Agent processing method, electronic device, and computer-readable storage medium

Abstract

The present disclosure relates to the technical fields of large model technology and machine learning. Disclosed are an agent processing method, an electronic device, and a computer-readable storage medium. The method comprises: determining a region to be monitored; acquiring map information of the region and at least one agent located in the region; acquiring simulation information associated with the map information and the at least one agent, wherein the simulation information is used for representing historical behavior information of the agent in the region corresponding to the map information, and / or simulation rule information used when the agent is simulated in the region; and using the simulation information to simulate the behavior of the at least one agent in the region to obtain at least one simulation result.

Description

[0001] Cross-Reference This disclosure claims priority to Chinese patent application No. 202410362639.8, filed with the Patent Office of the People's Republic of China on March 27, 2024, entitled "Processing Method, Electronic Device, and Computer-Readable Storage Medium for Intelligent Agents," the entire contents of which are incorporated herein by reference. Technical Field This disclosure relates to the fields of large model technology and machine learning technology, and more specifically, to a processing method, electronic device, and computer-readable storage medium for intelligent agents. Background: With the acceleration of urbanization and the rapid expansion of urban scale, the complexity of urban management and planning has increased. The concept of smart cities has emerged. Smart cities aim to utilize information and communication technologies to optimize urban functions, enhance urban operational efficiency, and improve the quality of life for residents. In related technologies, multi-agent behavior prediction neural networks and trajectory prediction neural networks can be used to generate behavior or trajectory prediction results. However, these predictions can only provide short-term behavior or trajectory predictions, such as simulating the short-term behavior of autonomous driving rules in urban traffic. They cannot accurately predict traffic conditions and agent behavior over longer periods of time, nor can they accurately simulate complex urban traffic conditions and the coordinated behavior between agents. Therefore, there is a technical problem of being unable to effectively simulate agent behavior. Currently, no effective solution has been proposed to address this issue. SUMMARY OF THE INVENTION The present disclosure provides an agent processing method, an electronic device, and a computer-readable storage medium to at least address the technical problem of being unable to effectively simulate agent behavior. According to one aspect of an embodiment of the present disclosure, a method for processing an intelligent agent is provided, comprising: determining an area to be monitored; obtaining map information of the area and at least one intelligent agent located within the area; obtaining simulation information associated with the map information and the at least one intelligent agent, wherein the simulation information is used to represent historical behavior information of the intelligent agent within the area corresponding to the map information and / or simulation rule information used when simulating the intelligent agent within the area; and using the simulation information, simulating the behavior of the at least one intelligent agent within the area to obtain at least one simulation result.According to another aspect of an embodiment of the present disclosure, a method for processing an intelligent agent is provided, comprising: determining an urban area to be monitored; obtaining map information of the urban area and at least one intelligent agent located within the urban area; obtaining simulation information associated with the map information and the at least one intelligent agent, wherein the simulation information is used to represent historical traffic behavior information of the intelligent agent within the urban area corresponding to the map information and / or simulation rule information used when simulating the intelligent agent within the area; using the simulation information, simulating the traffic behavior of the at least one intelligent agent within the urban area to obtain at least one simulation result; and transmitting the simulation result to an urban information monitoring platform. According to another aspect of an embodiment of the present disclosure, a method for processing an intelligent agent is provided, comprising: displaying an area to be monitored on the operating interface in response to an input instruction applied to the operating interface; and displaying at least one simulation result of at least one intelligent agent within the area on the operating interface in response to a simulation instruction applied to the operating interface, wherein the simulation result is obtained by simulating the behavior of the at least one intelligent agent within the area using simulation information, the simulation information being associated with map information of the area and the at least one intelligent agent, and being used to represent historical behavior information of the intelligent agent within the area corresponding to the map information, and / or simulation rule information used when simulating the intelligent agent within the area. According to another aspect of an embodiment of the present disclosure, a method for processing an intelligent agent is provided, comprising: obtaining map information of an area to be monitored by calling a first interface, wherein the first interface includes a first parameter, and the parameter value of the first parameter includes the map information; obtaining