIMIZATION OF A FILLING SYSTEM AND A FILLING PROCESS

Abstract

Optimization of a Filling System and a Filling Process. The present invention relates to an optimized filling device for containers with bulk products, as well as a corresponding process. [Fig. 1]

Description

Title of the invention: OPTIMIZATION OF A FILLING SYSTEM AND A FILLING METHOD Technical field

[0001] The present invention relates to a device and method for the optimized filling of containers with bulk products, such as blocks, beans, cereals, seeds, powders, etc. (and in particular agricultural products, such as frozen agricultural products). The aim of the invention is to make the filling of containers with bulk products more efficient and more accurate through automated image control and processing. STATE OF THE ART

[0002] A problem with known systems is that, for the moment, they are mainly adjusted manually by an operator. After leveling the quantity of bulk products poured, it is thus possible to see that there is too much or too little product intended in the container.

[0003] In the first case, it must be removed manually, which results in a loss. In addition, one is dependent on the operator who must report that there has been an overfill (which he often does not do due to time pressure), but who can only rarely quantify it, and certainly not correctly, which results in a prolonged period of compensations where the containers are either overfilled or underfilled.

[0004] Moreover, at this stage it is already a final product, which has gone through a large number of production stages (e.g. for agricultural products: washing, peeling, blanching, sorting, freezing, etc.), thus representing an even greater economic loss.

[0005] In the second case, this is usually no longer compensated because the container is leveled at another position and the filling station is already used by other containers. In such a situation, the available container volume is used suboptimally, and there is a loss of storage space. Often, these are relatively small loss volumes, but given the enormous total volumes, this quickly accumulates to a very large absolute loss volume. Furthermore, one is then again dependent on the operator to adjust the feed, but this is usually not quantifiable by the operator.

[0006] The present invention aims to find a solution to at least some of the aforementioned problems. Summary of the invention

[0007] The present invention relates to a device and a method for the optimized filling of containers with bulk products, in particular agricultural products, such as frozen agricultural products. The objective of the invention is to make the filling of containers with bulk products more efficient and more precise by means of automated image control and processing.

[0008] The device comprises several components that work together to achieve the optimized filling process. First, there is a feeding unit that is designed to place the bulk products from above into a container located in a lower position in a controlled manner. This allows for an orderly and structured filling of the container. The feeding can be done by a plurality of systems, such as a screw, a shaker, a conveyor, a “Redler®”, a chute, air transport, etc.

[0009] To monitor the filling of the container, the device comprises an image capture system. This system is composed of at least one image sensor, preferably a camera, which is oriented towards the lower position of the container. The image sensor(s) capture images of at least a portion of the bulk products in the container. The image sensors make it possible to continuously (or almost continuously) monitor the filling state.

[0010] The captured images are then processed by an image processing system. This system analyzes the images and estimates the total relative and / or absolute useful volume of the bulk products in the container. The estimation is based on the characteristics and properties of the bulk products that are visible in the images.

[0011] The optimized filling process is controlled by a processing unit. This processing unit receives the estimate of the volume of the bulk products in the container from the image processing system. Based on this estimate, the processing unit controls the feeding unit to add or not add bulk products into the container. The feeding stops when the estimate reaches a predetermined value. This predetermined value depends on the type of bulk products that are filled, allowing the device to flexibly adapt to different types of bulk products.

[0012] The automated filling process has several advantages. First, it minimizes the empty space after filling the container, which avoids wasting valuable space. This maximizes the efficiency of transport and storage capacity.

[0013] In addition, the device prevents overfilling of the container, which can cause problems such as product loss, damaged containers and possible damage to the contents due to excessive stacking. Especially for frozen products, this represents a real danger, because after filling, the products can expand due to the freezing process, weakening or thereby damaging the container. Thanks to this improvement, operational costs are reduced and product quality is maintained.

[0014] An additional advantage of the proposed device and method is that it is not necessary for an operator to be present in a hazardous area, where the filling process takes place and where forklifts, AGVs (automatically guided vehicles, therefore not piloted) and other machinery often circulate to move the containers. There is no need to manually monitor and readjust the filling process, because the automated system controls and directs the feeding unit based on the image information obtained.

