Object arrangement method and device and storage medium

[0039] Example one

[0040] The embodiment of the present invention provides an object organization method. Please refer to figure 1 , The method includes the following steps:

[0041] S102: Acquire the three-dimensional size of each object to be placed.

[0042] Hereinafter, for the convenience of description, the three-dimensional size involved in the embodiment of the present invention is defined as: x*y*z. That is, the three-dimensional size is represented by the numerical values ​​on the x-axis, y-axis, and z-axis in the spatial three-dimensional coordinate system.

[0043] Among them, based on the three-dimensional coordinate system where the storage box is located, the x-axis is used to indicate the length of the storage box in the three-dimensional coordinate system, the y-axis is used to indicate the width of the storage box in the three-dimensional coordinate system, and the z-axis is used to indicate the The height in the three-dimensional coordinate system.

[0044] It should be noted that the placement of the objects to be placed is different, and the three-dimensional dimensions of the objects are defined differently in each direction. For example, if the three-dimensional size of an object to be placed is 10, 5, and 8 (for example only, ignoring the unit), then in a possible placement posture, in the three-dimensional coordinate system where the storage box is located, it can be Expressed as 10*5*8, at this time, it means that its length in the x-axis direction is 10, its width in the y-axis direction is 5, and its height in the z-axis direction is 8; or, on another pendulum In the placement posture, in the three-dimensional coordinate system where the aforementioned finishing box is located, it can be expressed as 8*5*10, which means: its length in the x-axis direction is 8 and its width in the y-axis direction is 5. The height in the z-axis direction is 10. That is to say, when performing the acquisition step, only the three-dimensional size is used to define the size of the space occupied by it, and there is no need to limit the placement posture of each object to be placed.

[0045] S104: According to the three-dimensional size of each to-be-placed object and the three-dimensional size of the sorting box, a simulation calculation is performed on the scene in which each to-be-placed object is put into the sorting box to obtain multiple candidate placement solutions.

[0046] When this step is implemented, it is mainly based on simulation calculations based on different changes of the placement posture, placement order, placement position, etc. of each object to be placed, to obtain a candidate placement plan.

[0047] For example, in specific implementation, even if there is only one object to be placed, at least two candidate placement solutions can be obtained according to the different placement postures. However, the embodiment of the present invention does not specifically limit the specific number of objects to be placed.

[0048] In addition, it should be noted that for any candidate placement scheme, its termination conditions can include any of the following: all the objects to be placed are placed in the finishing box (at this time, there may be remaining space in the finishing box) ; Or, the finishing box can no longer place any remaining objects to be placed (at this time, the remaining space of the finishing box cannot satisfy the placement of any remaining objects to be placed, that is, there may be unplaced objects Objects to be placed in the finishing box).

[0049] S106: Obtain a candidate placement plan with the largest remaining space among the candidate placement plans, and use it as a target placement plan.

[0050] The remaining space refers to the size of the remaining space not occupied by the object to be placed after the finishing box is placed in the object to be placed. Specifically, under the premise that the sorting space of the sorting box and the objects to be placed are fixed, the larger the remaining space, the higher the space utilization rate.

[0051] The embodiment of the present invention is designed for the purpose of making full use of the sorting space. Therefore, in specific implementation, the candidate placement scheme with the largest remaining space is obtained as the target placement scheme, and each object to be placed in the scheme occupies The storage space is the smallest, the storage box can be placed into more objects to be placed, with a higher space utilization.

[0052] Hereinafter, the implementation of the foregoing steps will be specifically described in combination with specific implementation scenarios.

[0053] First, the step of obtaining the three-dimensional size of each object to be placed in S102 can be implemented in multiple ways. The following takes obtaining the three-dimensional size of an object to be placed as an example for description.

[0054] First, the three-dimensional size of the object to be placed is known.