simulation information associated with the map information and at least one intelligent agent located in the area, wherein the simulation information is used to represent historical behavior information of the intelligent agent in the area corresponding to the map information and / or simulation rule information used when simulating the intelligent agent in the area; using the simulation information, simulating the behavior of the at least one intelligent agent in the area to obtain at least one simulation result; and outputting the simulation result by calling a second interface, wherein the second interface includes a second parameter, and the parameter value of the second parameter includes the simulation result. According to another aspect of an embodiment of the present disclosure, a processing device for an intelligent agent is further provided, comprising: a first determination component configured to determine an area to be monitored; a first acquisition component configured to acquire map information of the area and at least one intelligent agent located in the area; a second acquisition component configured to acquire simulation information associated with the map information and the at least one intelligent agent, wherein the simulation information is configured to represent historical behavior information of the intelligent agent in the area corresponding to the map information, and / or simulation rule information used when simulating the intelligent agent in the area; and a first simulation component configured to use the simulation information to simulate the behavior of the at least one intelligent agent in the area to obtain at least one simulation result.According to another aspect of the embodiment of the present disclosure, a processing device for an intelligent agent is also provided, comprising: a second determination component, configured to determine an urban area to be monitored; a third acquisition component, configured to obtain map information of the urban area and at least one intelligent agent located in the urban area; a fourth acquisition component, configured to obtain simulation information associated with the map information and the at least one intelligent agent, wherein the simulation information is configured to represent historical traffic behavior information of the intelligent agent in the urban area corresponding to the map information, and / or simulation rule information used when simulating the intelligent agent in the area; a second simulation component, configured to use the simulation information to simulate the traffic behavior of at least one intelligent agent in the urban area to obtain at least one simulation result; and a sending component, configured to send the simulation result to the urban information monitoring platform. According to another aspect of the embodiment of the present disclosure, a processing device for an intelligent agent is also provided, comprising: a first display component, configured to respond to an input instruction acting on an operation interface, and display an area to be monitored on the operation interface; a second display component, configured to respond to a simulation instruction acting on the operation interface, and display at least one simulation result of at least one intelligent agent in the area on the operation interface, wherein the simulation result is obtained by simulating the behavior of at least one intelligent agent in the area using simulation information, the simulation information is associated with map information of the area and at least one intelligent agent, and is configured to represent historical behavior information of the intelligent agent in the area corresponding to the map information, and / or simulation rule information used when simulating the intelligent agent in the area. According to another aspect of the embodiments of the present disclosure, an agent processing device is provided, comprising: a fifth acquisition component for acquiring map information of a to-be-monitored area by invoking a first interface, wherein the first interface includes a first parameter, the parameter value of the first parameter including the map information; a sixth acquisition component for acquiring simulation information associated with the map information and at least one agent located within the area, wherein the simulation information represents historical behavior information of the agent within the area corresponding to the map information and / or simulation rule information used when simulating the agent within the area; a third simulation component for using the simulation information to simulate the behavior of the at least one agent within the area to obtain at least one simulation result; and an output component for outputting the simulation result by invoking a second interface, wherein the second interface includes a second parameter, the parameter value of the second parameter including the simulation result. According to another aspect of the embodiments of the present disclosure, an electronic device is provided, comprising: a memory storing an executable program; and a processor for executing the program, wherein the program executes the agent processing method of the embodiments of the present disclosure when executed.According to another embodiment of the present disclosure, a computer terminal is provided, comprising: a memory storing an executable program; and a processor configured to execute the program, wherein when the program is executed, the method for processing an agent in each embodiment of the present disclosure is executed. According to another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device containing the computer-readable storage medium is controlled to execute the method for processing an agent in each embodiment of the present disclosure. According to another embodiment of the present disclosure, a computer program