Method and device for evaluating and planning path of AGV (Automatic Guided Vehicle)

Abstract

The invention discloses an AGV path evaluation planning method and device, and the method comprises the steps: generating a plurality of driving paths based on a starting point and a target point of an AGV through a preset path planning model, and enabling the preset path planning model to comprise a long and short term memory module and a priority experience playback mechanism; performing quantitative evaluation on path obstacles, path energy consumption and path safety of each driving path to obtain a quantitative evaluation result corresponding to each driving path; and selecting an optimal driving path of the AGV from the driving paths based on the quantitative evaluation results. According to the invention, the preset path planning model comprising the long and short term memory module and the priority experience playback mechanism is used to generate the plurality of driving paths based on the starting point and the target point of the AGV, then quantitative evaluation is carried out on each driving path from multiple angles, and the optimal driving path is selected based on the quantitative evaluation result. And the AGV path planning quality and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an AGV (Automated Guided Vehicle) vehicle path evaluation and planning method and device. Background Art

[0002] Automated Guided Vehicle (AGV) refers to a transport vehicle with an automatic guidance system that can run between the starting point and the end point according to a specified route or an automatically generated route. It has certain safety protections and integrates sound, light, electricity, and computer technologies. It integrates the advanced theories and application technologies in today's science and technology field. It is widely used in flexible manufacturing systems and automated factories. It has many advantages such as high transportation efficiency, energy saving, reliable operation, and flexible transportation, which greatly improves the degree of automation and production efficiency.

[0003] The current AGV path planning technology mainly relies on classic algorithms, such as the A* algorithm and the Dijkstra algorithm. Although these algorithms are mature and widely used, they are not only inefficient in planning, but also show certain limitations when dealing with complex dynamic environments. Especially in high-dimensional dynamic environments, traditional centralized path planning algorithms are inefficient in dealing with real-time changes and multi-task requirements, and are difficult to flexibly adapt to temporarily changing tasks and environments. In addition, when AGVs perform work tasks, they often need to plan the path to the target point in advance to avoid collisions with the environment or other AGVs. Previous path planning methods often only consider the distance between the current point and the target point, or select the fastest path possible by adding turning angle constraints. When implementing different work tasks, the path generated by a single strategy cannot achieve global optimization, which requires multi-angle evaluation of the paths generated by different strategies to find the optimal path under each working environment.

[0004] Therefore, there is an urgent need for an AGV path evaluation and planning method that can effectively improve the quality and efficiency of AGV path planning. Summary of the invention

[0005] The main purpose of the present invention is to provide an AGV vehicle path evaluation and planning method and device, aiming to solve the technical problems in the prior art that the path planning of the AGV vehicle is inefficient when responding to real-time changes and multi-task requirements, and is difficult to flexibly adapt to temporarily changing tasks and environments, resulting in low quality and efficiency of the AGV vehicle path planning.

[0006] To achieve the above object, the present invention provides an AGV path evaluation and planning method, the method comprising the following steps:

[0007] Using a preset path planning model, multiple driving paths are generated based on the starting point and target point of the AGV, wherein the preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism;

[0008] Quantitatively evaluate the path obstacles, path energy consumption and path safety of each of the driving paths to obtain quantitative evaluation results corresponding to each of the driving paths;

[0009] Based on each of the quantitative evaluation results, the best driving path of the AGV is selected from each of the driving paths.

[0010] Optionally, before the step of using a preset path planning model to generate multiple driving paths based on the starting point and the target point of the AGV, the step further includes:

[0011] Building an initial path planning model based on a proximal strategy optimization algorithm, wherein the initial path planning model includes a long-short term memory module and a priority experience playback mechanism;

[0012] Obtain historical operation data of the AGV, and generate model training data based on the historical operation data;

[0013] Performing model training on the initial path planning model based on the model training data to obtain a training result;

[0014] The model parameters of the initial path planning model are optimized according to the training results to obtain a preset path planning model.

[0015] Optionally, the step of using a preset path planning model to generate multiple driving paths based on the starting point and the target point of the AGV vehicle includes:

[0016] Obtaining the operating environment information of the AGV, and constructing an operating environment map based on the operating environment information;

[0017] Rasterizing the operating environment map to obtain a grid map, and determining the starting point and the target point of the AGV in the grid map;

[0018] Using a preset path planning model, multiple driving paths from the starting point to the target point are searched in the grid map.

