Robot loading and stacking method and device, electronic equipment and storage medium

Abstract

The invention discloses a robot loading and stacking method and device, electronic equipment and a storage medium, and belongs to the technical field of data processing. The method comprises the steps that loading space information, target object information and placement attitude information of a target object are obtained, the loading space information is used for describing information of a three-dimensional space formed by combining one or more hexahedrons, and the target object information is used for describing information of a cube with a length value, a width value and a height value; based on a knapsack algorithm and a multi-target optimization algorithm, according to the loading space information, the target object information and the placement attitude information of the target object, generating a target object arrangement scheme; and according to the target object arrangement scheme, the movement track and the operation posture of the robot from the feeding point to the placing point are planned, and a loading and stacking scheme is obtained.

Description

Technical Field

[0001] This application belongs to the field of communication technologies, and particularly relates to a method, apparatus, electronic device, and storage medium for robot loading and palletizing. Background Art

[0002] In high-end mold manufacturing and logistics, due to the complex geometric shapes and large weights of mold parts, strict requirements are imposed on stacking and loading operations. Currently, mold loading operations mostly rely on manual labor or traditional mechanical assistance, resulting in problems such as low efficiency and unreasonable stacking, and are prone to causing mold collisions and damage. Especially when loading into containers, how to efficiently utilize the limited space to achieve compact stacking and stable placement of molds remains a technical challenge.

[0003] Under the existing technology, there is a lack of intelligent planning means for high-end mold loading and palletizing. Due to the diverse shapes and specifications of molds, traditional palletizing algorithms have large computational amounts and low optimization efficiency when dealing with objects of complex shapes, and cannot meet the requirements of efficient and accurate stacking. At the same time, existing loading systems do not adequately consider the space utilization rate and loading stability of containers, resulting in molds being prone to slipping and falling during transportation, increasing the risk of damage and logistics costs.

[0004] With the development of high-end mold manufacturing towards refinement and automation, the need for intelligent palletizing and loading technology is urgent. Although robot automatic palletizing technology brings new directions, existing robot palletizing systems are mostly applicable to objects of simple shapes. For molds with complex geometric shapes, there is a lack of planning algorithms and control logics with strong adaptability and high efficiency, making it difficult to achieve intelligent loading operations. Summary of the Invention

[0005] The objective of the embodiments of this application is to provide a method, apparatus, electronic device, and storage medium for robot loading and palletizing, which can solve the problem that it is currently difficult to achieve intelligent loading operations.

[0006] In a first aspect, the embodiments of this application provide a method for robot loading and palletizing, which includes:

[0007] Obtain loading space information, target object information, and the placement attitude information of the target object. The loading space information is used to describe the information of a three-dimensional space composed of one or more hexahedrons, and the target object information is used to describe the information of a cube with length, width, and height values;

[0008] Based on the knapsack algorithm and multi-objective optimization algorithm, generate a target object arrangement plan according to the loading space information, target object information, and the placement attitude information of the target object;

[0009] According to the target object arrangement plan, plan the movement trajectory and operation attitude of the robot from the loading point to the placement point to obtain a loading and palletizing plan.

[0010] Second aspect, an embodiment of the present application provides a robot loading and palletizing device, which includes:

[0011] An acquisition module, configured to acquire loading space information, target object information, and placement attitude information of the target object. The loading space information is used to describe information of a three-dimensional space composed of the combination of one or more hexahedrons, and the target object information is used to describe information of a cube with a length value, a width value, and a height value;

[0012] A generation module, configured to generate a target object arrangement plan based on the knapsack algorithm and the multi-objective optimization algorithm according to the loading space information, the target object information, and the placement attitude information of the target object;

[0013] A planning module, configured to plan the movement trajectory and operation attitude of the robot from the loading point to the placement point according to the target object arrangement plan to obtain a loading and palletizing plan.

[0014] Third aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory. The memory stores a program or instruction that can run on the processor. When the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented.

[0015] Fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0016] Fifth aspect, an embodiment of the present application provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is configured to run a program or instruction to implement the method described in the first aspect.

[0017] Sixth aspect, an embodiment of the present application provides a computer program product, which is stored in a storage medium. The program product is executed by at least one processor to implement the method described in the first aspect.

[0018] In an embodiment of the present application, by obtaining the loading space information, the target object information, and the placement attitude information of the target object, the loading space information is used to describe the information of a three-dimensional space composed of a combination of one or more hexahedrons, and the target object information is used to describe the information of a cube with a length value, a width value, and a height value, providing necessary data support for subsequent steps and ensuring that the planning of the entire loading and stacking scheme is based on the actual loading space and the situation of the target object. Based on the knapsack algorithm and the multi-objective optimization algorithm, according to the loading space information, the target object information, and the placement attitude information of the target object, multiple optimization objectives can be balanced, and the optimal target object arrangement scheme can be found, which can not only make full use of the loading space, improve the space utilization rate, achieve a good balance among multiple optimization objectives, make the stacking stable and efficient, thereby improving the quality and efficiency of the entire loading and stacking process. According to the target object arrangement scheme, plan the movement trajectory and operation attitude of the robot from the loading point to the placement point to obtain the loading and stacking scheme. A reasonable movement trajectory can reduce the movement time and energy consumption of the robot, and at the same time avoid safety problems such as collisions. A suitable operation attitude can ensure that the robot stably grabs and places the target object, improving the accuracy and stability of the stacking. Thus, intelligent loading operations can be realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flowchart of a robot loading and stacking method provided by an embodiment of the present application;

[0020] Figure 2 is a flowchart of another robot loading and stacking method provided by an embodiment of the present application;

[0021] Figure 3 is a flowchart of yet another robot loading and stacking method provided by an embodiment of the present application;

[0022] Figure 4 is a structural diagram of a robot loading and stacking device provided by an embodiment of the present application;

[0023] Figure 5 is a schematic hardware structure diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] The technical solutions of the embodiments of the present application will be clearly described below with reference to the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application fall within the scope of protection of the present application.

[0025] In the description and claims of this application, terms such as "first" and "second" are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances, so that the embodiments of this application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually of the same category, and do not limit the number of objects. For example, the first object can be one or more. In addition, "and / or" in the description and claims means at least one of the connected objects, and the character " / ", generally represents an "or" relationship between the associated objects before and after.

[0026] In view of the problems in the related art, the embodiments of this application provide a robot loading and palletizing method, device, electronic device and storage medium, which can solve the problem that it is currently difficult to realize intelligent loading operations in the related art.

[0027] The following combines the drawings and details the robot loading and palletizing method provided by the embodiments of this application through specific embodiments and their application scenarios.

[0028] Figure 1 It is a flowchart of a robot loading and palletizing method provided by the embodiments of this application.

[0029] As Figure 1 shown, the robot loading and palletizing method may include step 110-step 130. This method is applied to a robot loading and palletizing device, and is specifically as follows:

[0030] Step 110, obtain loading space information, target object information and the placement attitude information of the target object. The loading space information is used to describe the information of a three-dimensional space composed of one or more hexahedrons, and the target object information is used to describe the information of a cube with a length value, a width value and a height value;

[0031] Step 120, based on the knapsack algorithm and the multi-objective optimization algorithm, generate a target object arrangement plan according to the loading space information, the target object information and the placement attitude information of the target object;

[0032] Step 130, according to the target object arrangement plan, plan the movement trajectory and operation attitude of the robot from the loading point to the placement point to obtain a loading and palletizing plan.