[0055] For example, in an implementation scenario, the three-dimensional size of the object to be identified is marked on the outer surface of the object to be placed, and the outer surface of the object to be identified can be identified by image recognition technology to obtain its three-dimensional size.

[0056] For another example, in another implementation scenario, the three-dimensional size of the object to be identified has been measured by other means and stored in a certain fixed storage location, then only the identification of the object to be identified needs to be obtained from the fixed storage location. Just mark the corresponding three-dimensional size.

[0057] The second method is to obtain the three-dimensional size of the object to be placed through a three-dimensional size obtaining device or module.

[0058] For a specific design, please refer to figure 2 ,in figure 2 On the periphery of the storage box shown, a dot matrix projection module is provided, and the dot matrix projection module is rotatable, which specifically includes a plurality of dot matrix projectors, and each dot matrix projector is used to emit light signals. The embodiment of the present invention does not specifically limit the number of dot matrix projectors included in the dot matrix projection module, where: figure 2 There are 6 dot matrix projectors in the dot matrix projection module shown.

[0059] In addition, the peripheral part of the storage box is also provided with a light sensor. The light sensor is used to sense light and to sense the objects to be placed through light changes.

[0060] The object to be placed can be placed in the front area opposite to these components (light sensor, dot matrix projection module). Wherein, the placement position of the object to be placed is any position within the projection range of the dot matrix projection module and the sensing range of the light sensor. In specific implementation, for ease of processing, a fixed placement position can be designed. The distance between the placement position and the finishing box can be recorded as la.

[0061] in figure 2 In the implementation scene shown, the dot matrix projection module and the light sensor can be used to obtain the three-dimensional size of each object to be placed.

[0062] Specifically, using the dot matrix projection module and the light sensor to obtain the three-dimensional size of each object to be placed can refer to image 3 , S102 includes the following steps:

[0063] S1022: When the light sensor detects an object to be placed, the light signal emitted by the dot matrix projection module is controlled to focus on a focal point on a plane of the object to be placed.

[0064] Wherein, the focus point is a point on the plane of the object to be placed, and the focus point can be defined as required. For example, it can be the midpoint of the plane where the object to be placed currently faces the finishing box (dot matrix projection module and light sensor), or an inflection point at the edge, or any point on the plane.

[0065] When the steps are implemented, the light sensor detects whether there is an object to be placed at the designated position (distance from the storage bin la) through the change of light, and if the object to be placed can be sensed, the light signal from the dot matrix projection module is controlled to focus At the focus point; on the contrary, if the object to be placed cannot be sensed, there is no need to perform control for the dot matrix projection module. In addition, how to determine that the light signal emitted by each dot matrix projector is indeed focused on the focus point is also determined by the light sensor.

[0066] The step of controlling the focus of each dot matrix projector can be realized by controlling the rotation of the dot matrix projector. Specifically, it can include but not limited to the following two designs:

[0067] In a possible design, the dot matrix projection module is fixed on a rotatable shaft provided on the peripheral edge of the finishing box. In this way, when the light signal of the dot matrix projection module is controlled to move on the surface of the object to be placed, it can be realized by controlling the rotation of the rotatable shaft.

[0068] Or, in another possible design, each dot-matrix projector in the dot-matrix projection module is provided with a rotatable axis, and the movement of the optical signal is realized by controlling the rotatable axis.

[0069] And, considering such image 3 In the implementation shown, the size recognition is realized based on the rotation angle and the distance la. Therefore, in a specific implementation, in order to facilitate processing, the initial position of each dot matrix projector can be initialized before the step S1022 is executed. Combining the aforementioned dot matrix projector to realize the rotatable realization method, the initialization can be realized for the rotation axis of each dot matrix projector (inside of itself or set at the outer frame of the storage box), so that the initialized rotation axis is aligned with the initial position .

[0070] S1024: Control the dot matrix projection module to start rotating, so that the light signal moves from the focus point to the surroundings on the plane, until the light sensor detects that the number of light spots on the plane starts to decrease, and the rotation angle is recorded.