[0019] Optionally, the step of performing quantitative evaluation on path obstacles, path energy consumption and path safety of each of the driving paths to obtain the quantitative evaluation results corresponding to each of the driving paths includes:

[0020] Quantitatively evaluate the path obstacles of each of the driving paths using a conflict point waiting strategy and a conflict point avoidance strategy to obtain a first evaluation result;

[0021] constructing a weighted directed graph model based on each of the driving paths, and performing quantitative evaluation on the path energy consumption of each of the driving paths based on the weighted directed graph model to obtain a second evaluation result;

[0022] Based on the real-time operation data and safety evaluation index of the AGV, a quantitative evaluation is performed on the path safety of each of the driving paths to obtain a third evaluation result;

[0023] The quantitative evaluation results corresponding to each of the driving paths are determined according to the first evaluation result, the second evaluation result and the third evaluation result.

[0024] Optionally, the step of using the conflict point waiting strategy and the conflict point avoidance strategy to quantitatively evaluate the path obstacles of each of the driving paths to obtain a first evaluation result includes:

[0025] Based on the conflict point waiting strategy, determining the first operation data of the AGV vehicle in each of the driving paths;

[0026] Based on the conflict point avoidance strategy, determining the second operation data of the AGV vehicle in each of the driving paths;

[0027] A quantitative evaluation is performed on the path obstacles of each of the driving paths based on the first operating data and the second operating data to obtain a first evaluation result.

[0028] Optionally, the step of performing quantitative evaluation on the path obstacles of each of the driving paths based on the first operating data and the second operating data to obtain a first evaluation result includes:

[0029] Determine the number of conflict points, waiting time and conflict resolution efficiency in each of the driving paths according to the first operation data;

[0030] Determine path complexity change information, travel time change information, and energy consumption change information resulting from conflict point avoidance according to the second operation data;

[0031] Based on the number of conflict points, the waiting time, the conflict resolution efficiency, the path complexity change information, the driving time change information and the energy consumption change information, the path obstacles of each driving path are quantitatively evaluated to obtain a first evaluation result.

[0032] Optionally, the step of constructing a weighted directed graph model based on each of the driving paths, and quantitatively evaluating the path energy consumption of each of the driving paths based on the weighted directed graph model to obtain a second evaluation result includes:

[0033] Constructing a weighted directed graph model based on each of the driving paths, wherein the nodes in the weighted directed graph model are key points on the path, and the weights of the edges in the weighted directed graph model are corresponding energy consumption values;

[0034] Determining a path energy consumption value of each of the driving paths based on the weighted directed graph model;

[0035] A second evaluation result is determined according to the path energy consumption value of each of the driving paths and the transportation cost of each of the driving paths.

[0036] Optionally, the step of performing a quantitative evaluation on the path safety of each of the driving paths based on the real-time operation data and the safety evaluation index of the AGV to obtain a third evaluation result includes:

[0037] Based on the safety assessment index, determine the number of collisions, close-range events and emergency braking in the real-time operation data of the AGV;

[0038] It is beneficial to preset a path safety scoring formula to determine the safety score of each driving path according to the number of collisions, the close-range events and the number of emergency brakes;

[0039] The safety score of each of the driving routes is used as the third evaluation result.

[0040] In addition, to achieve the above-mentioned purpose, the present invention also proposes an AGV vehicle path evaluation and planning device, the device comprising:

[0041] A path generation module, which is used to generate multiple driving paths based on the starting point and target point of the AGV car by using a preset path planning model, wherein the preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism;

[0042] A quantitative evaluation module, used to quantitatively evaluate the path obstacles, path energy consumption and path safety of each of the driving paths, and obtain the quantitative evaluation results corresponding to each of the driving paths;

[0043] A path selection module is used to select the best driving path of the AGV from the driving paths based on the quantitative evaluation results.