[0033] In the embodiments of the present application, by obtaining the loading space information, the target object information, and the placement attitude information of the target object, the loading space information is used to describe the information of a three-dimensional space composed of one or more hexahedrons, and the target object information is used to describe the information of a cube with a length value, a width value, and a height value, providing the necessary data support for subsequent steps and ensuring that the planning of the entire vehicle loading and palletizing scheme is based on the actual loading space and the target object situation. Based on the knapsack algorithm and the multi-objective optimization algorithm, according to the loading space information, the target object information, and the placement attitude information of the target object, multiple optimization objectives can be balanced, and the optimal target object arrangement scheme can be found. This can not only make full use of the loading space, improve the space utilization rate, achieve a better balance among multiple optimization objectives, make the palletizing stable and efficient, but also improve the quality and efficiency of the entire vehicle loading and palletizing process. According to the target object arrangement scheme, the movement trajectory and operation attitude of the robot from the loading point to the placement point are planned to obtain the vehicle loading and palletizing scheme. A reasonable movement trajectory can reduce the movement time and energy consumption of the robot, and at the same time avoid safety problems such as collisions. A suitable operation attitude can ensure that the robot stably grasps and places the target object, improving the accuracy and stability of palletizing. Thus, intelligent vehicle loading operations can be realized.

[0034] Step 110 is involved: Obtaining the loading space information, the target object information, and the placement attitude information of the target object provides the basic data for subsequent calculations and planning. Only by accurately understanding the size, shape of the loading space, and the size and placement attitude requirements of the target object can the arrangement method of the target object in the loading space be reasonably planned.

[0035] Loading space information: Information used to describe a three-dimensional space composed of one or more hexahedrons, including parameters such as the size, shape, and position of the space. These parameters are crucial for determining the placement method of the target object in the loading space.

[0036] Target object information: Information used to describe a cube with a length value, a width value, and a height value, mainly referring to the size of the target object, which is the basic data for calculating how the target object is arranged in the loading space.

[0037] Placement attitude information of the target object: Information such as the direction and angle when the target object is placed in the loading space. Different placement attitudes will affect the occupied space and arrangement method of the target object in the loading space.

[0038] Accurately obtaining the relevant information provides the necessary data support for subsequent steps, ensuring that the planning of the entire vehicle loading and palletizing scheme is based on the actual loading space and the target object situation, and avoiding unreasonable schemes caused by inaccurate information.

[0039] Among them, the loading space information includes: three-dimensional space information of at least one hexahedron combination, and the three-dimensional space information includes a length value, a width value, a height value, and a spatial shape structure parameter value;

[0040] The target object information is the cube bounding box information of the target object constructed according to the oriented bounding box OBB;

[0041] The placement attitude information includes the first attitude information with the length and width faces of the target object facing down, the second attitude information with the width and height faces of the target object facing down, and the third attitude information with the length and height faces of the target object facing down.

[0042] The loading space information describes a three-dimensional space composed of at least one hexahedron combination. The three-dimensional space information here covers multiple key parameters. The length value, width value, and height value define the size of the space in three dimensions. For example, for a loading space in the shape of a cuboid, the specific values of its length, width, and height determine the volume of the space. The spatial shape structure parameter value further describes the specific shape characteristics of the space, which may include whether it is a regular cuboid, whether there are special depressions or protrusions, etc. If it is a space composed of multiple hexahedrons, information such as the relative positions and connection relationships between these hexahedrons also needs to be clarified.

[0043] Optionally, deduct the free space dimensions that must be executed during the movement of the robot gripper and the loading process in the loading space, and deduct the interference space dimensions that may appear such as the container crossbeam and truss.

[0044] Since the robot needs sufficient space for the gripper to grasp, transport, and place the target object during the palletizing operation. If this part of the space is not reserved, the robot gripper may collide with the stacked goods or the boundary of the loading space, resulting in operation failure and even possible damage to the robot or goods.

[0045] During the loading process, in addition to the movement space of the robot gripper, the space required for factors such as the swing and rotation of the goods during transportation also needs to be considered. In addition, to ensure the smooth progress of the loading process, a certain amount of space also needs to be reserved for the path planning and adjustment of the robot to avoid interference with other objects.

[0046] Structures such as container crossbeams and trusses are fixed in the loading space and they will occupy a certain space. If these structures are not considered when planning the loading scheme, then during the actual loading process, the goods or the robot may interfere with these structures and the normal palletizing and loading operations cannot be completed.

[0047] Therefore, after deducting these spatial dimensions, it can be ensured that the robot will not collide with surrounding objects during operation, protecting the safety of the robot and the goods, and reducing the risk of equipment damage and goods loss caused by collisions. Reserving free space and deducting interference space dimensions provide a reasonable spatial range for the movement of the robot and the loading of goods, enabling the robot to operate according to the preset path and posture, improving the accuracy and efficiency of the loading process, and avoiding operation interruptions or delays caused by insufficient space. Although certain spatial dimensions are deducted, overall, this approach can more reasonably plan the loading space, avoiding problems such as impossible loading or difficulties in subsequent operations caused by blindly pursuing space utilization rate. By reasonably deducting these spaces, on the premise of ensuring operation feasibility, the remaining space can be maximally utilized for goods stacking, achieving the optimization of space utilization.

[0048] The loading space information is crucial for subsequent target arrangement and robot trajectory planning. Accurate loading space information is the basis for determining whether and how the targets can be reasonably placed. Only by understanding the size and shape of the loading space can the optimal target arrangement plan be calculated through algorithms to make full use of the space and avoid situations of space waste or impossible placement of targets. At the same time, it also provides environmental information for the robot to plan its movement trajectory, ensuring that the robot will not collide with the boundaries of the loading space or other obstacles during operation.

[0049] The target information is to construct the cubic bounding box information of the target according to the oriented bounding box (OBB). OBB is a method for describing the boundary of an object. It can enclose the object more tightly and can more accurately reflect the shape of the object compared to the traditional axis-aligned bounding box (AABB). The cubic bounding box information constructed by OBB can accurately represent the size and direction of the target. For example, for an object with an irregular shape, the OBB algorithm can be used to find a cubic bounding box that best wraps the object and determine its direction and size parameters in space, such as length, width, and height.

[0050] During the target arrangement and stacking process, accurate target information is the key to calculating space utilization rate and generating the arrangement plan. It can help the algorithm more accurately calculate the occupied space of the target in the loading space, thereby finding the optimal arrangement method. At the same time, for the operations of the robot to grasp and place the target, the accurate information of the target also helps the robot determine the appropriate grasping position and posture, ensuring the accuracy and stability of the operation.

[0051] The placement attitude information describes different placement ways of the target object in the loading space. The first attitude information is that the length-width surface of the target object faces downwards, that is, the surface where the length and width of the target object are located is parallel to the bottom surface of the loading space; the second attitude information is that the width-height surface faces downwards, that is, the surface where the width and height of the target object are located is parallel to the bottom surface; the third attitude information is that the length-height surface faces downwards, that is, the surface where the length and height of the target object are located is parallel to the bottom surface. These different placement postures will result in different occupied spaces and arrangement ways of the target object in the loading space.