[0071] On a projection plane where the light number has been determined, the light sensor detects that the number of light spots of the light signal is constant, and when the edge of the plane is reached, the number of light spots starts to decrease. Therefore, the embodiment of the present invention is based on this To determine whether the edge position of the plane is reached.

[0072] In addition, considering that the plane of the object to be placed facing the finishing box may not have a fixed shape, each dot-matrix projector can expand and move from the focus point to the surrounding according to different directions or angles, and its movement can be It is a straight line method. At this time, a rotation angle is used to represent the rotation angle of a dot matrix projector moving linearly from the focus point to the edge point.

[0073] S1026: According to the rotation angle and the distance between the dot matrix projection module and the object to be placed, the length and width dimensions of the object to be placed on the plane are calculated;

[0074] Specifically, after determining the distance and rotation angle between the two points, the distance (size) from the focus point to the edge point can be determined based on the trigonometric function relationship. As before, the shape and size of the current plane of the object to be placed can be determined through the distance between the edge point and the focus point calculated by multiple dot matrix projectors.

[0075] In this implementation scenario, in order to distinguish the light signals projected by the dot matrix projectors, different identifiers can also be set for the light signals emitted by each dot matrix projector. For example, the dot matrix projectors in the dot matrix projection module can be distinguished according to at least one way: the color of the light signal, the shape of the light beam, or the intensity of the light signal.

[0076] In addition, when the solution is specifically implemented, it can also be implemented by a dot matrix projector in the dot matrix projection module. In this implementation scenario, the dot matrix projector needs to repeat the foregoing steps multiple times to complete the measurement on a plane. Compared with the previous implementation, the cost is lower, but the recognition efficiency is also lower.

[0077] S1028: Adjust the posture of the object to be placed, and repeat the foregoing steps for each posture until the three-dimensional size of the object to be placed is obtained.

[0078] The aforementioned S1022-S1026 can obtain the shape and size of the object to be placed on a plane facing the finishing box. Therefore, in order to obtain its three-dimensional size, the plane facing the finishing box can be adjusted to obtain the shape of each plane And the size, and then, the three-dimensional size of the object to be placed is obtained from this.

[0079] In addition, it should be noted that in the foregoing steps, the rotation axis of the dot matrix projection module can be controlled by the execution device of the present invention, or directly driven and controlled by the light sensor. Among them, if it is controlled by the execution device of this solution, the light sensor needs to send a control signal to the execution device when it senses the change of the light signal, and the execution device controls the rotation axis of the dot matrix projection module after receiving the control signal Rotate.

[0080] It should be noted that due to the different shapes of the objects to be placed, their three-dimensional shapes may be irregular. Therefore, in the specific implementation scenario of object placement, the objects to be placed can be obtained according to the maximum size of each plane. The three-dimensional size of the placement object.

[0081] In summary, the three-dimensional size of the object to be placed can be obtained.

[0082] Furthermore, when the simulation calculation of the placement scene is specifically performed, a three-dimensional coordinate system can be established according to the three-dimensional size of the finishing box, thereby establishing multiple correspondences between the three-dimensional size of each object to be placed and the three-dimensional coordinate system. Furthermore, according to each corresponding relationship, each object to be placed is placed in each direction of the three-dimensional coordinate system in turn, until any unplaced object to be placed can no longer be placed in that direction, and multiple candidates are obtained Placement program. Among them, each corresponding relationship is used to characterize a placement posture of the object to be placed.

[0083] Among them, the correspondence between the three-dimensional size and the three-dimensional coordinate system refers to the definition of the length, width, and height of the three-dimensional coordinate system where the storage box is located under different placement postures of the objects to be placed. In a specific implementation, the corresponding relationship can be implemented through a three-dimensional size representation of the object to be placed. As in the previous example, the three-dimensional size of an object to be placed is expressed as 10*5*8, which means that its length in the x-axis direction is 10, its width in the y-axis direction is 5, and its width in the z-axis direction The height is 8. No longer.