product is provided, comprising a computer program, wherein when executed by the processor, the computer program implements the method for processing an agent in each embodiment of the present disclosure. According to another embodiment of the present disclosure, a computer program product is provided, comprising a non-volatile computer-readable storage medium storing the computer program, wherein when executed by the processor, the computer program implements the method for processing an agent in each embodiment of the present disclosure. According to another embodiment of the present disclosure, a computer program is provided, wherein when executed by the processor, the computer program implements the method for processing an agent in each embodiment of the present disclosure. According to another aspect of the embodiments of the present disclosure, a computer program product is provided. The computer program product includes a non-volatile computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the agent processing methods described in various embodiments of the present disclosure. According to another aspect of the embodiments of the present disclosure, a computer program product is provided. When executed by a processor, the computer program product implements the agent processing methods described in various embodiments of the present disclosure. In an embodiment of the present disclosure, based on map information of a region to be monitored, historical behavior information of at least one agent within the region and simulation rule information for simulating the at least one agent within the region corresponding to the map information are obtained. Since the historical behavior information of an agent within the region can be used to indicate the agent's behavioral habits, the simulation rule information for simulating an agent within the region can be used to indicate the behavioral rules that the agent must comply with within the region. Therefore, by utilizing the historical behavior information of the agent within the region and the simulation rule information for simulating the agent within the region, the behavior of the agent within the region is simulated. This ensures a strong correlation between the simulation results and the historical behavior information and the simulation rule information, improving the accuracy of the agent simulation results. This results in a multi-agent simulation with controllable behaviors and solutions, resolving the technical issue of being unable to effectively simulate the behavior of the agents. It should be noted that the general description above and the detailed description that follow are merely illustrative and illustrative of the present disclosure and do not constitute a limitation of the present disclosure.BRIEF DESCRIPTION OF THE DRAWINGS The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the accompanying drawings: Figure 1 is a schematic diagram of an application scenario of a processing and generation method for an intelligent agent according to an embodiment of the present disclosure; Figure 2 is a flow chart of a processing method for an intelligent agent according to an embodiment of the present disclosure; Figure 3 is a flow chart of a processing method for another intelligent agent according to an embodiment of the present disclosure; Figure 4 is a flow chart of a processing method for another intelligent agent according to an embodiment of the present disclosure; Figure 5 is a flow chart of a processing method for another intelligent agent according to an embodiment of the present disclosure; Figure 6 is a flow chart of a processing method for another intelligent agent according to an embodiment of the present disclosure; Figure 7 is a schematic diagram of a matrixed semantic map line of a high-definition semantic map according to an embodiment of the present disclosure; Figure 8 is a schematic diagram of collaborative generation of map-related semantic features and multi-agent simulation trajectories according to an embodiment of the present disclosure; Figure 9 is a schematic diagram of a processing device for an intelligent agent according to an embodiment of the present disclosure; Figure 10 is a schematic diagram of a processing device for another intelligent agent according to an embodiment of the present disclosure; Figure 11 is a schematic diagram of a processing device for another intelligent agent according to an embodiment of the present disclosure; Figure 12 is a schematic diagram of a processing device for another intelligent agent according to an embodiment of the present disclosure; Figure 13 is a structural block diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION To help those skilled in the art better understand the present disclosure, the following will provide a clear and complete description of the technical solutions in the embodiments of the present disclosure, in conjunction with the accompanying drawings. It should be noted that the described embodiments represent only a portion of the embodiments of the present disclosure, and are not exhaustive. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present disclosure without inventive effort should fall within the scope of protection of the present disclosure. It should be noted that the terms "first," "second," and so on, in the specification and claims of the present disclosure, and in the accompanying drawings, are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that such terms are interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than that illustrated or described herein. