AMR path planning method, device and storage medium based on digital twin

Abstract

The present invention relates to an AMR path planning method, device and storage medium based on digital twins, which are applied to the technical field of AMR path planning, and include: obtaining all possible path trajectories through a path search algorithm, performing cube rasterization processing on an AMR operating environment of a preset size, and obtaining the path length of each path trajectory passing through each cube grid; constructing a digital twin model of the AMR operating environment, simulating various dangerous events on the path length through the digital twin model, and obtaining dangerous event impact parameters; using the obtained path length parameters as scheme layer factors, and using a hierarchical analysis method to solve a road communication state index of a target layer; obtaining a path equivalent length according to the road traffic state index; obtaining an obstacle avoidance cost for the AMR to avoid obstacles on the path trajectory; obtaining the path cost of the path trajectory according to all path equivalent lengths and obstacle avoidance costs in the path trajectory, and selecting a path trajectory with the smallest path cost as the optimal path.

Description

Technical Field

[0001] The present invention relates to the technical field of AMR path planning, and in particular to an AMR path planning method, device and storage medium based on digital twins. Background Art

[0002] An autonomous mobile robot (AMR) is an industrial device that can move automatically along a preset route. It combines computer hardware technology, parallel and distributed processing technology, automatic control technology, and sensor technology to complete handling tasks accurately and efficiently. It is usually used to transport goods in factories or warehouses. The operating environment of AMR is usually a complex dynamic system. How to accurately plan the path and avoid collisions, conflicts, and blockages is an urgent problem to be solved.

[0003] Existing AMR path planning methods include: AMR guidance technology and traditional search algorithms;

[0004] Among them, AMR guidance technology includes two categories: fixed channel guidance and free path guidance;

[0005] Fixed channel guidance: The representative technologies are electromagnetic navigation and ribbon navigation, which use buried magnetic strips and ribbons to make the AMR drive along a fixed road;

[0006] Free path guidance: Representative technologies include laser navigation and visual navigation. Laser navigation is to install reflectors in the AMR working environment and determine the current position of the AMR by the reflection of lasers. Visual guidance is to install visual sensors on the AMR and use image processing technology to enable the AMR to "sense" its own position and surrounding environment. For example, application number 201320260187.X discloses an AMR visual navigation system that uses cameras and LED lights for image acquisition, and the motion control module controls the movement of the AMR.

[0007] However, the fixed channel guidance technology requires the laying of magnetic strips and ribbons in advance, which has certain requirements for the flatness of the road surface and the layout of the equipment. Once the path is set, it is not easy to change, and the flexibility is poor. Laser navigation requires the installation of reflectors in the working environment, which has certain requirements for the building structure. The image acquisition equipment can only collect the environment within a viewing angle, and cannot obtain a global view of the surrounding environment. In addition, image acquisition has high requirements for the building structure and light conditions of the AMR working environment.

[0008] Traditional search methods, such as the A* algorithm, Dijkstra algorithm and other grid methods, are characterized by rasterizing the map, traversing the grid according to certain rules, and gradually searching for the optimal path; however, existing research on the algorithm is mostly conducted in a two-dimensional plane, treating the AMR as a volumeless particle and obstacles as undifferentiated objects. In particular, obstacles are also processed in a two-dimensional grid, ignoring the spatial structure of the obstacles, such as the space under the hanging beam mechanism, which can be passed by the AMR in actual production, but the two-dimensional grid map will regard it as an obstacle and cannot pass; at the same time, each road section is regarded as an undifferentiated grid, and the impact of different road conditions on the AMR's passage is not reflected, such as bends, slopes, narrow roads, etc. In addition, existing research lacks consideration of the dynamic constraints of different models of AMRs and ignores the posture requirements of the transportation operation, resulting in the path search results being unsuitable for actual material transportation tasks. Summary of the invention

[0009] In view of this, the purpose of the present invention is to provide an AMR path planning method, device and storage medium based on digital twins to solve the problem in the prior art that the fixed channel guidance technology requires the advance laying of magnetic strips and color ribbons, has certain requirements on the road surface flatness and equipment layout, and once the path is set, it is not easy to change and has poor flexibility; while laser navigation requires the installation of reflectors in the working environment, has certain requirements on the building structure, and has requirements on light; and traditional search methods are mostly carried out in a two-dimensional plane, ignoring the spatial structure of obstacles, treating each road section as an undifferentiated grid, and not reflecting the impact of different road conditions on the AMR traffic situation.