[0015] In another embodiment, the captured images can be used for additional applications, such as, on the one hand, detecting the presence of a container (and thus the automated start and stop of filling), and, on the other hand, recognizing the container (with or without additional subtypes, such as different volumes of the same container type), thus allowing the system to automatically implement the predetermined 'correct' value.

[0016] In addition, the verification of certain quality conditions can be automated based on these images, such as the presence of a film in the container, and whether it is correctly installed (on each edge, with sufficient margin, etc.), the correct position of the container (the filling system is often very precisely positioned, starting from a specific positioning of the container, to prevent the products from falling out of the container or colliding with each other), environmental hygiene, as well as other elements, such as the detection of the presence of personnel in certain areas, which can cause the feeding unit to automatically deactivate for safety reasons.

[0017] Optionally, the device may be equipped with an equalization system. This system comprises at least one automated manipulator, optionally robotic, adapted for the automated movement of the bulk products in the container towards a substantially flat horizontal filling. This allows for uniform distribution of the bulk products and minimizes height differences within the container. The equalization system may also use a second image capture system, comprising at least one image sensor, to capture images of the surface of the bulk products in the container. These images are used to control the equalization system, and if necessary, the predetermined value of the volume of products fed at which the feeding stops, may be adjusted based on the captured images.

[0018] The automated manipulator may comprise one or more mechanical arms (possibly robotic arms), capable of moving so as to reach each position (within a certain range). This is possible, for example, with an arm robotic with at least three degrees of freedom. A Cartesian (robotic) arm that can move in the xyz directions could in principle meet this requirement, as could articulated (robotic) arms.

[0019] The described device can be used in combination with the proposed method for the optimized filling of containers with bulk products. The method comprises the controlled feeding of the bulk products into the container based on the estimation of the image processing. The filling process stops when the estimation reaches the predetermined value.

[0020] The images may be captured with at least one image sensor, which is positioned to determine or extrapolate a three-dimensional image of the bulk products in the container. Thus, the volume estimation may be based on the detailed three-dimensional image.

[0021] The proposed device and method provide an efficient and accurate solution for optimizing the filling of containers with bulk products, thereby minimizing empty space and avoiding overfilling. In addition, they eliminate the need for manual monitoring and control, while ensuring a safer working environment without an operator in hazardous areas. DESCRIPTION OF FIGURES

[0022] [Fig. 1] [Fig. 1] shows a possible embodiment of a filling line on which the improved filling system is implemented.

[0023] [Fig.2] [Fig.2] shows an embodiment of the equalization system implemented in a production line. DETAILED DESCRIPTION

[0024] Unless otherwise indicated, all terms used in the disclosure of the invention, including technical and scientific terms, have the same meaning as commonly understood by a person skilled in the art of the invention. The following terms are explicitly explained to better appreciate the teaching of the present invention.

[0025] "About," as used herein, referring to a measurable value such as a quantity, parameter, time duration, and so forth, is intended to encompass variations of + / - 20% or less, preferably + / - 10% or less, more preferably + / - 5% or less, even more preferably + / - 1% or less, and still more preferably + / - 0.1% or less of and relative to the specified value, to the extent that such variations are appropriate for carrying out the disclosed invention. However, it will be understood that the value to which the modifier "about" refers is also specifically disclosed.

[0026] The terms "include", "comprising", "comprises" and "composed of" as used herein are synonymous with "included", "including", "comprises" or "contain", "containing", "contains" and are inclusive or flexible terms that specify the presence of the following, and do not exclude the presence of other components, features, elements, members, steps, known in the art or disclosed therein.

[0027] The enumeration of ranges of numerical values ​​by endpoints includes the numbers and fractions encompassed within that range, as well as the cited endpoints.

[0028] In a first aspect, the invention relates to a device for the optimized filling of containers with bulk products as described herein, namely a device for the optimized filling of containers with bulk products, preferably agricultural products, preferably frozen agricultural products comprising:

[0029] a. a feeding unit, for the controllable feeding of bulk products from above into a container in a lower position;

[0030] b. an image capture system, comprising at least one image sensor, preferably a camera, facing the lower position, for capturing images of at least a portion of the bulk products in the container;

[0031] c. an image processing system for processing the captured images, the processing comprising an estimation of the relative and / or absolute total useful volume of the bulk products in the container based on the captured images;

[0032] d. a processing unit configured to control the feeding unit, depending on the estimate, to add or not add bulk products into the container, the processing unit stopping the feeding when the estimate reaches a predetermined value;

[0033] the image capture system comprising at least two image sensors, at least two of the at least two image sensors being positioned relative to a central point of the lower position at an intermediate angle of at least 60°.