[0084] In addition, the foregoing correspondence relationship is described by taking an object to be placed as an example. In addition to this implementation, in specific implementation, a correspondence relationship can also be used to characterize the placement postures of multiple objects to be placed.

[0085] In the following, with respect to one of the corresponding relationships, the implementation of the candidate placement solution obtained by simulation calculation will be specifically described.

[0086] Considering the specific implementation scenario of item placement, when the placement of the object to be placed is specifically implemented, it can be executed in the direction of the coordinate axis, and the placement shall be carried out according to the principle of size from large to small, in order to place as much as possible. Many objects to be placed.

[0087] Specifically, the implementation steps can refer to Figure 4 , Specifically including the following steps:

[0088] S402: In the first direction of the three-dimensional coordinate system, according to the order of the size of the objects to be placed in the first direction, the objects to be placed are simulated and placed into the sorting box until they can no longer be placed in the first direction. Place any unplaced object to be placed.

[0089] S404: In the second direction of the three-dimensional coordinate system, according to the order of the size of the objects to be placed in the second direction, the remaining objects to be placed are placed in the sorting box by simulation until they are in the second direction. Cannot place any unplaced objects to be placed.

[0090] S406: In the third direction of the three-dimensional coordinate system, according to the order of the third dimension of the objects to be placed, the remaining objects to be placed are placed in the sorting box by simulation, until the third direction can no longer Place any unplaced object to be placed.

[0091] Wherein, the size in the first direction, the size in the second direction and the size in the third direction are obtained by the corresponding indication. Hereinafter, the first direction is the x-axis direction (length direction), the second direction is the y-axis direction (width direction), and the third direction is the z-axis direction (height direction) as an example.

[0092] The three-dimensional size of the object to be placed that has been determined for a corresponding relationship is expressed as: xi*yi*zi, and the three-dimensional size of the finishing box is X*Y*Z. It can be seen that, on the basis of the foregoing corresponding relationship has been set, xi And X are both used to represent the length dimension, yi and Y are used to represent the width dimension, and zi and Z are used to represent the height dimension.

[0093] When performing the design of a candidate placement plan, you can select the placement objects in xi in descending order and place them on the bottom of the sorting box until the x-axis direction (length direction) is full. That is, no other arbitrary objects to be placed can be placed. At this time, the sum of each xi is the occupied space, denoted as △xi, and the difference between X and △xi is used to indicate the x-axis direction Remaining size.

[0094] After that, in the y-axis direction, in the order of yi from largest to smallest, place the objects to be placed on the bottom of the finishing box until the y-axis (length direction) is full, that is, no other objects to be placed can be placed At this time, the sum of each yi is the occupied space, denoted as △yi, and the difference between Y and △yi is used to represent the remaining size in the y-axis direction.

[0095] In the same way, in the z-axis direction, in the order of zi from large to small, place the objects to be placed on the bottom of the finishing box until the z-axis (length direction) is full, that is, no other objects can be placed. Arbitrary objects placed, at this time, the sum of each zi is the occupied space, denoted as △zi, and the difference between Z and △zi is used to represent the remaining size in the z-axis direction.

[0096] Repeat the foregoing steps until any of the foregoing termination conditions is met, then stop placing the objects to be placed, and obtain a complete candidate placement solution.

[0097] As before, the placement postures of the objects to be placed are different, and their corresponding relationships are different. Therefore, based on the foregoing multiple different correspondence relationships, the foregoing steps are repeated to obtain multiple candidate placement solutions.

[0098] In the embodiment of the present invention, in order to improve the space utilization of the sorting box, it is necessary to determine a target solution among the multiple candidate placement solutions obtained above, which depends on the size of the remaining space for screening.