[0052] The placement attitude information provides more arrangement possibilities and increases the optimization space. When generating the target object arrangement plan, considering different placement postures can find a better space utilization rate and palletizing stability. For example, in some cases, placing the target object with the length-width surface facing downwards may be more conducive to filling the space, while in other cases, the width-height surface facing downwards or the length-height surface facing downwards may be more appropriate. At the same time, for the operation of the robot, different placement postures also require the robot to adjust the corresponding operation postures to perform grasping and placement. Therefore, the placement attitude information is also an important basis for the robot operation planning.

[0053] Involve step 120: Generate the target object arrangement plan based on the knapsack algorithm and the multi-objective optimization algorithm. The knapsack algorithm provides a basic idea for placing the target object into the loading space. By continuously trying different combination methods, a plan that can maximize the utilization of the loading space is found. The multi-objective optimization algorithm, on this basis, considers the balance between multiple optimization objectives. For example, while pursuing the space utilization rate, the palletizing stability also needs to be considered to avoid the pallet collapsing due to unreasonable placement of the target object. By weighting these objectives or adopting other optimization strategies, an optimal target object arrangement plan with the best comprehensive performance is found.

[0054] Knapsack algorithm: An algorithm for combinatorial optimization of NP-complete problems. In this scenario, it can be analogized to putting target objects (items) of different sizes into the loading space (knapsack) to achieve a certain optimal goal, such as maximizing the utilization of the loading space.

[0055] Multi-objective optimization algorithm: An algorithm used to optimize multiple conflicting objectives simultaneously. When generating the target object arrangement plan, multiple objectives may need to be considered simultaneously, such as maximizing the space utilization rate, the highest palletizing stability, and the shortest palletizing time. The multi-objective optimization algorithm is used to balance these objectives and find an optimal compromise plan.

[0056] Optionally, with maximizing the space utilization rate and minimizing the cycle time as the optimization objectives, and using the Pareto optimal method for decision-making.

[0057] During the process of loading goods, making full use of the loading space is a very important goal. Improving space utilization can reduce the number of transportation containers used, lower transportation costs, and at the same time improve logistics efficiency, enabling more full use of limited transportation resources.

[0058] The takt time refers to the cycle time to complete one palletizing operation, that is, the palletizing time. Minimizing the takt time means being able to complete the palletizing and loading of goods faster, improving work efficiency, reducing the idle time of manpower and equipment, and thus enhancing the operation efficiency of the entire logistics system to meet the business requirements of rapid turnover.

[0059] According to the experience of manual nesting, placing the layer with a larger height in the vertical direction at the bottom helps to improve the overall stability and can better utilize the vertical space. The bottom bears a larger weight. Placing the layer with a larger height can make the center of gravity lower, reducing the risk of pallet collapse. At the same time, it also provides a more stable foundation for placing the upper-layer goods, which is conducive to further improving space utilization.

[0060] To improve space utilization, more complex palletizing schemes may be required, which will increase the complexity of operations and result in an extended takt time; while simply pursuing the minimization of the takt time may sacrifice some space utilization. The Pareto optimal method can find a balance between these two conflicting goals, helping to determine the set of solutions where one goal cannot be improved without making the other goal worse, so as to select a better solution that can meet certain requirements for both goals.

[0061] By simultaneously considering the maximum utilization of nesting space and the minimization of takt time, it is possible to achieve a dual optimization of space and time in the logistics link. It not only makes full use of the space of transportation containers but also improves the efficiency of palletizing and loading, helping to reduce logistics costs and enhance the benefits of the entire logistics system.

[0062] The Pareto optimal method provides a scientific basis for decision-making, avoiding the one-sidedness of decision-making caused by simply relying on experience or only focusing on a single goal. It can systematically analyze the performance of different solutions in terms of the two goals, providing a set of representative better solutions for decision-makers, enabling decision-makers to make more practical choices according to the actual situation.

[0063] This optimization method can enable enterprises to more reasonably allocate logistics resources, including manpower, equipment, and transportation containers, etc. It reduces the waste and idleness of resources, improves the utilization efficiency of resources, and helps enterprises enhance their competitiveness in the fierce market competition. Placing the layer with a larger height at the bottom not only improves space utilization but also enhances the stability of the pallet. This helps to reduce the possibility of goods collapse or damage caused by shaking, jolting, etc. during transportation, improves the reliability and stability of the logistics system, and reduces potential losses and risks.

[0064] Therefore, by combining the knapsack algorithm and the multi-objective optimization algorithm, an optimized arrangement plan for the objects can be generated. This plan can not only make full use of the loading space and improve the space utilization rate, but also achieve a good balance among multiple optimization objectives, making the palletizing both stable and efficient, thus improving the quality and efficiency of the entire loading and palletizing process.

[0065] Involve step 130: Plan the motion trajectory and operation posture of the robot from the loading point to the placement point according to the object arrangement plan. After determining the arrangement mode of the objects, it is necessary to enable the robot to accurately grasp the objects from the specified loading point and place them at the corresponding placement point. This requires planning a reasonable motion trajectory based on the position and posture of the objects, as well as the motion ability and constraint conditions of the robot, and at the same time determining the operation posture of the robot during the process of grasping, transporting, and placing the objects, so as to ensure that the robot can successfully complete the palletizing task and will not collide with the surrounding environment during the motion process.

[0066] The motion trajectory and operation posture of the robot planned according to the object arrangement plan can enable the robot to accurately and efficiently complete the tasks of transporting and palletizing the objects from the loading point to the placement point. A reasonable motion trajectory can reduce the motion time and energy consumption of the robot, and at the same time avoid safety problems such as collisions; a suitable operation posture can ensure that the robot stably grasps and places the objects, improving the accuracy and stability of the palletizing, and ultimately realizing efficient, safe, and accurate loading and palletizing operations.

[0067] Such as Figure 2 As shown, in a possible embodiment, step 120 may specifically include the following steps:

[0068] Step 210, determine the first priority weight value corresponding to the palletizing time and the second priority weight value corresponding to the space utilization rate;

[0069] Step 220, based on the knapsack algorithm and the multi-objective optimization algorithm, generate an object arrangement plan according to the first priority weight value, the second priority weight value, the loading space information, the object information, and the placement posture information of the objects.

[0070] In the actual loading and palletizing scenario, the palletizing time and the space utilization rate are two important considerations, but their importance may vary depending on the specific requirements. By determining the priority weight values of these two factors, their relative importance in the multi-objective optimization can be quantified.

[0071] For example, if the current task has a very urgent time requirement, a relatively high weight value can be assigned to the palletizing time; if the space cost is high and it is desired to improve the space utilization rate as much as possible, a relatively high weight value can be assigned to the space utilization rate. These weight values will play a key role in the subsequent algorithm calculations and affect the generation of the target object arrangement plan.

[0072] The knapsack algorithm provides a basic framework for placing target objects into the loading space, and it tries different combination methods to find the possibilities of space utilization. The multi-objective optimization algorithm then comprehensively considers different objectives by using the priority weight values of the palletizing time and space utilization rate determined previously. During the calculation process, the algorithm will evaluate indicators such as the palletizing time and space utilization rate under various target object arrangement methods, and perform weighted calculations on these indicators according to the weight values to obtain a comprehensive evaluation score.