[0044] Optionally, the device further comprises:

[0045] A model building module, used to build an initial path planning model based on a proximal strategy optimization algorithm, wherein the initial path planning model includes a long-short term memory module and a priority experience playback mechanism;

[0046] A data generation module is used to obtain the historical operation data of the AGV and generate model training data based on the historical operation data;

[0047] A model training module, used to perform model training on the initial path planning model based on the model training data to obtain a training result;

[0048] The model optimization module is used to optimize the model parameters of the initial path planning model according to the training results to obtain a preset path planning model.

[0049] In addition, to achieve the above-mentioned purpose, the present invention also proposes an AGV vehicle path evaluation and planning device, which includes: a memory, a processor, and an AGV vehicle path evaluation and planning program stored in the memory and executable on the processor, and the AGV vehicle path evaluation and planning program is configured to implement the steps of the AGV vehicle path evaluation and planning method described above.

[0050] In addition, to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which an AGV vehicle path evaluation planning program is stored. When the AGV vehicle path evaluation planning program is executed by a processor, the steps of the AGV vehicle path evaluation planning method described above are implemented.

[0051] The present invention discloses the use of a preset path planning model to generate multiple driving paths based on the starting point and target point of an AGV vehicle, wherein the preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism; quantitatively evaluate the path obstacles, path energy consumption and path safety of each of the driving paths to obtain a quantitative evaluation result corresponding to each of the driving paths; and select the best driving path of the AGV vehicle from each of the driving paths based on the quantitative evaluation results. Since the present invention is conducive to the preset path planning model including a long-term and short-term memory module and a priority experience playback mechanism to generate multiple driving paths based on the starting point and target point of the AGV vehicle, and then quantitatively evaluate each driving path from multiple angles, and select the best driving path based on the quantitative evaluation results, compared with the prior art, the present invention improves the quality and efficiency of AGV vehicle path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flow chart of the first embodiment of the AGV vehicle path evaluation and planning method of the present invention;

[0053] Figure 2 This is a flow chart of a second embodiment of the AGV vehicle path evaluation and planning method of the present invention;

[0054] Figure 3 It is a flow chart of the third embodiment of the AGV vehicle path evaluation and planning method of the present invention;

[0055] Figure 4It is a structural block diagram of the first embodiment of the AGV vehicle path evaluation and planning device of the present invention;

[0056] Figure 5 It is a structural diagram of an AGV vehicle path evaluation and planning device in a hardware operating environment involved in an embodiment of the present invention.

[0057] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0058] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0059] The embodiment of the present invention provides an AGV vehicle path evaluation and planning method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the AGV path evaluation and planning method of the present invention.

[0060] In this embodiment, the AGV vehicle path evaluation and planning method includes the following steps S10 to S30:

[0061] Step S10: Generate multiple driving paths based on the starting point and target point of the AGV by using a preset path planning model, wherein the preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism.

[0062] It should be noted that the execution subject of this embodiment can be a computer server device with data processing, network communication and program running functions used in the AGV vehicle path evaluation and planning scenario, such as a server, a tablet computer, a smart phone, a smart watch, etc., or an electronic device capable of realizing the above functions, an AGV vehicle path evaluation and planning device, etc. The following takes a multi-agent distributed control system (hereinafter referred to as the system) including an AGV vehicle path evaluation and planning device as an example to illustrate this embodiment and the following embodiments.

[0063] It should be understood that the existing technology mostly uses a centralized control system for path planning, while the present invention uses a multi-agent distributed control system for path planning. Each AGV independently selects and optimizes the path, does not rely on the central control system, and avoids the single point failure problem caused by centralized control. Each AGV can make independent decisions based on the real-time environment and can quickly respond to task changes, which improves the robustness and real-time performance of the system.

[0064] In a specific implementation, in order to improve the efficiency of data processing, the operating environment information of the AGV can be obtained, and an operating environment map can be constructed based on the operating environment information; the operating environment map can be rasterized to obtain a grid map, and the starting point and target point of the AGV can be determined in the grid map; and a preset path planning model can be used to search for multiple driving paths from the starting point to the target point in the grid map.

[0065] It should be explained that the above-mentioned preset path planning model can be constructed based on the proximal policy optimization algorithm (PPO algorithm), and the above-mentioned preset path planning model also includes a long short-term memory module and a priority experience playback mechanism. The PPO) algorithm is used for strategy optimization, combined with the long short-term memory module to process timing information, and multiple driving paths can be generated in a dynamic environment.