[0010] According to a first aspect of an embodiment of the present invention, a method for AMR path planning based on digital twin is provided, the method comprising:

[0011] Get the starting point and end point of the AMR, and use the path search algorithm to obtain all the path trajectories from the starting point to the end point;

[0012] The AMR operating environment is processed into a cube grid of a preset size, and any path track passes through multiple cube grids, and the intersection point of any path track and each cube grid is obtained, and the path length of the path track in a cube grid is obtained according to the coordinates of the intersection point;

[0013] Acquire path parameters of the path length, and acquire flow rate, road width and slope, and surrounding environment parameters of the path length according to the path parameters;

[0014] Constructing a digital twin model of the AMR operating environment, simulating various dangerous events on the path length through the digital twin model, obtaining severity parameters of each dangerous event, and obtaining dangerous event impact parameters according to the severity parameters of each dangerous event;

[0015] The flow rate of the path length, the road width and slope, the surrounding environment parameters and the impact parameters of dangerous events are used as the scheme layer factors, and the hierarchical analysis method is used to solve the road communication status index of the target layer;

[0016] Obtaining a path equivalent length of the path length according to a road traffic status index of the path length; obtaining the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory;

[0017] Obtaining an obstacle avoidance cost of the AMR in avoiding obstacles on a path trajectory;

[0018] Obtaining a path cost of the path trajectory according to the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory and the obstacle avoidance cost;

[0019] Obtain the path costs corresponding to all path trajectories, and select the path trajectory with the smallest path cost as the optimal path.

[0020] Preferably,

[0021] The path parameters of the path length are obtained, and the flow rate, road width and slope, and surrounding environment parameters of the path length are obtained according to the path parameters, including:

[0022] Obtain the number of vehicles and people passing through the path length in a preset time period, and obtain the flow rate of the path length according to the number of vehicles and people and the preset time period;

[0023] Obtaining the width, horizontal length and plumb height difference of the path length, and obtaining the road width and slope according to the width, horizontal length and plumb height difference of the path length;

[0024] Obstacles within a certain area around the path length are obtained, and the surrounding environment parameters are obtained according to the ratio of the area occupied by the obstacles.

[0025] Preferably,

[0026] The digital twin model of the AMR operating environment is constructed as follows:

[0027] Obtain the external dimensions and internal structure data of AMR and other equipment in the operating environment;

[0028] Importing the external dimension data and internal structure data of the AMR and other equipment into 3ds Max software for model building;

[0029] Rendering and generating a virtual environment in Unity3D, and placing the models of the AMR and other devices at corresponding positions of the generated virtual environment;

[0030] The rules for setting the models of the AMR and other devices in the Unity3D include:

[0031] Physical models are used to analyze the mechanical characteristics of AMR and other equipment;

[0032] The behavior model is used to define the specific motion behaviors of AMR and other devices;

[0033] The rule model is used to define the paths, scopes, and constraints of AMRs and other devices;

[0034] All real-time data generated by the physical entity movement of AMR and other equipment are collected and transmitted to the virtual system. The virtual system drives the virtual entity to perform corresponding actions based on the input real-time data, completes the real-time mapping from physical to virtual, and obtains the digital twin model of the AMR operating environment.

[0035] Preferably,

[0036] The digital twin model is used to simulate various dangerous events on the path length to obtain severity parameters of each dangerous event, including:

[0037] Configure different environmental variables in the digital twin model, obtain historical dangerous event case data, input the historical dangerous event case data into the digital twin model, and obtain the impact size, impact range, and impact time of different dangerous event case data;

[0038] Obtain severity parameters of different hazardous events based on impact size, impact scope, and impact time.

[0039] Preferably,

[0040] The method of using the flow rate of the path length, the road width and slope, the surrounding environment parameters and the risk event impact parameters as the scheme layer factors and using the hierarchical analysis method to solve the road communication status index of the target layer includes:

[0041] The analytic hierarchy process is used to divide the solution of the road traffic status into a target layer, a criterion layer and a solution layer. The target layer is the road traffic status, the criterion layer is the road conditions and the road surrounding environment, the solution layer factors corresponding to the road conditions are the flow of the path length and the road width and slope, and the solution layer factors corresponding to the road surrounding environment are the surrounding environment parameters of the path length and the risk event impact parameters.