[0073] By continuously trying and comparing different arrangement methods, the method with the optimal comprehensive evaluation score is selected as the target object arrangement plan. For example, for a certain arrangement method, calculate the score corresponding to its palletizing time and the score corresponding to the space utilization rate, and then perform weighted summation according to the corresponding weight values to obtain the comprehensive score of this arrangement method. Finally, select the arrangement method with the highest comprehensive score from all possible arrangement methods.

[0074] Defining the priority weight values of the palletizing time and space utilization rate enables the system to flexibly adjust the degree of emphasis on these two key factors according to actual needs. This helps to formulate a target object arrangement plan that is more in line with the actual situation in different application scenarios. For example, during the peak logistics period, assigning a higher weight to the palletizing time can speed up the loading speed and improve the logistics efficiency; while in the case of limited space and high costs, increasing the weight of the space utilization rate can reduce the transportation cost and improve the economic benefits.

[0075] The target object arrangement plan generated by combining the knapsack algorithm and the multi-objective optimization algorithm and considering information such as weight values is an optimization plan that comprehensively considers multiple objectives. This plan can find a better balance between the palletizing time and space utilization rate. It can not only shorten the time required for palletizing to a certain extent and improve the palletizing efficiency, but also make full use of the loading space and reduce space waste. Thus, it realizes the efficient and economic operation of the loading and palletizing process and meets the requirements of actual business.

[0076] In a possible embodiment, as Figure 3 shown, in step 220, it may specifically include the following steps:

[0077] Step 310, in the one-dimensional strip planning, according to the loading space information, the target object information, and the placement attitude information of the target object, calculate the space utilization rate of the target object arrangement, and determine the one-dimensional strip arrangement plan according to the space utilization rate of the target object arrangement;

[0078] Step 320, in the two-dimensional layer planning, use the knapsack algorithm to update the one-dimensional strip arrangement plan and determine the arrangement plan of the target object on the two-dimensional plane;

[0079] Step 330, in the three-dimensional space planning, according to the first priority weight value and the second priority weight value, perform multi-objective optimization on the arrangement plan on the two-dimensional plane to generate the target object arrangement plan.

[0080] In the one-dimensional strip planning, consider the loading space as a one-dimensional strip. Based on the loading space information, the target object information, and the placement attitude information of the target object, try different arrangements of the target objects. For each arrangement, calculate the ratio of the space occupied by the arrangement on the one-dimensional strip to the total space of the one-dimensional strip. This ratio is the space utilization rate. This process is to preliminarily screen out the arrangements with higher space utilization efficiency in the one-dimensional direction.

[0081] Based on the calculated space utilization rate, select the arrangement with the highest space utilization rate or the arrangement that meets specific requirements as the one-dimensional strip arrangement plan. This plan is the basis for subsequent two-dimensional and three-dimensional planning.

[0082] In the two-dimensional layer planning, regard the one-dimensional strip arrangement plan as an item and the two-dimensional plane as a knapsack. The knapsack algorithm will try different combinations of one-dimensional strips to achieve the optimal utilization of the space on the two-dimensional plane. By continuously adjusting the arrangement and combination of the one-dimensional strips, update the one-dimensional strip arrangement plan, and thus determine the arrangement plan of the target object on the two-dimensional plane. This process is to further optimize the layout of the target object on the two-dimensional plane based on the one-dimensional planning.

[0083] In the three-dimensional space planning, it is necessary to comprehensively consider two objectives: the palletizing time and the space utilization rate. According to the first priority weight value and the second priority weight value determined previously, evaluate and adjust the arrangement plan on the two-dimensional plane. Through the multi-objective optimization algorithm, find a compromise plan among different arrangement plans to make the comprehensive effect of the palletizing time and the space utilization rate optimal, and finally generate the target object arrangement plan.

[0084] Specifically, in the one-dimensional strip planning, starting from the lower left corner point of the container, perform an exhaustive planning for three different postures of the palletizing target. Exhaustive planning means trying all possible permutations and combinations to find the optimal arrangement of the palletizing target on the one-dimensional strip, so as to achieve the highest space utilization rate in this one-dimensional direction.

[0085] In the two-dimensional layer planning, based on the one-dimensional strip planning, the dynamic programming knapsack algorithm is used to plan the one-dimensional strips. The core idea of the knapsack algorithm is to select items in a knapsack with a given capacity so that the total value of the items in the knapsack is maximized. Here, the two-dimensional layer is equivalent to the knapsack, and the one-dimensional strip is equivalent to the item. Through the dynamic programming algorithm, the optimal arrangement combination of the one-dimensional strips on the two-dimensional plane can be found, further improving the space utilization rate.

[0086] In the three-dimensional space planning, multi-objective optimization and dynamic programming algorithms are used to plan the layering. In the three-dimensional space, multiple objectives need to be considered simultaneously, such as maximizing the space utilization rate and the highest palletizing stability, etc. Through the multi-objective optimization algorithm, an optimal solution that balances multiple objectives can be found among different layering schemes; dynamic programming is used to solve sub-problems and gradually construct the optimal palletizing scheme for the entire three-dimensional space.

[0087] Through step-by-step planning and optimization, the three-dimensional space of the container can be fully utilized, space waste can be reduced, and the loading capacity of the goods can be increased. The application of multi-objective optimization enables the consideration of palletizing stability while pursuing the space utilization rate. By reasonably arranging the arrangement and layering of palletizing objectives, the risk of pallet collapse can be reduced. The use of the dynamic programming algorithm avoids repeated calculations, improves the computational efficiency of the algorithm, and enables a relatively optimal palletizing layout scheme to be obtained in a short time.

[0088] Thus, by calculating the space utilization rate and determining the one-dimensional strip arrangement scheme, a relatively efficient arrangement method in the one-dimensional direction can be initially screened out, providing a relatively reasonable basis for subsequent two-dimensional and three-dimensional planning, reducing the search space for subsequent planning, and improving the overall planning efficiency.

[0089] Updating the one-dimensional strip arrangement scheme using the knapsack algorithm can further optimize the arrangement of the target objects on the two-dimensional plane, improve the space utilization rate of the two-dimensional plane, and make the layout of the target objects on the two-dimensional level more reasonable.

[0090] Performing multi-objective optimization on the arrangement scheme on the two-dimensional plane according to the weight value can comprehensively consider the two important objectives of palletizing time and space utilization rate in the three-dimensional space, generate an arrangement scheme of the target objects that can meet the time requirements and fully utilize the space, and realize the efficient and economical operation of the loading and palletizing process.

[0091] In a possible embodiment, in step 120, it may specifically include the following steps:

[0092] Establish a hierarchical constraint mechanism, which is used to constrain the first positional relationship between the center of gravity position of the upper-layer objects and the edge position of the lower-layer objects in the object arrangement scheme, and the second positional relationship between the projected edge position of the upper-layer objects on the horizontal plane and the edge position of the lower-layer objects on the horizontal plane;

[0093] Generate an object arrangement scheme based on the hierarchical constraint mechanism, loading space information, object information, and the placement attitude information of the objects.

[0094] During the palletizing process, there is an important relationship between the center of gravity position of the upper-layer objects and the edge position of the lower-layer objects. If the center of gravity of the upper-layer objects exceeds the edge of the lower-layer objects too much, the stability of the palletizing structure will be threatened and it is prone to tipping over. By constraining the first positional relationship, it is ensured that the center of gravity of the upper-layer objects falls within the support range of the lower-layer objects as much as possible, thereby enhancing the overall stability of the palletizing. For example, when placing a box on another box, it should be ensured that the center of gravity of the upper box is within the area enclosed by the edges of the lower box.