[0066] It should be understood that the PPO algorithm is a strategy optimization algorithm. By limiting the range of changes between the new and old strategies, the stability of the training process is ensured. The PPO algorithm performs well in processing high-dimensional continuous action spaces and is suitable for AGV path planning. The objective function of the PPO algorithm is:

[0067]

[0068]

[0069] In the formula, r t (θ) represents the importance sampling ratio, which is used to measure the difference between the new policy and the old policy. π e and are the new and old strategies, A t is the advantage function, used to estimate t Take action a t The relative value to the average, clip(r t (θ), 1-∈, 1+∈) is the clipping function, ∈ is a hyperparameter, θ represents the parameters of the proximal policy optimization algorithm, is the estimated advantage function value.

[0070] The long short-term memory module can be a module containing a recursive neural network that processes time series information, which can remember important information in a long time series and ignore irrelevant information. In the path planning of the AGV, the long short-term memory module can process the time step information in the path planning process and improve the efficiency and effect of the path planning of the AGV.

[0071] Prioritized Experience Replay (PER) mechanism is an important method in deep reinforcement learning, which aims to improve training efficiency and performance. The priority experience replay mechanism in this embodiment can be based on the priority sorting of AGV car task rewards: sorting by factors such as task completion, time consumption, number of collisions, etc., giving priority to training key paths, improving learning efficiency and model convergence speed. Priority calculation formula:

[0072]

[0073] Where, d c is the distance from the AGV to the target point, r ah is the distance threshold to reach the target point, P(e) represents the priority of event e, and reward represents the reward value.

[0074] Step S20: performing quantitative evaluation on the path obstacles, path energy consumption and path safety of each of the driving paths, and obtaining the quantitative evaluation results corresponding to each of the driving paths.

[0075] It can be understood that quantitative evaluation is a method of evaluating and measuring the characteristics, behaviors or results of things through mathematical methods and data analysis techniques. It involves the collection, processing, analysis and interpretation of relevant data, and converting these complex problems into specific quantitative data to provide more accurate judgment and decision-making basis.

[0076] It should be understood that the quantitative evaluation of path obstacles can focus on the obstacles that the AGV may encounter during driving and its obstacle avoidance capabilities. The quantitative evaluation of path energy consumption can focus on the energy consumed by the AGV during driving. The quantitative evaluation of path safety can focus on the safety risks and protective measures that the AGV may encounter during driving.

[0077] Step S30: selecting the best driving path of the AGV from the driving paths based on the quantitative evaluation results.

[0078] In a specific implementation, the driving path with the highest score can be selected from the quantitative evaluation results corresponding to each of the driving paths, and then the driving path with the highest score can be used as the optimal driving path for the AGV. Compared with the prior art, this embodiment performs quantitative evaluations on each driving path from multiple angles, and selects the optimal driving path based on the quantitative evaluation results, which effectively solves the technical problem that the path planning of the AGV in the prior art is inefficient when responding to real-time changes and multi-task requirements, and is difficult to flexibly adapt to temporarily changing tasks and environments, resulting in low quality and efficiency of the path planning of the AGV.

[0079] This embodiment discloses the use of a preset path planning model to generate multiple driving paths based on the starting point and target point of the AGV vehicle, wherein the preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism; quantitatively evaluate the path obstacles, path energy consumption and path safety of each of the driving paths to obtain a quantitative evaluation result corresponding to each of the driving paths; and select the best driving path of the AGV vehicle from each of the driving paths based on the quantitative evaluation results. Since this embodiment facilitates the preset path planning model including a long-term and short-term memory module and a priority experience playback mechanism to generate multiple driving paths based on the starting point and target point of the AGV vehicle, and then quantitatively evaluate each driving path from multiple angles, and select the best driving path based on the quantitative evaluation results, compared with the prior art, this embodiment improves the quality and efficiency of AGV vehicle path planning.

[0080] refer to Figure 2 , Figure 2 This is a flow chart of the second embodiment of the AGV path evaluation and planning method of the present invention.

[0081] Based on the first embodiment described above, in this embodiment, before step S10, steps S01 to S04 are further included:

[0082] Step S01: constructing an initial path planning model based on a proximal strategy optimization algorithm, wherein the initial path planning model includes a long-short term memory module and a priority experience playback mechanism.