[0042] The relative importance of each factor in the criterion layer and the scheme layer to the road traffic state is set, and a road traffic state judgment matrix, a road condition judgment matrix, and a road surrounding environment judgment matrix are constructed;

[0043] Obtain the influence weights of the scheme layer factors on the target layer according to the road traffic state judgment matrix, the road condition judgment matrix and the road surrounding environment judgment matrix;

[0044] The road traffic status index of the path length is obtained according to the influence weights of the scheme layer factors on the target layer.

[0045] Preferably,

[0046] The step of obtaining the road traffic status index of the path length according to the influence weights of the scheme layer factors on the target layer includes:

[0047] The scheme layer factors are dimensionally processed respectively, and the products of all dimensionally processed scheme layer factors and the corresponding scheme layer factors' influence weights on the target layer are accumulated to obtain the road traffic status index of the path length.

[0048] Preferably,

[0049] The obtaining of the obstacle avoidance cost of the AMR avoiding obstacles on the path trajectory includes:

[0050] A bounding box with preset length, width and height is used to replace the outline of the AMR, the current distance between any obstacle in the path trajectory and the bounding box is obtained, and the response distance, buffer distance and danger distance are set;

[0051] When the current distance between the obstacle in the path trajectory and the bounding box is between the response distance and the buffer distance, the obstacle avoidance cost is a preset avoidance fixed value;

[0052] When the current distance between the obstacle and the bounding box in the path trajectory is between the buffer distance and the danger distance, the obstacle avoidance cost is obtained according to the preset avoidance fixed value and the difference between the current distance and the danger distance.

[0053] According to a second aspect of an embodiment of the present invention, a digital twin-based AMR path planning device is provided, the device comprising:

[0054] Path search module: used to obtain the starting point and end point of the AMR, and use the path search algorithm to obtain all path trajectories from the starting point to the end point;

[0055] Path length acquisition module: used to process the AMR operating environment into a cube grid of a preset size, then any path track passes through multiple cube grids, obtain the intersection of any path track and each cube grid, and obtain the path length of the path track in a cube grid according to the intersection coordinates;

[0056] Path parameter acquisition module: used to acquire the path parameters of the path length, and acquire the flow rate, road width and slope, and surrounding environment parameters of the path length according to the path parameters;

[0057] Dangerous event impact acquisition module: used to build a digital twin model of the AMR operating environment, simulate various dangerous events on the path length through the digital twin model, obtain the severity parameter of each dangerous event, and obtain the dangerous event impact parameter according to the severity parameter of each dangerous event;

[0058] Hierarchical analysis module: used to use the flow of the path length, road width and slope, surrounding environmental parameters and dangerous event impact parameters as scheme layer factors, and use hierarchical analysis method to solve the road communication status index of the target layer;

[0059] An equivalent length acquisition module: used to acquire the path equivalent length of the path length according to the road traffic status index of the path length; and to acquire the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory;

[0060] Obstacle avoidance cost acquisition module: used to obtain the obstacle avoidance cost of the AMR avoiding obstacles on the path trajectory;

[0061] A path cost acquisition module: used to acquire the path cost of the path trajectory according to the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory and the obstacle avoidance cost;

[0062] Optimal path confirmation module: used to obtain the path costs corresponding to all path trajectories and select the path trajectory with the smallest path cost as the optimal path.

[0063] According to a third aspect of an embodiment of the present invention, there is provided a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a main controller, each step in the above method is implemented.

[0064] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0065] The present application obtains all possible path trajectories through a path search algorithm, performs cube rasterization processing on the AMR operating environment of a preset size, and obtains the path length of each path trajectory passing through each cube grid; by constructing a digital twin model of the AMR operating environment, the digital twin model is used to simulate various dangerous events on the path length, obtain the severity parameters of each dangerous event, and obtain the dangerous event impact parameters according to the severity parameters; the obtained path length parameters are used as solution layer factors, and the hierarchical analysis method is used to solve the road communication state index of the target layer; the path equivalent length of the path length is obtained according to the road traffic state index; the obstacle avoidance cost of the AMR avoiding obstacles on the path trajectory is obtained; the path cost of the path trajectory is obtained according to the path equivalent length corresponding to all cube grids passed by the AMR in the path trajectory and the obstacle avoidance cost, and the path trajectory with the smallest path cost is selected as the optimal path; the method of the present application is carried out in three-dimensional space, and multiple constraints such as motion, path, and obstacle avoidance are added. The obtained path is accurate and meets the actual production needs, and there are no requirements for the environment and other special equipment.