[0034] The optimized filling takes into account certain user requirements. For example, it may be set to provide a free space of up to about 2 cm below the rim of the containers. This may be to allow, among other things, the placement of a lid, stacking of the containers without damaging the bulk products, allowing for an expected expansion of the bulk products, etc. The objective is that the container is nevertheless filled optimally taking into account these requirements, and that when a free space of 2 cm is required, this is not 3 cm or 1 cm in practice.

[0035] The containers in question can take any possible shape, material or type and are used to contain products. For example, they can be (wooden) crates, optionally provided with (plastic) film inside (so-called a Flemish crate), bulk boxes or octabins, crates, etc. The invention does not only apply to substantially rigid containers, although the advantages are most pronounced there, but also to flexible containers, such as big bags, bags, tarpaulins, etc. Preferably, however, these are rigid containers.

[0036] The term "bulk products" refers to solid products whose particle size varies from a few micrometers (in the case of powders) to several centimeters, or even decimeters, when it comes to pieces and clearly distinct units.

[0037] In principle, the invention also applies to solid products of larger particle size. In this case, it may be necessary to provide an equalization system adapted to the fragility, weight and / or size of the particles.

[0038] Preferably, at least 2, 3 or even 4 image sensors are provided, placed at different positions around the lower position where the container is placed during filling. The aim is to obtain as correct a three-dimensional image as possible of the bulk products in the container, in order to estimate the volume as accurately as possible. Preferably, the image sensors are positioned so that the axes of their orientations (central axis of the sensor direction) intersect somewhere above the lower position, the axes of at least two of the image sensors not being in the same vertical plane, which allows developing a three-dimensional perspective, for example, one image sensor is positioned at 12 o'clock relative to the central container, and another image sensor is positioned at 3 or 9 o'clock.Preferably, the image capture system is adapted to capture a 360° image of the bulk products in the container.

[0039] The image capture system comprises at least two image sensors, at least two of the at least two image sensors being positioned relative to a central point of the lower position at an intermediate angle of at least 60°, preferably at least 75° and more preferably at least 90°.

[0040] In one possible embodiment, there are at least three image sensors distributed symmetrically around the lower position. Depending on the number, this may be at an intermediate angle of 120°, 90°, 72°, 60°, etc. Preferably, however, the number is limited.

[0041] In one possible embodiment, a single image sensor is used. This may optionally be placed above the container, or laterally to it (while remaining at or above the edge of the container).

[0042] In some embodiments, multiple lower positions are monitored with a single image capture system, and multiple power systems are provided.

[0043] In a preferred embodiment, at least one image sensor is positioned laterally relative to the lower position and at or above the height of the container height. The advantage is that this makes it easier to estimate the height, which means less calculation work during processing. Another advantage is improved hygiene. By not placing sensors above the container, nothing can fall into the container either. Dust or dirt can accumulate on the sensor, which can then detach and fall into the container.

[0044] Preferably, the image sensors are oriented horizontally or downwards. The downward orientation may for example be at an angle between 5° and 60°, or between 10° and 45° or between 15° and 30°. The above angles may also be considered separately as minimum or maximum angles.

[0045] By orienting the image sensors downwards, it is also possible to observe the volume present in the container, so that it is not necessary to make (or make fewer) assumptions about it during the estimation. Typically, it is necessary to have at least two distinct viewpoints, not located in the same vertical plane, so that the three-dimensional image can be extrapolated, for example by assuming symmetry.

[0046] In a preferred embodiment, at least one image sensor is positioned vertically above the lower position and above the height of the containers, preferably at the center point above the lower position, with the image sensor preferably facing vertically downward.

[0047] This position makes it possible to have a complete view of the upper surface of the products in the container at all times. Thus, the number of image sensors required can be limited, so that even one may be sufficient. Based on the captured images, it is possible to determine the volume of the bulk products in the container.