[0083] Step S02: Acquire historical operation data of the AGV, and generate model training data based on the historical operation data.

[0084] Step S03: Performing model training on the initial path planning model based on the model training data to obtain a training result.

[0085] Step S04: Optimizing model parameters of the initial path planning model according to the training results to obtain a preset path planning model.

[0086] This embodiment discloses constructing an initial path planning model based on a proximal strategy optimization algorithm, wherein the initial path planning model includes a long-term and short-term memory module and a priority experience playback mechanism; obtaining historical operation data of the AGV car, and generating model training data based on the historical operation data; performing model training on the initial path planning model based on the model training data to obtain training results; optimizing model parameters of the initial path planning model according to the training results to obtain a preset path planning model. Since this embodiment constructs an initial path planning model based on a proximal strategy optimization algorithm, and then performs model training on the initial path planning model based on the model training data, and optimizes model parameters according to the training results, compared with the prior art, this embodiment improves the accuracy and generalization ability of the AGV car path planning.

[0087] refer to Figure 3 , Figure 3 It is a flow chart of the third embodiment of the AGV vehicle path evaluation and planning method of the present invention.

[0088] Based on the above embodiments, in this embodiment, step S20 includes steps S201 to S204:

[0089] Step S201: Quantitatively evaluate the path obstacles of each of the driving paths using the conflict point waiting strategy and the conflict point avoiding strategy to obtain a first evaluation result.

[0090] It should be noted that the conflict point waiting (CPW) strategy can be used to identify potential conflict points when multiple AGVs are running in the same area. The AGV with lower priority waits at the conflict point until the AGV with higher priority passes. Specifically, all possible conflict points are identified during the path planning process; each conflict point is assigned a priority, usually based on the distance from the conflict point or the urgency of the task; the running status of the AGV is monitored in real time, and when a conflict is detected, the speed of the AGV with lower priority is adjusted or its driving is suspended until the conflict is resolved.

[0091] The conflict point avoidance (CPA) strategy can be to re-plan the path to avoid conflicts, choose alternative routes to bypass conflict points, and ensure the smoothness and safety of the AGV operation. Specifically, all possible conflict points are identified during the path planning process; a dynamic path planning algorithm (such as the Dijkstra algorithm or the A* algorithm) is used to adjust the path of the AGV in real time to bypass the conflict points; the running status of the AGV is monitored, and when a conflict is detected, the path is immediately recalculated to avoid the conflict point.

[0092] In a specific implementation, the first operating data of the AGV vehicle in each of the driving paths can be determined based on a conflict point waiting strategy; the second operating data of the AGV vehicle in each of the driving paths can be determined based on a conflict point avoidance strategy; and the path obstacles of each of the driving paths can be quantitatively evaluated based on the first operating data and the second operating data to obtain a first evaluation result.

[0093] It should be noted that the above-mentioned step of performing quantitative evaluation on the path obstacles of each of the driving paths based on the first operating data and the second operating data to obtain the first evaluation result may include: determining the number of conflict points, waiting time and conflict resolution efficiency in each of the driving paths according to the first operating data; determining the path complexity change information, driving time change information and energy consumption change information caused by conflict point avoidance according to the second operating data; performing quantitative evaluation on the path obstacles of each of the driving paths based on the number of conflict points, the waiting time, the conflict resolution efficiency, the path complexity change information, the driving time change information and the energy consumption change information to obtain the first evaluation result.

[0094] Step S202: constructing a weighted directed graph model based on each of the driving paths, and performing quantitative evaluation on the path energy consumption of each of the driving paths based on the weighted directed graph model to obtain a second evaluation result.

[0095] In a specific implementation, a weighted directed graph model can be constructed based on each of the driving paths, wherein the nodes in the weighted directed graph model are key points on the path, and the weights of the edges in the weighted directed graph model are corresponding energy consumption values; the path energy consumption values ​​of each of the driving paths are determined based on the weighted directed graph model; and the second evaluation result is determined according to the path energy consumption values ​​of each of the driving paths and the transportation cost of each of the driving paths. The calculation formula of the second evaluation result can be:

[0096] C total =∑ (i，j)∈Path (e ij +λt ij );

[0097] In the formula, a node pair on the path, that is, the edge from node i to node , , Path represents the set of all possible paths in the weighted directed graph model. ij represents the path energy consumption value from node i to node , t ij represents the transportation cost from node i to node , and λ represents the weight value.