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or components is not necessarily limited to the steps or components expressly listed, but may include other steps or components not expressly listed or inherent to such process, method, product, or apparatus.The technical solutions provided herein are primarily implemented using large-scale model technology. A large-scale model here refers to a deep learning model with large-scale model parameters, typically including hundreds of millions, tens of billions, hundreds of billions, trillions, or even more than ten trillion model parameters. Large-scale models, also known as foundation models, are pre-trained using large-scale unlabeled corpora to produce pre-trained models with over 100 million parameters. Such models are adaptable to a wide range of downstream tasks and exhibit good generalization capabilities. Examples include large language models (LLMs) and multi-modal pre-training models. It should be noted that in practical applications, large-scale models can be fine-tuned using a small number of samples, allowing them to be applied to different tasks. For example, large models can be widely applied in fields such as natural language processing (NLP), computer vision, and speech processing. Specifically, they can be applied to computer vision tasks such as visual question answering (VQA), image captioning (IC), and image generation. They can also be widely applied to natural language processing tasks such as text-based sentiment classification, text summarization, and machine translation. Therefore, the main application scenarios of large models include, but are not limited to, digital assistants, intelligent robots, search, online education, office software, e-commerce, and intelligent design. In the embodiments of this disclosure, data processing using the Contrastive Language-Image Pre-Training (CLIP) model in a smart city scenario is used as an example for explanation. First, some nouns or terms used in the description of the embodiments of this disclosure are subject to the following interpretation:

[0002] The CLIP model, a text-image correlation matching model, is used to understand and process the semantic relationship between text and images; an agent, an entity capable of making autonomous decisions, such as a pedestrian, motor vehicle, or non-motor vehicle; and simulation rule information, information about the behavioral rules that the agent must follow when operating within an area. According to an embodiment of the present disclosure, a method for processing an agent is provided. It should be noted that the steps shown in the flowcharts of the accompanying figures can be executed in a computer system, such as a set of computer-executable instructions. Furthermore, while the flowcharts illustrate a logical order, in some cases, the steps shown or described may be executed in a different order. Given the large number of model parameters in large models and the limited computing resources of mobile terminals, the agent processing method provided in the embodiments of the present disclosure can be applied to the application scenario shown in Figure 1, but is not limited thereto. In the application scenario shown in Figure 1, the large model is deployed on a server 10. Server 10 can connect to one or more client devices 20 via a local area network (LAN), a wide area network (WAN), the Internet, or other types of data networks. Client devices 20 herein include, but are not limited to, smartphones, tablets, laptops, PDAs, personal computers, smart home devices, and in-vehicle devices. Client devices 20 can interact with users via a graphical user interface to invoke the large model and thereby implement the methods provided in the embodiments of the present disclosure. In the embodiments of the present disclosure, a system comprising client devices and a server can perform the following steps: the client device sends a simulation request for an agent within a region to be monitored, and the server determines the region to be monitored; obtains map information for the region and at least one agent located within the region; obtains simulation information associated with the map information and the at least one agent, wherein the simulation information represents historical behavior information of the agent within the region and / or simulation rules used when simulating the agent within the region; and uses the simulation information to simulate the behavior of the at least one agent within the region to obtain at least one simulation result. It should be noted that, if the client device's operating resources meet the requirements for deploying and operating a large model, the embodiments of the present disclosure can be implemented on the client device. Within the aforementioned operating environment, the present disclosure provides an agent processing method as shown in Figure 2. Figure 2 is a flow chart of an agent processing method according to an embodiment of the present disclosure. As shown in Figure 2, the method may include the following steps: Step S201: Determine an area to be monitored. In the technical solution provided in step S201 of the present disclosure, the area to be monitored may be an urban area. The area to be monitored may be determined when a simulation request is received from an agent within the area to be monitored.An agent is an entity capable of autonomous action and decision-making. For example, it can be a pedestrian, motor vehicle, or non-motor vehicle within the area to be monitored. For example, when a simulation request for an agent within a city is received, the city area can be identified as the area to be monitored. Step S202: Obtain map information for the area and at least one agent located within the area. In the technical solution provided in step S202 of the present disclosure, after determining the area to be monitored according to step S201, further map information for the area and at least one agent located within the area can be obtained. The map information for the area can include a global map, a physical map, a global semantic map, a high-definition semantic map, or other information corresponding to the area, without specific limitation. The map information for the area can include information