[0048] More preferably, the image sensor is oriented vertically downwards, the image processing system being configured to calculate the relative and / or absolute total usable volume based on an estimated height profile of the upper surface of the bulk products in the container. The vertical orientation of the sensor ensures that every point is visible in the container, with the sensor essentially oriented downwards.

[0049] It is understood that, in certain embodiments, the image sensors located both directly (vertically) above and laterally relative to the lower position where the container will be placed can be used in combination, in order to facilitate and accelerate the determination of the volume.

[0050] In a preferred embodiment, at least one image sensor is height adjustable and preferably automatically adjusts to the height of the container in the lower position. This allows it to adapt to different types of containers (of different heights).

[0051] In a preferred embodiment, the estimation is made on the basis of the volume of bulk products which protrude from the edge of the container.

[0052] Alternatively, bulk products below the container edge can also be taken into account, if downward-facing image sensors are provided. In this way, the volume can be estimated even more precisely.

[0053] In a preferred embodiment, the predetermined value is defined as at least 100% of the useful volume of the container (maximum contents that can be used). As discussed in more detail, the stacked volume is far from optimized, and the volume will decrease after "settling" and equalization, due to more compact stacking (as well as compression of the product, which should be avoided as much as possible).

[0054] In a preferred embodiment, the bulk products are adjustable in the device depending on the type, and the predetermined value depends on the type of bulk products. Preferably, other factors, such as the type of product, its size, etc., may also be taken into account.

[0055] In addition, the storage density when feeding into the container can also be taken into account, compared to the storage density of the bulk products after equalization, which is generally higher. Thus, the difference in this storage density will vary according to each type of product (e.g., cauliflower versus peas, with also further subdivisions such as large peas versus fine peas, frozen peas versus unfrozen peas, etc.). Depending on this, the predetermined value can be set to a certain additional volume compared to the useful volume in the container (more than 100%), this additional volume being larger or smaller depending on additional information such as the type of bulk products, etc.

[0056] By allowing such further subdivision, and especially by making it adjustable, the system can also become self-learning and, thanks to the feedback, automatically return to previously confirmed or corrected values ​​of identical or similar types of bulk products. For similar types, one can for example consider the size and / or shape of the pieces.

[0057] An additional advantage is that the system can optionally use image recognition to identify bulk products during or before they are fed into the container, and thus select or even adjust on the fly the correct predetermined value during filling (if it detects for example that they are small fine peas when it had initially assumed large peas, due to misidentification or incorrect entry by an operator).

[0058] In a preferred embodiment, the device further comprises an equalizing system, comprising an automated, optionally robotic, manipulator adapted to automatically move the bulk products in the container towards a substantially flat horizontal filling in the container, preferably the manipulator comprising at least one leveling bar. This equalizing system may be provided at the lower position where the container is filled, or at another position, so that equalizing can be carried out while another container is being filled, for example.

[0059] This step involves filling the container optimally so that when the stacked bulk products are distributed, the container is filled to the rim but not beyond.

[0060] This is done for example by a leveling bar, but can also be achieved by a gripping unit that handles the products one by one. It is also possible to opt for one or more different leveling bars and / or manipulators per type of bulk products (fragile or not, large or not, etc.). The advantage over manual leveling is obvious.

[0061] Preferably, the manipulator comprises an interchangeable part for handling bulk products. Thus, the part can be changed when another type of bulk products is to be handled (between containers), in order to avoid cross-contamination, but also to use a specific tool that is more efficient or better suited to a certain type of bulk products. Again, with human operators, there is a risk that this exchange is not carried out due to negligence, lack of time or other reasons. For example, the part can be automatically changed between each container or between successive containers with different contents. It can, for example, be cleaned in a specific subsystem or another system after each exchange, or simply be collected for later cleaning in larger quantities.

[0062] In a preferred embodiment, the device comprises a second image capturing system, comprising at least one image sensor, preferably a camera, oriented towards the surface of the bulk products in the container, for capturing images of said surface, the system also being controlled on the basis of the captured images of said surface. The second image capturing system may be the same as the image capturing system for capturing the bulk products in the container during filling, particularly preferably if the equalization is carried out at the same location as the filling. Alternatively, it may be a different image capturing system if the equalization is done at a different location. The same remarks as for the original image capturing system apply in terms of positioning and number of image sensors, in order to form a precise image and optimally control the manipulator.