[0098] Step S203: quantitatively evaluating the path safety of each of the driving paths based on the real-time operation data and safety evaluation index of the AGV to obtain a third evaluation result.

[0099] It should be noted that the above safety assessment indicators may include the number of collisions, close-range incidents and emergency braking times.

[0100] In a specific implementation, the number of collisions, close-range events and emergency brakes in the real-time operation data of the AGV can be determined based on the safety evaluation index; the preset path safety scoring formula is used to determine the safety score of each driving path based on the number of collisions, close-range events and emergency brakes; and the safety score of each driving path is used as the third evaluation result.

[0101] The above preset path safety score formula is:

[0102]

[0103] In the formula, S safe represents the safety score, N can represent the number of steps in the path, and w collide Represents the weight of the collision event, n collide represents the number of collisions, w near represents the weight of close events, n near represents the number of close events, w brake Indicates the weight of the emergency braking event, n brake Indicates the number of emergency brakes.

[0104] Step S204: Determine the quantitative evaluation result corresponding to each of the driving paths according to the first evaluation result, the second evaluation result and the third evaluation result.

[0105] This embodiment discloses using a conflict point waiting strategy and a conflict point avoidance strategy to quantitatively evaluate the path obstacles of each of the driving paths to obtain a first evaluation result; constructing a weighted directed graph model based on each of the driving paths, and quantitatively evaluating the path energy consumption of each of the driving paths based on the weighted directed graph model to obtain a second evaluation result; quantitatively evaluating the path safety of each of the driving paths based on the real-time operation data and safety evaluation indicators of the AGV car to obtain a third evaluation result; and determining the quantitative evaluation results corresponding to each of the driving paths based on the first evaluation result, the second evaluation result, and the third evaluation result. Compared with the prior art, this embodiment quantitatively evaluates each of the driving paths from multiple angles, effectively improving the reliability of the quantitative evaluation results, and thereby improving the accuracy of path planning.

[0106] In addition, an embodiment of the present invention further proposes a storage medium, on which an AGV vehicle path evaluation and planning program is stored. When the AGV vehicle path evaluation and planning program is executed by a processor, the steps of the AGV vehicle path evaluation and planning method described above are implemented.

[0107] Reference Figure 4 , Figure 4 This is a structural block diagram of the first embodiment of the AGV vehicle path evaluation and planning device of the present invention.

[0108] like Figure 4 As shown, the AGV vehicle path evaluation and planning device proposed in the embodiment of the present invention includes: a path generation module block 401, a quantitative evaluation module 402 and a path selection module 403.

[0109] The path generation module 401 is used to generate multiple driving paths based on the starting point and target point of the AGV vehicle using a preset path planning model. The preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism.

[0110] The quantitative evaluation module 402 is used to perform quantitative evaluation on the path obstacles, path energy consumption and path safety of each of the driving paths, and obtain the quantitative evaluation results corresponding to each of the driving paths.

[0111] The path selection module 403 is used to select the best driving path of the AGV from the driving paths based on the quantitative evaluation results.

[0112] The path generation module 401 is also used to obtain the operating environment information of the AGV vehicle and construct an operating environment map based on the operating environment information; rasterize the operating environment map to obtain a raster map, and determine the starting point and target point of the AGV vehicle in the raster map; and use a preset path planning model to search for multiple driving paths from the starting point to the target point in the raster map.

[0113] The present device embodiment discloses the use of a preset path planning model to generate multiple driving paths based on the starting point and target point of the AGV trolley, wherein the preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism; quantitatively evaluates the path obstacles, path energy consumption and path safety of each of the driving paths to obtain a quantitative evaluation result corresponding to each of the driving paths; and selects the best driving path of the AGV trolley from each of the driving paths based on the quantitative evaluation results. Since the present device embodiment is conducive to the preset path planning model including a long-term and short-term memory module and a priority experience playback mechanism to generate multiple driving paths based on the starting point and target point of the AGV trolley, and then quantitatively evaluates each driving path from multiple angles, and selects the best driving path based on the quantitative evaluation results, compared with the prior art, the present device embodiment improves the quality and efficiency of the AGV trolley path planning.