[0066] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0068] Figure 1 is a flowchart of an AMR path planning method based on digital twins according to an exemplary embodiment;

[0069] Figure 2 is a schematic diagram of an architecture for solving a road traffic state coefficient using a hierarchy analysis method according to another exemplary embodiment;

[0070] Figure 3 is a schematic diagram showing that the outline of an AMR is replaced by a bounding box according to another exemplary embodiment;

[0071] Figure 4 is a schematic diagram showing a path trajectory passing through a cubic grid according to another exemplary embodiment;

[0072] Figure 5 is a schematic diagram of a digital twin model construction process according to another exemplary embodiment;

[0073] Figure 6 is a schematic diagram of a 3D cube grid according to another exemplary embodiment;

[0074] Figure 7is a system schematic diagram of an AMR path planning device based on digital twin according to another exemplary embodiment;

[0075] In the attached figure: 1-path search module, 2-path length acquisition module, 3-path parameter acquisition module, 4-danger event impact acquisition module, 5-hierarchical analysis module, 6-equivalent length acquisition module, 7-obstacle avoidance cost acquisition module, 8-path cost acquisition module, 9-optimal path confirmation module. DETAILED DESCRIPTION

[0076] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0077] Embodiment 1

[0078] Figure 1 is a flow chart of a digital twin-based AMR path planning method according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0079] S1, obtain the starting point and end point of the AMR, and use the path search algorithm to obtain all the path trajectories from the starting point to the end point;

[0080] S2, the AMR operating environment is processed into a cube grid of a preset size, and any path track passes through multiple cube grids, and the intersection point of any path track and each cube grid is obtained, and the path length of the path track in a cube grid is obtained according to the coordinates of the intersection point;

[0081] S3, obtaining path parameters of the path length, and obtaining flow, road width and slope, and surrounding environment parameters of the path length according to the path parameters;

[0082] S4, constructing a digital twin model of the AMR operating environment, simulating various dangerous events on the path length through the digital twin model, obtaining severity parameters of each dangerous event, and obtaining dangerous event impact parameters according to the severity parameters of each dangerous event;

[0083] S5, using the flow rate of the path length, the road width and slope, the surrounding environment parameters and the impact parameters of dangerous events as scheme layer factors, and using the hierarchical analysis method to solve the road communication status index of the target layer;

[0084] S6, obtaining a path equivalent length of the path length according to the road traffic status index of the path length; obtaining the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory;

[0085] S7, obtaining an obstacle avoidance cost of the AMR for avoiding obstacles on the path trajectory;

[0086] S8, obtaining a path cost of the path trajectory according to the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory and the obstacle avoidance cost;

[0087] S9, obtaining the path costs corresponding to all path trajectories, and selecting a path trajectory with the smallest path cost as the optimal path;

[0088] It can be understood that the starting point and the end point of the AMR are obtained, and all possible path trajectories are obtained through the path search algorithm;

[0089] Next, the road status information is obtained through cameras and other facilities, and the path map is processed into a cube grid. A suitable unit length a is selected to divide the space occupied by the scene into a 3D grid map composed of cubes with a length, width and height of a. Different from traditional 2D maps, 3D maps can more accurately reflect the structure of irregular obstacles, such as the attached Figure 6 As shown in the figure, the number of path re-planning is reduced, the efficiency and accuracy of the search are improved, and then the state information of the road surface in different areas is obtained according to the model material of the road surface and the photos taken by the camera in the scene. According to the road surface state information extraction results in each grid area, a road traffic state coefficient is assigned to each grid;

[0090] The road traffic status in each grid is considered the same, as shown in the attached Figure 4 As shown in the figure, the intersection of the path trajectory and the cube grid is obtained. It is approximately assumed that the AMR moves in a straight line in a grid. The curved trajectory is decomposed into broken line segments, and the intersection coordinates A (x1, y1, z1) and B (x2, y2, z2) are obtained. The length of the path segment is calculated using the distance formula between two points. d :

[0091]

[0092] For any length of any path trajectory, as shown in the attached Figure 2As shown in the figure, the analytic hierarchy process is used to solve the road traffic state coefficient. The target layer road traffic state is affected by two factors: the criterion layer road section conditions (S1) and the road section surrounding environment (S2). The road conditions are affected by two subdivided factors: the solution layer flow (E1) and road width (E2) and slope. The road surrounding environment is affected by two factors: the solution layer surrounding environment parameters (E3) and the dangerous event impact parameters (E4).