such as the area's geographic location, topography, roads, building distribution, and natural resource distribution. In this embodiment, semantic segmentation is performed on camera perception data within the area and the corresponding surveying and mapping modeling data using a general semantic segmentation network (SAM). Cameras within an area can collect historical state information for at least one agent within the area. This historical state information can include the agent's location, size, speed, and type. For example, assuming the area is a city, video data can be collected using city-level image acquisition equipment, and corresponding mapping and modeling data can be obtained. The city-level image acquisition equipment can be a city-level camera. A SAM semantic segmentation network is then used to identify objects and regions in the video data and mapping and modeling data, segmenting them into semantically meaningful components to generate map information corresponding to the city. Since cities contain agents, such as pedestrians, motor vehicles, and non-motor vehicles, historical state information for at least one agent can be obtained based on the video data collected by the city-level camera. Alternatively, the map information can be viewed as a map composed of multiple polygons with semantic attributes. The map information can be vectorized into an N*M*d matrix, where N represents the number of polygons, M represents the maximum number of edges, and d represents certain characteristics of the polygon edges, such as their position, length, direction, and semantic attributes. Step S203: Acquire simulation information associated with the map information and at least one agent.In the technical solution provided in step S203 of the present disclosure, after obtaining map information for a region and at least one agent located within the region in step S202, simulation information associated with the map information and the at least one agent can be further obtained. The simulation information represents the agent's historical behavior within the region corresponding to the map information and / or the simulation rules used by the agent during simulation within the region. In this embodiment, as previously described, the map information can indicate information such as the region's geographic location, topography, transportation routes, building distribution, and natural resource distribution, while the simulation rule information can indicate information used by the agent during simulation within the region, such as the safety and traffic regulations that the agent must follow when navigating within the region. In other words, the map information and simulation rule information are both for the same region. For example, the agent's historical behavior information within the region indicates the agent's historical trajectory and behavior within the region corresponding to the map information. This information can be obtained using video data from cameras within the region. The simulation rule information used by the agent during simulation within the area indicates the behavioral rules that the agent must follow when operating within the area corresponding to the map information. For example, these include the safety and traffic rules that autonomous entities such as pedestrians, motor vehicles, and non-motor vehicles must follow when operating on roads. The simulation rule information used by the agent during simulation within the area corresponding to the map information can be obtained from road traffic guidelines published by traffic management departments. This is merely an example and does not limit the specific method for obtaining the simulation rule information. In step S204, the behavior of at least one agent within the area is simulated using the simulation information to obtain at least one simulation result. In the technical solution provided in step S204 of the present disclosure, after obtaining the simulation information associated with the map information and at least one agent in step S203, the behavior of the at least one agent within the area can be simulated using the simulation information to obtain at least one simulation result. This simulation result can at least indicate a simulated trajectory of the agent within the area, which is used to simulate the dynamic behavior of the agent within the area. In this embodiment, after obtaining the simulation information, the simulation information associated with the map information and at least one intelligent agent can be input into a simulation generation component. The simulation generation component can simulate the behavior and interactions of the at least one intelligent agent in the area based on the historical behavior information of the intelligent agent in the area and / or the simulation rule information used when simulating the intelligent agent in the area, to obtain at least one simulation result.For example, the simulation generation component includes a deep multi-agent behavior prediction network. By analyzing historical behavior information of agents within a region and / or simulation rules used by agents during simulation within the region, the deep multi-agent behavior prediction network can derive the behavioral habits of at least one agent within the region and the behavioral rules that at least one agent must follow when operating within the region. When the agents include pedestrians, vehicles, and non-motorized vehicles, simulations can be performed on pedestrian compliance with traffic rules, traffic flow between vehicles, and shared space utilization between non-motorized vehicles and pedestrians. Simulation results can be generated to help urban planners and decision makers better understand traffic flow and crowd behavior within cities, thereby optimizing urban planning and management. Based on steps S201 to S204 of the above embodiment, historical behavior