[0114] Based on the first embodiment of the AGV path evaluation and planning device of the present invention, a second embodiment of the AGV path evaluation and planning device of the present invention is proposed.

[0115] In this embodiment, the AGV vehicle path evaluation and planning device further includes:

[0116] The model building module is used to build an initial path planning model based on a proximal strategy optimization algorithm, wherein the initial path planning model includes a long-short term memory module and a priority experience playback mechanism.

[0117] The data generation module is used to obtain the historical operation data of the AGV and generate model training data based on the historical operation data.

[0118] The model training module is used to perform model training on the initial path planning model based on the model training data to obtain a training result.

[0119] The model optimization module is used to optimize the model parameters of the initial path planning model according to the training results to obtain a preset path planning model.

[0120] Other embodiments or specific implementations of the AGV vehicle path evaluation and planning device of the present invention can refer to the above-mentioned method embodiments and will not be described in detail here.

[0121] The present application provides an AGV vehicle path evaluation and planning device, the AGV vehicle path evaluation and planning device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the AGV vehicle path evaluation and planning method in the above-mentioned embodiment one.

[0122] Reference below Figure 5 , which shows a schematic diagram of the structure of an AGV vehicle path evaluation and planning device suitable for implementing the embodiment of the present application. The AGV vehicle path evaluation and planning device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The AGV path evaluation and planning device shown is merely an example and should not impose any limitations on the functions and scope of use of the embodiments of the present application.

[0123] like Figure 5 As shown, the AGV trolley path evaluation and planning device may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the AGV trolley path evaluation and planning device are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the AGV trolley path evaluation and planning device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows an AGV trolley path evaluation and planning device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0124] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0125] The AGV trolley path evaluation and planning device provided by the present application adopts the AGV trolley path evaluation and planning method in the above-mentioned embodiment, which can solve the technical problem that the path planning of the AGV trolley in the prior art is inefficient when responding to real-time changes and multi-task requirements, and it is difficult to flexibly adapt to temporarily changing tasks and environments, resulting in low quality and efficiency of the AGV trolley path planning. Compared with the prior art, the beneficial effects of the AGV trolley path evaluation and planning device provided by the present application are the same as the beneficial effects of the AGV trolley path evaluation and planning method provided by the above-mentioned embodiment, and the other technical features in the AGV trolley path evaluation and planning device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0126] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0127] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0128] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0129] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0131] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for evaluating and planning the path of an AGV vehicle, characterized in that: The method comprises: Using a preset path planning model, multiple driving paths are generated based on the starting point and target point of the AGV, wherein the preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism; Quantitatively evaluate the path obstacles, path energy consumption and path safety of each of the driving paths to obtain quantitative evaluation results corresponding to each of the driving paths; Based on each of the quantitative evaluation results, the best driving path of the AGV is selected from each of the driving paths.

2. The AGV vehicle path evaluation and planning method according to claim 1, characterized in that: Before the step of using the preset path planning model to generate multiple driving paths based on the starting point and the target point of the AGV, it also includes: Building an initial path planning model based on a proximal strategy optimization algorithm, wherein the initial path planning model includes a long-short term memory module and a priority experience playback mechanism; Obtain historical operation data of the AGV, and generate model training data based on the historical operation data; Performing model training on the initial path planning model based on the model training data to obtain a training result; The model parameters of the initial path planning model are optimized according to the training results to obtain a preset path planning model.

3. The AGV vehicle path evaluation and planning method according to claim 1, characterized in that: The step of using the preset path planning model to generate multiple driving paths based on the starting point and the target point of the AGV vehicle includes: Obtaining the operating environment information of the AGV, and constructing an operating environment map based on the operating environment information; Rasterizing the operating environment map to obtain a grid map, and determining the starting point and the target point of the AGV in the grid map; Using a preset path planning model, multiple driving paths from the starting point to the target point are searched in the grid map.