[0095] The relationship between the projected edge position of the upper-layer objects on the horizontal plane and the edge position of the lower-layer objects on the horizontal plane also affects the stability and space utilization rate of the palletizing. By constraining the second positional relationship, it is possible to prevent the upper-layer objects from exceeding the lower-layer objects excessively in the horizontal direction, prevent the occurrence of unstable structures, and at the same time contribute to more reasonable use of space. For example, the degree of overlap between the projected edge of the upper-layer objects and the edge of the lower-layer objects can be used as a constraint condition to ensure that the objects are arranged tightly and stably.

[0096] On the basis of the established hierarchical constraint mechanism, combine the loading space information, object information, and the placement attitude information of the objects to generate an object arrangement scheme. At this time, when the algorithm tries different object arrangement methods, it will judge whether the arrangement meets the requirements of stability and space utilization according to the hierarchical constraint mechanism. For example, for a specific loading space and a group of objects, the algorithm will try different placement postures and arrangement orders, and at the same time check whether the two positional relationships in the hierarchical constraint mechanism are satisfied. If satisfied, it will continue to evaluate other indicators. If not satisfied, it will abandon the arrangement method, and finally screen out the optimal object arrangement scheme.

[0097] First, clearly define the allowable range of the center of gravity position of the upper-layer objects and the edge position of the lower-layer objects, that is, determine the specific constraint conditions of the first positional relationship. For example, a distance threshold between the center of gravity of the upper-layer objects and the edge of the lower-layer objects can be set.

[0098] Then define the constraint conditions for the projected edge position of the upper-layer objects on the horizontal plane and the edge position of the lower-layer objects on the horizontal plane. For example, stipulate the overlap ratio of the projected edge, etc., so as to establish a complete hierarchical constraint mechanism.

[0099] Divide the loading space into different areas, and determine the size and shape of the target object according to the target object information. Try different placement postures of the target object in sequence. For each placement posture, try to arrange the target objects in the loading space in a certain order, and check the rationality of the arrangement according to the hierarchical constraint mechanism. Calculate indicators such as the space utilization rate of the arrangement method that conforms to the hierarchical constraint mechanism. Compare the indicators of different arrangement methods, and select the optimal arrangement method as the target object arrangement plan.

[0100] Establishing a hierarchical constraint mechanism can improve the stability of palletizing, reduce the risk of pallet collapse caused by unreasonable arrangement of target objects, and ensure the safety of palletizing during transportation or storage. It provides clear stability constraint conditions for subsequent generation of the target object arrangement plan, making the generated plan more practically feasible.

[0101] According to the hierarchical constraint mechanism, loading space information, target object information, and target object placement posture information, the generated target object arrangement plan not only meets the stability requirements, but also optimizes the space utilization rate to a certain extent, achieving a balance between stability and space utilization. It enables the robot to work according to a more reasonable plan during palletizing operations, improving the efficiency and quality of palletizing, and reducing repeated operations or adjustments caused by unreasonable plans.

[0102] In a possible embodiment, in step 130, it may specifically include the following steps:

[0103] Through the heuristic breadth-first search algorithm, generate candidate directions for the robot's movement path according to the target object arrangement plan;

[0104] Through Dijkstra's algorithm, iteratively update the robot's movement path length and operation time according to the target object arrangement plan to generate the robot's motion trajectory;

[0105] According to the candidate directions of the robot's movement path and the robot's motion path, plan the robot's motion trajectory and operation posture from the loading point to the placement point to generate a loading and palletizing plan; the loading and palletizing plan includes the coordinate information, placement posture information of the target object in the loading space, as well as the movement path and operation sequence information of the robot from the loading point to the placement point.

[0106] Heuristic breadth-first search algorithm: This is a graph search algorithm. During the search process, it traverses the nodes in the graph in a breadth-first manner. The heuristic function is used to evaluate the "goodness" of each node, guiding the search towards a direction more likely to find the optimal solution. In this scenario, it can help quickly find the possible directions of the robot's movement path.

[0107] Dijkstra's algorithm: A typical single-source shortest path algorithm used to find the shortest paths from a starting node to all other nodes in a weighted graph. By continuously iterating and updating the path lengths, the optimal paths are gradually determined. Here, it is used to calculate the optimal length of the robot's movement path and the corresponding operation time.

[0108] Convert the information such as the loading space and the positions of the target objects described in the target object arrangement plan into a graph structure, where the nodes can represent the position points in space and the edges represent the connection relationships between the nodes. The heuristic breadth-first search algorithm starts from the starting position of the robot and expands the search nodes layer by layer in a breadth-first manner. During the expansion process, a heuristic function is used to evaluate the value of each node, and nodes with greater potential are preferentially selected for expansion, thereby quickly generating a set of candidate directions for the robot's movement path. These candidate directions are the set of directions in which the robot may move, providing a basis for determining the specific path subsequently.

[0109] Based on the graph structure and weights determined by the target object arrangement plan, Dijkstra's algorithm starts from the starting node, i.e., the loading point, and gradually calculates the shortest paths to all other nodes. During the iteration process, the shortest path length from each node to the starting node and the corresponding operation time are continuously updated. In this way, the shortest path from the starting node to the target node, i.e., the placement point, which is the movement trajectory of the robot, is finally determined. This process takes into account the layout of the entire loading space and the movement cost, and can find the optimal movement path.

[0110] Combining the candidate directions of the movement path generated previously and the finally determined movement path, and comprehensively considering the mechanical structure and operation requirements of the robot, plan the operation postures of the robot at each position. At the same time, integrate the coordinate information of the target objects in the loading space, the placement posture information, and the movement path and operation sequence information of the robot from the loading point to the placement point to form a complete loading and palletizing plan. This plan details the key information such as the actions of the robot and the position postures of the target objects during the entire palletizing process.

[0111] By quickly generating candidate directions of the movement path through the heuristic breadth-first search algorithm, the search range can be narrowed, unnecessary search paths can be reduced, the efficiency of path planning can be improved, and time can be saved for subsequently determining the precise movement trajectory.

[0112] The movement trajectory determined by Dijkstra's algorithm is the optimal path considering the entire loading space and the movement cost, which can make the movement path of the robot the shortest and the operation time the most reasonable when completing the palletizing task, thereby improving the palletizing efficiency and reducing energy consumption.

[0113] The palletizing and loading plan formulated by integrating the candidate directions and the motion path describes in detail and accurately the motion and operation of the robot during palletizing and the relevant information of the target objects, provides clear guidance for the actual palletizing and loading operation, ensures the accuracy and efficiency of the palletizing process, and is also convenient for monitoring and managing the palletizing process.

[0114] In a possible embodiment, in step 130, it may specifically include the following steps:

[0115] Through the three-dimensional palletizing layout algorithm, generate the palletizing sequence and operation postures according to the arrangement plan of the target objects;

[0116] Plan the first motion trajectory of the robot from the feeding point to the picking transition point;

[0117] According to the operation postures of the robot, use the three-dimensional A* algorithm to calculate the second motion trajectory from the picking transition point to the placement point;

[0118] Perform motion simulation on the first motion trajectory and the second motion trajectory until the preset safety conditions are met to obtain the palletizing and loading plan.