[0093] Flow rate (E1), calculated as:

[0094]

[0095] In the formula, Q For other vehicles and people passing through, t For statistical time;

[0096] Road width and slope (E2). Under normal conditions, road passability is proportional to road width and inversely proportional to road slope. The calculation formula is:

[0097]

[0098] In the formula, w is the road width, x is the horizontal length of the road section, y is the plumb height difference of the road;

[0099] Surrounding environment parameter (E3), the area ratio of obstacles in a certain area around the road section, calculated by the formula:

[0100]

[0101] In the formula, is the area of ​​obstacles in a certain area around it, S is the area of ​​a certain area around the road section;

[0102] Dangerous event parameter (E4), around the AMR driving path, there are often other operations and unexpected situations, collectively referred to as dangerous events. The calculation formula for its impact on the AMR path cost is:

[0103]

[0104] In the formula, i Number the dangerous event. T is the severity of the dangerous event, which can be obtained by simulating dangerous events with the digital twin model. d is the distance from the event to the road section, where the construction of the digital twin model is as shown in the attached figure. Figure 5 As shown, including:

[0105] Use 3D modeling software to build geometric models of AMR, road surface, equipment, etc. at a 1:1 ratio, and give them appearances such as materials, colors, skins, and textures. Import the models into Unity and render them using Unity's built-in renderer;

[0106] 3dsMax is also the mainstream 3D modeling software. It focuses more on the modeling of the appearance rather than the construction of the complex internal structure. However, the rendering capability of this software is very complete and it is widely used in the field of visual monitoring. Since the equipment models of the AMR working environment are complex and numerous, the entire scene model is divided into three parts: dynamic model, static model and UI interaction model. The establishment of the 3D model is mainly for the dynamic and static models. 3dsMax is used to model the AMR and other equipment. The construction of the motor, roller, bearing and other parts is completed first, and then the whole machine is assembled.

[0107] Unity3D is a mainstream game engine in the world. With the increasing demand for equipment monitoring, more and more researchers choose Unity3D as a visual monitoring software. According to the appearance data observed in the early stage, the equipment is arranged in the virtual environment in Unity3D. The entire virtual scene is composed of AMR, other equipment, roads, buildings and other parts. The model is arranged according to its actual spatial layout and coordination relationship. The Light effect in Unity3D is added to simulate the lighting effect in the actual scene. The component function of the software is used to set auxiliary equipment, floors and other objects to complete the construction of the entire equipment scene and ensure the high-fidelity restoration of the equipment. At the same time, to prevent the subsequent system from running In order to prevent the occurrence of movements that do not conform to the actual situation, it is necessary to establish hierarchical dependencies for all components in the virtual scene, connect all subcomponents according to the subordinate relationship, create a virtual entity that is highly consistent with the physical entity, and set the motion constraints of AMR and other devices, including: the physical model is used to analyze the mechanical change characteristics of AMR and other devices; the behavior model is used to clarify the specific motion behavior of AMR and other devices; the rule model is used to clarify the path, range and constraint conditions of AMR and other devices; all real-time data generated by the movement of physical entities are collected and transmitted to the virtual system. After the system processes the data, it uses these real-time data to drive the virtual entity to perform corresponding actions, and complete the real-time mapping from physical to virtual. The dynamic behavior of virtual entities is mainly translation, rotation and scaling. By applying the above three transformations, the dynamic behavior of virtual entities can be realized. The Transform property in Unity is used to complete a series of translation, rotation and scaling operations on objects. By combining basic movements, the simulation of all running actions of AMR is realized, the construction of the behavior model is completed, and the real-time dynamic mapping from physical entities to virtual entities is realized.

[0108] The digital twin model has good real-time interactive characteristics. It can freely configure environmental variables and reflect them in real time in the simulation of the digital twin system. Therefore, by configuring different environmental variables - such as equipment production status, number of AMRs, fault frequency, etc., the digital twin model can simulate dangerous event situations by learning from historical dangerous event cases and combining historical data with environmental variables. The severity of dangerous events is mainly measured by three factors: impact size, impact range, and impact time. The impact size includes the reduction in output, the scrapping or maintenance costs of equipment, and the economic losses caused by casualties. The impact range is the number of equipment affected by the dangerous event, and the impact time is the length of time from the occurrence of the accident to complete recovery. The impact size, impact range, and impact time data are obtained based on the simulation results. These data can be used to more accurately calculate the severity of the dangerous event. T .