4. The AGV vehicle path evaluation and planning method according to claim 1, characterized in that: The step of quantitatively evaluating the path obstacles, path energy consumption and path safety of each of the driving paths to obtain the quantitative evaluation results corresponding to each of the driving paths includes: Quantitatively evaluate the path obstacles of each of the driving paths using a conflict point waiting strategy and a conflict point avoidance strategy to obtain a first evaluation result; constructing a weighted directed graph model based on each of the driving paths, and performing quantitative evaluation on the path energy consumption of each of the driving paths based on the weighted directed graph model to obtain a second evaluation result; Based on the real-time operation data and safety evaluation index of the AGV, a quantitative evaluation is performed on the path safety of each of the driving paths to obtain a third evaluation result; The quantitative evaluation results corresponding to each of the driving paths are determined according to the first evaluation result, the second evaluation result and the third evaluation result.

5. The AGV vehicle path evaluation and planning method according to claim 4, characterized in that: The step of using the conflict point waiting strategy and the conflict point avoidance strategy to quantitatively evaluate the path obstacles of each of the driving paths to obtain a first evaluation result includes: Based on the conflict point waiting strategy, determining the first operation data of the AGV vehicle in each of the driving paths; Based on the conflict point avoidance strategy, determining the second operation data of the AGV vehicle in each of the driving paths; A quantitative evaluation is performed on the path obstacles of each of the driving paths based on the first operating data and the second operating data to obtain a first evaluation result.

6. The AGV vehicle path evaluation and planning method according to claim 5, characterized in that: The step of performing quantitative evaluation on the path obstacles of each of the driving paths based on the first operating data and the second operating data to obtain a first evaluation result comprises: Determine the number of conflict points, waiting time and conflict resolution efficiency in each of the driving paths according to the first operation data; Determine path complexity change information, travel time change information, and energy consumption change information resulting from conflict point avoidance according to the second operation data; Based on the number of conflict points, the waiting time, the conflict resolution efficiency, the path complexity change information, the driving time change information and the energy consumption change information, the path obstacles of each driving path are quantitatively evaluated to obtain a first evaluation result.

7. The AGV vehicle path evaluation and planning method according to claim 4, characterized in that: The step of constructing a weighted directed graph model based on each of the driving paths, and quantitatively evaluating the path energy consumption of each of the driving paths based on the weighted directed graph model to obtain a second evaluation result includes: Constructing a weighted directed graph model based on each of the driving paths, wherein the nodes in the weighted directed graph model are key points on the path, and the weights of the edges in the weighted directed graph model are corresponding energy consumption values; Determining a path energy consumption value of each of the driving paths based on the weighted directed graph model; A second evaluation result is determined according to the path energy consumption value of each of the driving paths and the transportation cost of each of the driving paths.

8. The AGV vehicle path evaluation and planning method according to claim 4, characterized in that: The step of quantitatively evaluating the path safety of each of the driving paths based on the real-time operation data and safety evaluation index of the AGV to obtain a third evaluation result includes: Based on the safety assessment index, determine the number of collisions, close-range events and emergency braking in the real-time operation data of the AGV; It is beneficial to preset a path safety scoring formula to determine the safety score of each driving path according to the number of collisions, the close-range events and the number of emergency brakes; The safety score of each of the driving routes is used as the third evaluation result.

9. An AGV vehicle path evaluation and planning device, characterized in that: The device comprises: A path generation module, which is used to generate multiple driving paths based on the starting point and target point of the AGV car by using a preset path planning model, wherein the preset path planning model includes a long-term and short-term memory module and a priority experience playback mechanism; A quantitative evaluation module, used to quantitatively evaluate the path obstacles, path energy consumption and path safety of each of the driving paths, and obtain the quantitative evaluation results corresponding to each of the driving paths; A path selection module is used to select the best driving path of the AGV from the driving paths based on the quantitative evaluation results.

10. The AGV vehicle path evaluation and planning device according to claim 9, characterized in that: The device also includes: A model building module, used to build an initial path planning model based on a proximal strategy optimization algorithm, wherein the initial path planning model includes a long-short term memory module and a priority experience playback mechanism; A data generation module is used to obtain the historical operation data of the AGV and generate model training data based on the historical operation data; A model training module, used to perform model training on the initial path planning model based on the model training data to obtain a training result; The model optimization module is used to optimize the model parameters of the initial path planning model according to the training results to obtain a preset path planning model.

Images