[0119] Three-dimensional palletizing layout algorithm: It is an algorithm used to determine the arrangement method, palletizing sequence and operation postures of target objects in three-dimensional space. It comprehensively considers factors such as the size, shape, weight of the target objects and the limitations of the palletizing space to achieve an efficient and stable palletizing layout.

[0120] Three-dimensional A* algorithm: The A* algorithm is a classic algorithm for finding the shortest path in a graph or network, and the three-dimensional A* algorithm extends its application to three-dimensional space. In robot trajectory planning, it can search for an optimal path from the starting point to the ending point according to the given starting point, ending point and environmental information, while considering constraints such as the operation postures of the robot.

[0121] Motion simulation is to simulate and analyze the motion process of the robot through computer simulation. In this process, the motion trajectory, operation postures of the robot and its interaction with the surrounding environment can be observed to evaluate the feasibility and safety of the motion.

[0122] Preset safety conditions are a series of pre-set standards or rules used to judge whether the motion trajectory of the robot is safe and feasible. These conditions may include avoiding collisions with obstacles, maintaining the stability of the robot, meeting the requirements of operation postures, etc.

[0123] The three-dimensional palletizing layout algorithm is based on the arrangement plan of the target objects and further analyzes the sequence of each target object during the palletizing process. This requires considering the spatial relationship between the target objects, the stability requirements, and the overall structure of the palletizing. At the same time, according to the shape, size of the target objects and the requirements of palletizing, determine the appropriate operating postures when the robot grasps and places each target object to ensure the smooth progress of the operation. For example, for target objects with irregular shapes, specific postures may be required to accurately grasp and place them.

[0124] Based on the position information of the loading point and the picking transition point, combined with the motion capabilities and constraints of the robot, such as joint movement range, maximum speed, etc., plan a motion trajectory from the starting position, i.e., the loading point, to the intermediate transition position, i.e., the picking transition point. This may involve path planning algorithms, such as geometric planning, sampling-based planning, etc., to find a safe and efficient path.

[0125] On the premise of knowing the robot's operating posture, taking the picking transition point as the starting point and the placement point as the ending point, use the three-dimensional A* algorithm to search for the optimal path in the three-dimensional space. The algorithm will consider the motion constraints of the robot, obstacle information, and the impact of the operating posture on the path, continuously evaluate and select the optimal nodes for expansion until the best motion trajectory from the picking transition point to the placement point is found.

[0126] Input the two planned motion trajectories into the motion simulation system to simulate the actual motion process of the robot. During the simulation, check whether the robot will collide with surrounding objects and whether the robot's motion meets the requirements of the operating posture and stability. If the preset safety conditions are not met, adjust and optimize the motion trajectory and perform the simulation again until all safety conditions are met. The finally obtained motion trajectory and related information that meet the safety conditions constitute the loading and palletizing plan.

[0127] The robot trajectory planning executes operations according to the palletizing sequence generated by the three-dimensional palletizing layout algorithm. Starting from the fixed loading mechanism, grab the goods in the predetermined order and reach the picking transition point through the picking trajectory. This is done to ensure the consistency between the robot's operations and the palletizing plan and make the palletizing process proceed in an orderly manner. For example, if the palletizing plan stipulates to pick the goods at a certain position first, the robot will grab the goods from the fixed loading mechanism in this order and move to the picking transition point to prepare for subsequent placement.

[0128] According to the placing posture, preferably use the three-dimensional A-Star algorithm to calculate the placing trajectory. The three-dimensional A-Star algorithm is an algorithm for finding the optimal path in the three-dimensional space. It selects the path by evaluating the cost function F(n) = g(n) + h(n) of each state.

[0129] Among them, F(n) is the cost estimate from the initial state through state n to the target state; g(n) is the actual cost from the initial state to the current state n; h(n) is the cost estimate of the optimal path from the current state n to the target state. When placing goods, this algorithm can find the optimal path from the material-taking transition point to the placement point according to the requirements of the feeding posture, enabling the robot to accurately place the goods at the predetermined position.

[0130] After completing the trajectory planning, motion simulation interference judgment is carried out. This is to check whether the robot will collide with surrounding objects when executing the planned trajectory. Through motion simulation, potential interference problems can be discovered before actual operation, avoiding the situation of colliding and damaging the robot or goods during actual operation.

[0131] The core of the A-Star algorithm lies in its combination of the actual cost g(n) from the starting point to the current point and the estimated cost h(n) from the current point to the target point. During the search process, the algorithm will preferentially expand the node with the smallest F(n) value and gradually construct the path from the starting point to the target point. This way enables the algorithm to find the optimal solution while being more efficient in finding the path compared to some blind search algorithms (such as breadth-first search), because it uses the heuristic function h(n) to guide the search direction and reduces unnecessary search nodes.

[0132] In the three-dimensional case, the three-dimensional A-Star algorithm is restricted by the spatial segmentation resolution. Spatial segmentation divides the three-dimensional space into small units (grids) so that the algorithm can search on these units. If the resolution is too low, it means that the size of each unit is relatively large, and then it is not precise enough to represent complex spatial structures and obstacles. At this time, the algorithm needs to consider more possible paths because there may be various different situations within each large unit, which will geometrically increase the algorithm complexity. For example, within a relatively large unit, there may be multiple obstacles, but due to low resolution, they cannot be accurately distinguished, and the algorithm needs to calculate various possible bypass methods, resulting in a significant increase in the amount of calculation.

[0133] When an interference occurs, it indicates that the path previously planned by the three-dimensional A-Star algorithm is not feasible. Since the three-dimensional A-Star algorithm may be less efficient or unable to find a suitable path when dealing with complex interference situations, at this time, the non-convex high-dimensional space random sampling search algorithm is used to re-plan the path. The non-convex high-dimensional space random sampling search algorithm finds a feasible path by randomly sampling points in the space and attempting to connect these points. It does not rely on a deterministic search method like the A-Star algorithm and has better adaptability to complex non-convex spaces and situations with interference.

[0134] This algorithm can quickly try different path combinations when interference occurs and find a feasible path to avoid interference. Although it samples randomly, through reasonable sampling strategies and path evaluation methods, it is still possible to find a path close to the optimal one, enabling the robot to successfully complete trajectory planning in a complex spatial environment and ensuring the continuation of the palletizing operation.

[0135] Generating the palletizing sequence and operation postures can ensure the orderly progress of the palletizing process, improve the efficiency and quality of palletizing, and at the same time ensure that the robot can accurately grasp and place the target object, avoiding damage or accidents caused by improper operations. Planning the first motion trajectory enables the robot to move safely and efficiently from the initial position to the pick-up transition point, preparing for subsequent pick-up and placement operations and reducing unnecessary motion time and energy consumption.

[0136] Calculating the second motion trajectory can find the optimal path from the pick-up transition point to the placement point, ensuring that the robot can avoid obstacles when placing the target object and at the same time meet the requirements of the operation postures, improving the accuracy and stability of placement. Through repeated simulation and adjustment, it is ensured that the motion trajectory of the robot meets the preset safety conditions, avoiding collisions or other safety problems in actual operations, thereby improving the safety and reliability of the loading and palletizing process and finally obtaining a feasible loading and palletizing plan.