[0063] Equalization can be carried out alternatively or additionally using vibrations, which is already partly done during transport (with a forklift, on the conveyor, etc.).

[0064] In a preferred embodiment, the predetermined value is adjusted based on the captured images of said surface after achieving a complete and substantially flat horizontal filling. As previously indicated, when, during filling, an under-quantity of bulk products is constant, this can be quantified to some extent based on the images from the second image capturing system and be fed back to the feeding system. Thus, the predetermined value can then be increased according to the under-quantity. Conversely, the same can be done for an over-quantity, when there is too much bulk product, and the predetermined value can be reduced according to the over-quantity. Ideally, this is linked to the type of bulk products, so that the device develops increasingly correct predetermined values ​​for each type of bulk products and can refine them further.By working from images, it is also possible to quantify this value effectively, instead of relying on a manual operator who has to estimate it by guesswork.

[0065] In a second aspect, the invention relates to a method for the optimized filling of containers with bulk products, preferably agricultural products, preferably frozen agricultural products, comprising the following steps: a. controllable feeding of bulk products into a container in a lower position; b. capturing images by means of at least one image sensor, preferably a camera, of at least a portion of the bulk products in the container; c. processing of the captured images, the processing including an estimation of the total relative and / or absolute useful volume of the bulk products in the container, based on the captured images; d. controlling the feeding unit based on the estimate to add or not add bulk products into the container, the processing unit stopping feeding when the estimate reaches a predetermined value.

[0066] The estimation of the total relative and / or absolute useful volume of the bulk products in the container is then based on at least two images, captured with at least two image sensors which are positioned relative to a central point of the lower position at an intermediate angle of at least 60°.

[0067] In a preferred embodiment, the predetermined value depends on the type of bulk products, and the type of bulk products can be set manually and / or automatically determined. As discussed, automatic determination can be done by object recognition based on images captured during filling.

[0068] In a preferred embodiment, the predetermined value is at least 100% of the container volume. As previously indicated, the stacked volume will be reduced upon equalization.

[0069] In a preferred embodiment, the method is carried out using a device according to the first aspect of the invention.

[0070] Preferably, a device is used having an equalization system, in which the predetermined value is adjusted on the basis of the images captured by the second image capturing system of said surface after having reached complete and substantially flat horizontal filling.

[0071] Even more preferably, the capturing of the images is done with at least two image sensors, positioned to determine or extrapolate a three-dimensional image of the bulk products in the container, and in which the estimate is determined on the basis of the three-dimensional image. Examples

[0072] [Fig.l] shows a possible configuration of a line with the device of the invention. A number of feeding systems are provided (6 in this case) which fill 6 containers at lower positions. These containers are monitored by an image capture system (not visible) which determines the volume of bulk products in the container and controls (stops or not) the feeding system for the container accordingly.

[0073] Once the feeding is stopped, when the volume of bulk products reaches the predetermined value, a transport unit, such as the forklift in the figure, can move the container to an equalization system, as shown in [Fig.2].

[0074] The equalization system may be part of a conveyor, the conveyor itself also contributing to the equalization due to the small vibrations that occur during transport. In a first installation position (1), a newly filled container is placed and moved to the equalization position (3) where the upper surface of the bulk products in the container is equalized. Between the installation position (1) and the equalization position (3), one or more intermediate positions (2) may be provided, where the equalization process is often already partially carried out.

[0075] After equalization, the container is moved further to a closing position (5), with again an intermediate position (6), with typically a further 'deposition' of the bulk products, and then to exit positions (7).

[0076] It is assumed that the present invention is not limited to the embodiments described above and that certain modifications or changes may be made to the examples described without re-evaluating what is written herein. For example, the present invention has been described with reference to frozen foods, but it should be clear that the invention can be applied to all types of foods or, more generally, to all types of bulk products. The invention can also be extended to smaller items, such as fruits, vegetables (e.g., cauliflower, broccoli, etc.).