[0099] Like the aforementioned distance, in each candidate placement scheme, there are some remaining sizes, and the space formed by these remaining sizes is the remaining space.

[0100] In a specific field, step S106 can be implemented as follows: in each candidate placement plan, obtain the remaining size of the sorting box in all directions; then, according to the remaining size, obtain the remaining size of each candidate placement plan The candidate placement plan with the largest space is used as the target placement plan. For example, in the candidate placement scheme in the foregoing example, the remaining size in the x-axis direction is X-△xi, the remaining size in the y-axis direction is Y-△yi, and the remaining size in the z-axis direction is Z-△zi .

[0101] If in a candidate placement plan, only one round of placement and arrangement process is carried out, the so-called round of placement and arrangement process means that the x-axis direction, y-axis direction, and z-axis direction are sequentially arranged once, then At this time, there is only one remaining space. At this time, the product of the remaining dimensions of the aforementioned three coordinate axes is obtained to obtain the first volume of the remaining space. Therefore, when the target placement scheme is screened, the candidate placement scheme with the first largest volume can be obtained.

[0102] In addition, considering that the shapes of the objects to be placed are different, there may be more than one remaining space. At this time, you can obtain the volume of each remaining space in each candidate placement scheme according to the remaining size, and then obtain each remaining space. The sum of the volume of the space obtains the remaining space capacity of each candidate placement plan, and then obtains the candidate placement plan with the largest remaining space capacity as the target placement plan.

[0103] Alternatively, the target placement plan can be filtered out based on the difference between the total volume of the finishing box and the sum of the placed objects.

[0104] Based on the foregoing process, the embodiment of the present invention can determine a target placement solution with the largest remaining space according to the three-dimensional size of the object to be placed and the storage box.

[0105] As a result, during specific implementation, the target placement solution can be output to the user end, so that the user end can implement the object placement according to the target placement solution, and quickly find the placed objects.

[0106] Among them, the target placement scheme can have multiple output modes.

[0107] In a possible design, the target placement solution may consist of the identification, placement posture and coordinates of the object to be placed. The output result can be a placement list or a placement sequence. In this way, when manual or automated equipment implements item placement according to the target placement plan, the items to be placed can be placed into the sorting box in sequence according to the placement list or placement sequence.

[0108] In another possible design, the embodiment of the present invention provides another more pictorial output method: according to the target placement scheme, a three-dimensional placement image in the finishing box is generated.

[0109] Specifically, generating the three-dimensional placement images in the sorting box can be achieved by following the steps: collect the images of the objects to be placed, and then combine the images of the objects to be placed according to the target placement scheme to get the sorting The three-dimensional placement image in the box.

[0110] Among them, the image of each object to be placed can be collected through figure 2 The image collector shown is implemented. Or, it can be achieved by directly recalling existing data.

[0111] This implementation mode can intuitively display the position and placement posture of each object to be placed in the finishing box, so that if the user needs to take a target object in the finishing box, the intuitively displayed three-dimensional The placement image quickly determines the location of the target object.

[0112] In this application scenario, in order to make it easier for the user to take and process, in addition to directly outputting the three-dimensional placement image of the storage box, it is also possible to further output the three-dimensional placement image containing the target object mark. At this time, the method may further include the following steps: receiving a retrieval request for any target object in the sorting box, thereby determining and marking the position of the target object in the three-dimensional placement image, and further, outputting a three-dimensional return containing the location mark Set the image.

[0113] The position mark may be a highlight mark, that is, the target object is displayed protrudingly in the three-dimensional placement image. For example, color contrast can be used to display the three-dimensional images of other objects in grayscale, and the target object can be displayed in color, so as to mark the position of the target object.

[0114] It can be understood that some or all of the steps or operations in the above-mentioned embodiments are only examples, and the embodiments of the present application may also perform other operations or variations of various operations. In addition, each step may be executed in a different order presented in the foregoing embodiment, and it may not be necessary to perform all operations in the foregoing embodiment.