[0137] In the embodiments of the present application, by obtaining the loading space information, target object information, and the placement posture information of the target object, the loading space information is used to describe the information of a three-dimensional space composed of the combination of one or more hexahedrons, and the target object information is used to describe the information of a cube with length value, width value, and height value, providing the necessary data support for the subsequent steps and ensuring that the planning of the entire loading and palletizing plan is based on the actual loading space and target object conditions. Based on the knapsack algorithm and multi-objective optimization algorithm, according to the loading space information, target object information, and the placement posture information of the target object, multiple optimization objectives can be balanced to find the optimal target object arrangement plan, which can not only make full use of the loading space, improve the space utilization rate, achieve a good balance among multiple optimization objectives, make the palletizing both stable and efficient, thereby improving the quality and efficiency of the entire loading and palletizing process. According to the target object arrangement plan, plan the motion trajectory and operation postures of the robot from the loading point to the placement point to obtain the loading and palletizing plan. A reasonable motion trajectory can reduce the motion time and energy consumption of the robot and at the same time avoid safety problems such as collisions, and a suitable operation posture can ensure that the robot stably grasps and places the target object, improving the accuracy and stability of palletizing. Thus, intelligent loading operations can be realized.

[0138] The robot loading and palletizing method provided by the embodiment of the present application may have a robot loading and palletizing device as the execution subject. In the embodiment of the present application, taking the robot loading and palletizing device executing the robot loading and palletizing method as an example, the robot loading and palletizing device provided by the embodiment of the present application is described.

[0139] Figure 4 It is a block diagram of a robot loading and palletizing device provided by an embodiment of the present application. The device 400 includes:

[0140] An acquisition module 410, configured to acquire loading space information, target object information, and the placement attitude information of the target object. The loading space information is used to describe the information of a three-dimensional space composed of the combination of one or more hexahedrons, and the target object information is used to describe the information of a cube with a length value, a width value, and a height value;

[0141] A generation module 420, configured to generate a target object arrangement plan based on the knapsack algorithm and the multi-objective optimization algorithm according to the loading space information, the target object information, and the placement attitude information of the target object;

[0142] A planning module 430, configured to plan the movement trajectory and operation attitude of the robot from the loading point to the placement point according to the target object arrangement plan to obtain a loading and palletizing plan.

[0143] In a possible embodiment, the generation module 420 is specifically configured to:

[0144] Determine a first priority weight value corresponding to the palletizing time and a second priority weight value corresponding to the space utilization rate;

[0145] Based on the knapsack algorithm and the multi-objective optimization algorithm, generate a target object arrangement plan according to the first priority weight value, the second priority weight value, the loading space information, the target object information, and the placement attitude information of the target object.

[0146] In a possible embodiment, the generation module 420 is specifically configured to:

[0147] In the one-dimensional strip planning, calculate the space utilization rate of the target object arrangement according to the loading space information, the target object information, and the placement attitude information of the target object;

[0148] Determine a one-dimensional strip arrangement plan according to the space utilization rate of the target object arrangement;

[0149] In the two-dimensional layer planning, use the knapsack algorithm to update the one-dimensional strip arrangement plan to determine the arrangement plan of the target object on the two-dimensional plane;

[0150] In the three-dimensional space planning, perform multi-objective optimization on the arrangement plan on the two-dimensional plane according to the first priority weight value and the second priority weight value to generate a target object arrangement plan.

[0151] In a possible embodiment, the planning module 430 is specifically configured to:

[0152] Generate candidate directions for the movement path of the robot according to the target arrangement scheme through a heuristic breadth-first search algorithm;

[0153] Iteratively update the movement path length and operation time of the robot according to the target arrangement scheme through Dijkstra's algorithm to generate the motion trajectory of the robot;

[0154] Plan the motion trajectory and operation posture of the robot from the loading point to the placement point according to the candidate directions of the robot's movement path and the robot's motion path, and generate a loading and palletizing scheme; the loading and palletizing scheme includes the coordinate information of the target in the loading space, the placement posture information, and the movement path and operation sequence information of the robot from the loading point to the placement point.

[0155] In a possible embodiment, the loading space information includes: the three-dimensional space information of at least one hexahedron combination, and the three-dimensional space information includes a length value, a width value, a height value, and a spatial shape structure parameter value;

[0156] The target information is the cube bounding box information of the target constructed according to the oriented bounding box OBB;

[0157] The placement posture information includes the first posture information with the length and width surfaces of the target facing down, the second posture information with the width and height surfaces of the target facing down, and the third posture information with the length and height surfaces of the target facing down.

[0158] In a possible embodiment, the generation module 420 is specifically configured to:

[0159] Establish a hierarchical constraint mechanism, which is used to constrain the first position relationship between the center of gravity position of the upper-layer target and the edge position of the lower-layer target in the target arrangement scheme, and the second position relationship between the projected edge position of the upper-layer target on the horizontal plane and the edge position of the lower-layer target on the horizontal plane;

[0160] Generate a target arrangement scheme according to the hierarchical constraint mechanism, the loading space information, the target information, and the placement posture information of the target.

[0161] In a possible embodiment, the planning module 430 is specifically configured to:

[0162] Generate a palletizing sequence and operation posture according to the arrangement scheme of the target through a three-dimensional palletizing layout algorithm;

[0163] Plan the first motion trajectory of the robot from the loading point to the pick-up transition point;

[0164] According to the operating posture of the robot, the three-dimensional A* algorithm is used to calculate the second motion trajectory from the picking transition point to the placing point;

[0165] Perform motion simulation on the first motion trajectory and the second motion trajectory until the preset safety conditions are met, and obtain the loading and palletizing plan.

[0166] In the embodiments of the present application, by obtaining the loading space information, the target object information, and the placing posture information of the target object, the loading space information is used to describe the information of a three-dimensional space composed of a combination of one or more hexahedrons, and the target object information is used to describe the information of a cube with a length value, a width value, and a height value, which provides necessary data support for the subsequent steps and ensures that the planning of the entire loading and palletizing plan is based on the actual loading space and the target object situation. Based on the knapsack algorithm and the multi-objective optimization algorithm, according to the loading space information, the target object information, and the placing posture information of the target object, multiple optimization objectives can be balanced, and the optimal target object arrangement plan can be found, which can not only make full use of the loading space, improve the space utilization rate, achieve a good balance among multiple optimization objectives, make the palletizing stable and efficient, thereby improving the quality and efficiency of the entire loading and palletizing process. According to the target object arrangement plan, plan the motion trajectory and operating posture of the robot from the feeding point to the placing point, and obtain the loading and palletizing plan. A reasonable motion trajectory can reduce the motion time and energy consumption of the robot, and at the same time avoid safety problems such as collisions. A suitable operating posture can ensure that the robot stably grasps and places the target object, improving the accuracy and stability of palletizing. Thus, intelligent loading operations can be realized.

[0167] Figure 5 The figure shows a schematic hardware structure diagram of an electronic device provided by an embodiment of the present application.

[0168] The electronic device may include a processor 501 and a memory 502 storing computer program instructions.

[0169] Specifically, the above-mentioned processor 501 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured as one or more integrated circuits implementing the embodiments of the present application.

[0170] The memory 502 may include a mass memory for data or instructions. By way of example and not limitation, the memory 502 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 502 may include removable or non-removable (or fixed) media. Where appropriate, the memory 502 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, the memory 502 is a non-volatile solid-state memory. In a particular embodiment, the memory 502 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or a flash memory, or a combination of two or more of these.