[0109] Next, determine the influence weight of each factor on the road traffic status index. The calculation steps are as follows:

[0110] First, construct a judgment matrix, and the assignment method is shown in Table 1 below:

[0111] Table 1

[0112]

[0113] Cij represents the value of the influencing factor of factor i on factor j. By comparing the relative importance of each factor in the established evaluation index system, the judgment matrix RS (as shown in Table 2), S1-E (as shown in Table 3), and S2-E (as shown in Table 4) are constructed for the road traffic status (R), the road conditions (S1), and the surrounding environment of the road section (S2) through empirical judgment:

[0114] Table 2

[0115]

[0116] Table 3

[0117]

[0118] Table 4

[0119]

[0120] After completing the construction of the judgment matrix, it is necessary to further calculate the weight vector of each factor. The calculation formula is as follows:

[0121]

[0122] In the formula, n is the number of factors at the scenario level,i , j is the factor number, a ij Indication factors i Equivalent to factor j The importance rating of k The judgment matrix k This formula means that after normalizing each column, i Average all elements in the row; first determine the influence weight of the target layer on the criterion layer through the judgment matrix RS and , as shown in Table 5 below, and then use the judgment matrix S1-E and S2-E to determine the influence weight of the criterion layer on the solution layer , , as well as As shown in Tables 6 and 7 below, the weight of the influence of the scheme layer on the criterion layer is multiplied by the weight of the influence of the criterion layer on the target layer to obtain the weight of the influence of the scheme layer E1-E4 on the road traffic state R , , ;

[0123] The calculation results are as follows:

[0124] Table 5

[0125]

[0126] Table 6

[0127]

[0128] Table 7

[0129]

[0130] The influence weights of the obtained flow (E1), road width and slope (E2), surrounding environment parameters (E3) and dangerous event impact parameters (E4) on the target layer are shown in Table 8:

[0131] Table 8

[0132]

[0133] The data of the four factors in the scheme layer corresponding to a certain road section are dimensionlessly processed, and the processed data are multiplied by the corresponding influence weight of each factor. Finally, the results of the multiplication of the data and the weight are added together to obtain the road traffic status index of this road section. φ , the calculation formula is as follows:

[0134]

[0135] In the formula, m It represents the dimensionless value of flow (E1), road width and slope (E2), surrounding environment parameters (E3) and dangerous event impact parameters (E4) after dimensionless processing;

[0136] The path equivalent length D is used to describe the actual length after weighted calculation of the road traffic status index. d The "equivalent length" of the path is calculated as:

[0137]

[0138] AMR will interact with various obstacles during operation. To reduce potential safety hazards, a certain safety distance must be kept, such as the following: Figure 3 As shown in the figure, the safety distance threshold s is the minimum distance limit between the AMR edge and the obstacle during the path search process. To simplify the calculation, a bounding box is used instead of the AMR outline. The AMR is circumscribed by a bounding box with a length of r, a width of d, and a height of h. As long as the bounding box does not intersect with other obstacle outlines, it is considered that no collision has occurred. In order to reduce the number of path replanning and simulate a more realistic collision process, the response distance is set. S 1. Buffer distance S 2. Danger distance S 3. The response distance represents the distance at which the AMR detects an obstacle and takes it into consideration for obstacle avoidance. A small, fixed obstacle avoidance cost is set to encourage the AMR to follow the original path. The buffer distance represents the distance at which the AMR makes a decision and executes the obstacle avoidance strategy. At this time, the AMR is encouraged to execute the obstacle avoidance strategy. An obstacle avoidance cost that gradually increases as the distance decreases is set. The danger distance represents the distance at which the AMR has a potential collision risk with the obstacle. At this time, the original path should be abandoned immediately to execute the obstacle avoidance strategy. The current distance between the AMR and the obstacle is I , C is the obstacle avoidance cost, and the calculation formula of the obstacle avoidance cost is as follows:

[0139]

[0140] Always, C 1 is the preset basic obstacle avoidance cost, which is a fixed value. k is a preset constant;

[0141] The path cost P can be obtained by accumulating the path equivalent length D corresponding to all n grids passed by the AMR in the path trajectory and then adding the obstacle avoidance cost C. The calculation formula is:

[0142]

[0143] After obtaining the path costs P of all path trajectories, a path trajectory with the smallest path cost P is selected as the optimal path output.