Claims

Claims

1. A device for the optimized filling of containers with bulk products, preferably agricultural products, and more preferably frozen agricultural products, comprising: a. a feeding unit, for the controllable feeding of the bulk products from above into a container in a lower position; b. an image capturing system, comprising at least one image sensor, preferably a camera, oriented towards the lower position, for capturing images of at least a portion of the bulk products in the container; c. an image processing system for processing the captured images, the processing comprising an estimation of the relative and / or absolute total useful volume of the bulk products in the container based on the captured images; d.a processing unit configured to control the feeding unit, based on the estimate, to add or not add bulk products into the container, the processing unit stopping the feeding when the estimate reaches a predetermined value; e. characterized in that the image capture system comprises at least two image sensors, at least two of the at least two image sensors being positioned relative to a central point of the lower position at an intermediate angle of at least 60°.

2. The device according to claim 1, wherein the image capturing system comprises at least two image sensors, at least two of the at least two image sensors being positioned relative to a central point of the lower position at an intermediate angle of at least 75° and preferably at least 90°.

3. The device according to any one of the preceding claims 1 to 2, wherein the at least one image sensor is positioned laterally relative to the lower position and at or above the height of the container.

4. The device according to any one of the preceding claims 1 to 3, wherein the at least one image sensor is positioned vertically above the lower position and above the height of the containers, preferably at the central point above the bottom position, with the image sensor preferably oriented vertically downwards.

5. The device according to the preceding claim 4, wherein the image sensor is oriented vertically downwards, the image processing system being configured to calculate the relative and / or absolute total useful volume on the basis of an estimated height profile of the upper surface of the bulk products in the container.

6. The device according to any one of the preceding claims 1 to 5, wherein the image capturing system is adapted to capture a 360° image of the bulk products in the container.

7. The device according to any one of the preceding claims 1 to 6, wherein the at least one image sensor is height adjustable and preferably automatically adjusts to the height of the container in the lower position.

8. The device according to any one of the preceding claims 1 to 7, wherein at least one, and preferably each, of the at least one image sensor is partially oriented downwards towards the lower position and is adapted to capture a portion of the interior of the container.

9. The device according to any one of the preceding claims 1 to 8, wherein the device further comprises an equalizing system, comprising an automated, preferably robotic, manipulator adapted for the automated movement of the bulk products in the container towards a substantially flat horizontal filling in the container, preferably the manipulator comprising at least one leveling bar.

10. The device according to the preceding claim 9, further comprising a second image capturing system, comprising at least one image sensor, preferably a camera, oriented towards the surface of the bulk products in the container, for capturing images of said surface, the equalization system being controlled on the basis of the captured images of said surface.

11. A method for optimizing the filling of containers with bulk products, preferably agricultural products, preferably frozen agricultural products, comprising the following steps: a. controllably feeding bulk products into a container in a lower position; b. capturing images by means of at least one image sensor, preferably a camera, of at least a portion of the bulk products in the container; c. processing the captured images, the processing comprising an estimation of the total relative and / or absolute useful volume of the bulk products in the container, based on the captured images; d. controlling the feeding unit depending on the estimation to add or not add bulk products into the container, the processing unit stopping the feeding when the estimation reaches a predetermined value.characterized in that the method comprises the step of estimating the relative and / or absolute total useful volume of the bulk products in the container on the basis of at least two images captured by means of at least 2 image sensors positioned relative to a central point of the lower position at an intermediate angle of at least 60°, preferably at least 75° and even more preferably at least 90°.

12. The method according to the preceding claim 11, wherein the predetermined value depends on the type of bulk products, and wherein the type of bulk products can be set manually and / or determined automatically.

13. The method of any preceding claim 11 or 12, wherein the predetermined value is at least 100% of the useful volume of the container.

14. The method of any one of preceding claims 11 to 13, wherein the estimation is based on the volume of bulk products which protrude from the edge of the container.

15. The method according to any one of the preceding claims 11 to 14, wherein the method is carried out by means of a device according to one of the preceding claims 1 to 10.

16. The method according to any one of the preceding claims 11 to 15, wherein the method is carried out by means of a device according to any one of the preceding claims 9 to 10, wherein the predetermined value is adjusted on the basis of the images captured by the second image capture system of said surface after having reached complete and substantially flat horizontal filling.

17. The method of any one of the preceding claims 11 to 16, wherein the capturing of the images is carried out by means of at least two image sensors, positioned so as to determine or extrapolate a three-dimensional image of the bulk products in the container, and wherein the estimate is determined on the basis of the three-dimensional image.

Images