[0171] The processor 501 reads and executes the computer program instructions stored in the memory 502 to implement any of the information processing methods in the illustrated embodiments.

[0172] In one example, the electronic device may further include a communication interface 503 and a bus 510. Among them, as Figure 5 shown, the processor 501, the memory 502, and the communication interface 503 are connected via the bus 510 to complete communication with each other.

[0173] The communication interface 503 is mainly used to implement communication between the various modules, devices, units, and / or devices in the embodiments of the present application.

[0174] The bus 510 includes hardware, software, or both, and couples the components of the electronic device to each other. By way of example and not limitation, the bus may include an accelerated graphics port (AGP) or other graphics bus, an enhanced industry standard architecture (EISA) bus, a front-side bus (FSB), a hyperTransport (HT) interconnect, an industry standard architecture (ISA) bus, an InfiniBand interconnect, a low-pin count (LPC) bus, a memory bus, a microChannel architecture (MCA) bus, a peripheral component interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a serial advanced technology attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, the bus 510 may include one or more buses. Although the embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0175] The electronic device can execute the information processing method in the embodiments of the present application, so as to implement the combination of Figure 2 the described information processing method.

[0176] In addition, in combination with the information processing method in the above embodiments, the embodiments of the present application can provide a computer-readable storage medium to implement. Computer program instructions are stored on the computer-readable storage medium; when the computer program instructions are executed by a processor, the Figure 1 information processing method is implemented.

[0177] It should be clear that the present application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, the detailed description of known methods is omitted here in the above embodiments. A number of specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present application.

[0178] The functional blocks shown in the above structural block diagrams can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and so on. When implemented in software, the elements of the present application are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium or transmitted via a data signal carried in a carrier wave on a transmission medium or a communication link. "Machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, and so on. The code segment can be downloaded via a computer network such as the Internet, an intranet, and so on.

[0179] It should also be noted that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be executed in the order mentioned in the embodiments, or different from the order in the embodiments, or several steps can be executed simultaneously.

[0180] As described above, this is only the specific implementation manner of the present application. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein. It should be understood that the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present application.

Claims

1. A method for a robot to load and stack goods, characterized in that, The method includes: Obtaining loading space information, target object information, and the placement attitude information of the target object. The loading space information is used to describe the information of a three-dimensional space composed of a combination of one or more hexahedrons, and the target object information is used to describe the information of a cube with a length value, a width value, and a height value; Based on the knapsack algorithm and the multi-objective optimization algorithm, generating a target object arrangement plan according to the loading space information, the target object information, and the placement attitude information of the target object; According to the target object arrangement plan, planning the movement trajectory and operation attitude of the robot from the loading point to the placement point to obtain a loading and stacking plan.

2. The method according to claim 1, wherein The generating of the target object arrangement plan based on the knapsack algorithm and the multi-objective optimization algorithm according to the loading space information, the target object information, and the placement attitude information of the target object includes: Determining a first priority weight value corresponding to the stacking time and a second priority weight value corresponding to the space utilization rate; Based on the knapsack algorithm and the multi-objective optimization algorithm, generating a target object arrangement plan according to the first priority weight value, the second priority weight value, the loading space information, the target object information, and the placement attitude information of the target object.

3. The method according to claim 2, wherein The generating of the target object arrangement plan according to the first priority weight value, the second priority weight value, the loading space information, the target object information, and the placement attitude information of the target object includes: In the one-dimensional strip planning, calculating the space utilization rate of the target object arrangement according to the loading space information, the target object information, and the placement attitude information of the target object; Determining a one-dimensional strip arrangement plan according to the space utilization rate of the target object arrangement; In the two-dimensional layer planning, using the knapsack algorithm to update the one-dimensional strip arrangement plan to determine the arrangement plan of the target object on the two-dimensional plane; In the three-dimensional space planning, performing multi-objective optimization on the arrangement plan on the two-dimensional plane according to the first priority weight value and the second priority weight value to generate a target object arrangement plan.

4. The method according to claim 1, characterized in that, The planning of the movement trajectory and operation attitude of the robot from the loading point to the placement point according to the target object arrangement plan to obtain a loading and stacking plan includes: Generating candidate directions for the movement path of the robot according to the target object arrangement plan through the heuristic breadth-first search algorithm; Iteratively updating the movement path length and operation time of the robot according to the target object arrangement plan through the Dijkstra algorithm to generate the movement trajectory of the robot; According to the candidate directions of the movement path of the robot and the movement path of the robot, planning the movement trajectory and operation attitude of the robot from the loading point to the placement point to generate a loading and stacking plan; the loading and stacking plan includes the coordinate information of the target object in the loading space, the placement attitude information, and the movement path and operation sequence information of the robot from the loading point to the placement point.

5. The method according to claim 1, wherein The loading space information includes: the three-dimensional space information of at least one hexahedron combination, and the three-dimensional space information includes a length value, a width value, a height value, and a spatial shape structure parameter value; The target object information is the cube bounding box information of the target object constructed according to the oriented bounding box OBB; The placement attitude information includes a first attitude information with the length and width surface of the target object facing down, a second attitude information with the width and height surface of the target object facing down, and a third attitude information with the length and height surface of the target object facing down.

6. The method according to claim 1, characterized in that, Generating an object arrangement plan according to the loading space information, object information, and the placement attitude information of the object, includes: Establishing a hierarchical constraint mechanism, which is used to constrain the first positional relationship between the center of gravity position of the upper-layer object and the edge position of the lower-layer object in the object arrangement plan, and the second positional relationship between the projected edge position of the upper-layer object on the horizontal plane and the edge position of the lower-layer object on the horizontal plane; Generating an object arrangement plan according to the hierarchical constraint mechanism, the loading space information, the object information, and the placement attitude information of the object.

7. The method according to claim 1, wherein Planning the movement trajectory and operation attitude of the robot from the loading point to the placement point according to the object arrangement plan to obtain a loading and palletizing plan, includes: Generating a palletizing sequence and operation attitude according to the object arrangement plan through a three-dimensional palletizing layout algorithm; Planning the first movement trajectory of the robot from the loading point to the picking transition point; Calculating the second movement trajectory from the picking transition point to the placement point according to the operation attitude of the robot by using a three-dimensional A* algorithm; Performing motion simulation on the first movement trajectory and the second movement trajectory until a preset safety condition is met to obtain a loading and palletizing plan.

8. A robot loading and palletizing device, characterized in that, The device includes: An acquisition module, which is used to acquire the loading space information, the object information, and the placement attitude information of the object. The loading space information is used to describe the information of a three-dimensional space composed of the combination of one or more hexahedrons, and the object information is used to describe the information of a cube with a length value, a width value, and a height value; A generation module, which is used to generate an object arrangement plan based on a knapsack algorithm and a multi-objective optimization algorithm according to the loading space information, the object information, and the placement attitude information of the object; A planning module, which is used to plan the movement trajectory and operation attitude of the robot from the loading point to the placement point according to the object arrangement plan to obtain a loading and palletizing plan.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus; The memory is used to store a computer program; When the processor is used to execute the program stored on the memory, it implements the method described in any one of claims 1-7.

10. A computer-readable storage medium, on which instructions are stored. When executed by one or more processors, the instructions cause the processor to execute the method described in any one of claims 1-7.

Images