[0144] Embodiment 2

[0145] Figure 7 is a system schematic diagram of an AMR path planning device based on digital twin according to another exemplary embodiment, the device comprising:

[0146] Path search module 1: used to obtain the starting point and end point of the AMR, and use the path search algorithm to obtain all path trajectories from the starting point to the end point;

[0147] Path length acquisition module 2: used to process the AMR operating environment into a cube grid of a preset size, then any path track passes through multiple cube grids, obtain the intersection point of any path track and each cube grid, and obtain the path length of the path track in a cube grid according to the intersection coordinates;

[0148] Path parameter acquisition module 3: used to acquire the path parameters of the path length, and acquire the flow rate, road width and slope, and surrounding environment parameters of the path length according to the path parameters;

[0149] Dangerous event impact acquisition module 4: used to build a digital twin model of the AMR operating environment, simulate various dangerous events on the path length through the digital twin model, obtain the severity parameter of each dangerous event, and obtain the dangerous event impact parameter according to the severity parameter of each dangerous event;

[0150] Hierarchical analysis module 5: used to use the flow of the path length, road width and slope, surrounding environment parameters and dangerous event impact parameters as scheme layer factors, and use hierarchical analysis method to solve the road communication status index of the target layer;

[0151] Equivalent length acquisition module 6: used to acquire the path equivalent length of the path length according to the road traffic status index of the path length; acquire the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory;

[0152] Obstacle avoidance cost acquisition module 7: used to obtain the obstacle avoidance cost of the AMR avoiding obstacles on the path trajectory;

[0153] Path cost acquisition module 8: used to acquire the path cost of the path trajectory according to the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory and the obstacle avoidance cost;

[0154] The optimal path confirmation module 9 is used to obtain the path costs corresponding to all path trajectories, and select a path trajectory with the smallest path cost as the optimal path.

[0155] Embodiment three:

[0156] This embodiment provides a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a main controller, each step in the above method is implemented;

[0157] It is understandable that the storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0158] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0159] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0160] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0161] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0162] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0163] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0164] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0165] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0166] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. AMR path planning method based on digital twin, characterized in that: The method comprises: Get the starting point and end point of the AMR, and use the path search algorithm to obtain all the path trajectories from the starting point to the end point; The AMR operating environment is processed into a cube grid of a preset size, and any path track passes through multiple cube grids, and the intersection point of any path track and each cube grid is obtained, and the path length of the path track in a cube grid is obtained according to the coordinates of the intersection point; Acquire path parameters of the path length, and acquire flow rate, road width and slope, and surrounding environment parameters of the path length according to the path parameters; Constructing a digital twin model of the AMR operating environment, simulating various dangerous events on the path length through the digital twin model, obtaining severity parameters of each dangerous event, and obtaining dangerous event impact parameters according to the severity parameters of each dangerous event; The flow rate of the path length, the road width and slope, the surrounding environment parameters and the impact parameters of dangerous events are used as the scheme layer factors, and the hierarchical analysis method is used to solve the road communication status index of the target layer; Obtaining a path equivalent length of the path length according to a road traffic status index of the path length; obtaining the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory; Obtaining an obstacle avoidance cost of the AMR in avoiding obstacles on a path trajectory; Obtaining a path cost of the path trajectory according to the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory and the obstacle avoidance cost; Obtain the path costs corresponding to all path trajectories, and select the path trajectory with the smallest path cost as the optimal path.

2. The method according to claim 1, characterized in that The path parameters of the path length are obtained, and the flow rate, road width and slope, and surrounding environment parameters of the path length are obtained according to the path parameters, including: Obtain the number of vehicles and people passing through the path length in a preset time period, and obtain the flow rate of the path length according to the number of vehicles and people and the preset time period; Obtaining the width, horizontal length and plumb height difference of the path length, and obtaining the road width and slope according to the width, horizontal length and plumb height difference of the path length; Obstacles within a certain area around the path length are obtained, and the surrounding environment parameters are obtained according to the ratio of the area occupied by the obstacles.

3. The method according to claim 2, characterized in that The digital twin model of the AMR operating environment is constructed as follows: Obtain the external dimensions and internal structure data of AMR and other equipment in the operating environment; Importing the external dimension data and internal structure data of the AMR and other equipment into 3ds Max software for model building; Rendering and generating a virtual environment in Unity3D, and placing the models of the AMR and other devices at corresponding positions of the generated virtual environment; The rules for setting the models of the AMR and other devices in the Unity3D include: Physical models are used to analyze the mechanical characteristics of AMR and other equipment; The behavior model is used to define the specific motion behaviors of AMR and other devices; The rule model is used to define the paths, scopes, and constraints of AMRs and other devices; All real-time data generated by the physical entity movement of AMR and other equipment are collected and transmitted to the virtual system. The virtual system drives the virtual entity to perform corresponding actions based on the input real-time data, completes the real-time mapping from physical to virtual, and obtains the digital twin model of the AMR operating environment.

4. The method according to claim 3, characterized in that The digital twin model is used to simulate various dangerous events on the path length to obtain severity parameters of each dangerous event, including: Configure different environmental variables in the digital twin model, obtain historical dangerous event case data, input the historical dangerous event case data into the digital twin model, and obtain the impact size, impact range, and impact time of different dangerous event case data; Obtain severity parameters of different hazardous events based on impact size, impact scope, and impact time.

5. The method according to claim 4, characterized in that The method of using the flow rate of the path length, the road width and slope, the surrounding environment parameters and the risk event impact parameters as the scheme layer factors and using the hierarchical analysis method to solve the road communication status index of the target layer includes: The analytic hierarchy process is used to divide the solution of the road traffic status into a target layer, a criterion layer and a solution layer. The target layer is the road traffic status, the criterion layer is the road conditions and the road surrounding environment, the solution layer factors corresponding to the road conditions are the flow of the path length and the road width and slope, and the solution layer factors corresponding to the road surrounding environment are the surrounding environment parameters of the path length and the risk event impact parameters. The relative importance of each factor in the criterion layer and the scheme layer to the road traffic state is set, and a road traffic state judgment matrix, a road condition judgment matrix, and a road surrounding environment judgment matrix are constructed; Obtain the influence weights of the scheme layer factors on the target layer according to the road traffic state judgment matrix, the road condition judgment matrix and the road surrounding environment judgment matrix; The road traffic status index of the path length is obtained according to the influence weights of the scheme layer factors on the target layer.

6. The method according to claim 5, characterized in that The step of obtaining the road traffic status index of the path length according to the influence weights of the scheme layer factors on the target layer includes: The scheme layer factors are dimensionally processed respectively, and the products of all dimensionally processed scheme layer factors and the corresponding scheme layer factors' influence weights on the target layer are accumulated to obtain the road traffic status index of the path length.

7. The method according to claim 6, characterized in that The obtaining of the obstacle avoidance cost of the AMR avoiding obstacles on the path trajectory includes: A bounding box with preset length, width and height is used to replace the outline of the AMR, the current distance between any obstacle in the path trajectory and the bounding box is obtained, and the response distance, buffer distance and danger distance are set; When the current distance between the obstacle in the path trajectory and the bounding box is between the response distance and the buffer distance, the obstacle avoidance cost is a preset avoidance fixed value; When the current distance between the obstacle and the bounding box in the path trajectory is between the buffer distance and the danger distance, the obstacle avoidance cost is obtained according to the preset avoidance fixed value and the difference between the current distance and the danger distance.

8. AMR path planning device based on digital twin, characterized in that: The device comprises: Path search module: used to obtain the starting point and end point of the AMR, and use the path search algorithm to obtain all path trajectories from the starting point to the end point; Path length acquisition module: used to process the AMR operating environment into a cube grid of a preset size, then any path track passes through multiple cube grids, obtain the intersection of any path track and each cube grid, and obtain the path length of the path track in a cube grid according to the intersection coordinates; Path parameter acquisition module: used to acquire the path parameters of the path length, and acquire the flow rate, road width and slope, and surrounding environment parameters of the path length according to the path parameters; Dangerous event impact acquisition module: used to build a digital twin model of the AMR operating environment, simulate various dangerous events on the path length through the digital twin model, obtain the severity parameter of each dangerous event, and obtain the dangerous event impact parameter according to the severity parameter of each dangerous event; Hierarchical analysis module: used to use the flow of the path length, road width and slope, surrounding environmental parameters and dangerous event impact parameters as scheme layer factors, and use hierarchical analysis method to solve the road communication status index of the target layer; An equivalent length acquisition module: used to acquire the path equivalent length of the path length according to the road traffic status index of the path length; and to acquire the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory; Obstacle avoidance cost acquisition module: used to obtain the obstacle avoidance cost of the AMR avoiding obstacles on the path trajectory; A path cost acquisition module: used to acquire the path cost of the path trajectory according to the path equivalent lengths corresponding to all cubic grids passed by the AMR in the path trajectory and the obstacle avoidance cost; Optimal path confirmation module: used to obtain the path costs corresponding to all path trajectories and select the path trajectory with the smallest path cost as the optimal path.

9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the main controller, each step in the AMR path planning method based on digital twins as described in any one of claims 1 to 7 is implemented.

Images