information of at least one agent within the area and simulation rule information for simulating at least one agent within the area are obtained based on map information of the area to be monitored. Since the historical behavior information of the agents within the area can be used to indicate the behavioral habits of the agents, the simulation rule information for simulating the agents within the area can be used to indicate the behavioral rules that the agents must comply with within the area. Therefore, by simulating the behavior of the agents within the area using the historical behavior information of the agents within the area and the simulation rule information for simulating the agents within the area, the simulation results can be strongly correlated with the historical behavior information and simulation rule information of the agents, thereby improving the accuracy of the generated simulation results for the agents. Furthermore, multi-agent simulation results with controllable behaviors and solutions are obtained, thus resolving the technical issue of being unable to effectively simulate the behavior of the agents. The above-mentioned method of this embodiment is further described below. As an optional implementation, step S203, obtaining simulation information associated with map information and at least one agent, includes: determining first noise information based on the map information and the at least one agent, where the first noise information indicates the type of simulation result; and determining simulation information based on the first noise information. In this embodiment, the first noise information may be Gaussian noise, which can be used to learn about simulation diversity and simulate noise and uncertainty in a real environment. Therefore, after determining the map information and the at least one agent, the first noise information may be determined, and then the simulation information may be determined based on the first noise information. For example, the first noise information may be Gaussian noise, and the simulation information may be determined based on the first noise information from the map information and the historical state information of the at least one agent.As an optional implementation, determining first noise information based on map information and at least one agent includes: obtaining historical state information of the at least one agent within the area; and determining the first noise information based on the amount of information in the historical state information and the amount of information in the map information. In this embodiment, the historical state information of the at least one agent within the area can be used to indicate the state of the at least one agent at a historical moment. The amount of information in this historical state information may include the location, size, speed, and type of the at least one agent at that historical moment. This is merely an example and does not limit the amount of information in the historical state information. For example, the historical state information of the at least one agent within the area can be obtained using video data captured by a camera within the area. For example, the location, size, speed, and type of the at least one agent at that historical moment can be obtained using video data captured by the camera. Alternatively, as can be seen from the foregoing description, the map can be a global semantic map corresponding to a region. This global semantic map can be considered a map composed of multiple polygons with semantic attributes and can be vectorized into an N*M*d matrix, where N represents the number of polygons, M represents the maximum number of edges in a polygon, and d represents some characteristics of the polygon edges. The information content of the map information can include the information represented by each edge of the vectorized matrix. As an optional implementation, determining the first noise information based on the information content of the historical state information and the information content of the map information includes: converting the historical state information into a first matrix and converting the map information into a second matrix, wherein the information content of the first and second matrices in the same direction has the same dimension; and obtaining first noise information that matches the dimensions of the first and second matrices. In this embodiment, when determining the first noise information based on the information content of the historical state information and the information content of the map information, the historical state information can be converted into the first matrix and the map information can be converted into the second matrix. Thereafter, obtaining the first noise information that matches the dimensions of the first and second matrices. For example, assuming the number of agents is K, the historical state information is used to indicate the state information of K agents at a historical time T. The state information of each agent includes dl elements, such as the position of the agent, the size of the agent, the speed of the agent, and the type of the agent.In this case, the dl pieces of historical state information for K agents at time T can be vectorized into a first matrix, where this first matrix can be expressed as a K*T*dl matrix, denoted as matrix QL. Alternatively, since map information can be global semantic map information, which can be viewed as a map of multiple polygons with semantic attributes, this can be vectorized into a second matrix. This second matrix can be expressed as an N*M*d matrix, denoted as matrix Q2, where N represents the number of polygons, M represents the maximum number of edges in a polygon, and d represents some characteristics of the polygon edges. Optionally, after obtaining the first and second matrices, a structure similar to point cloud data mapping (Point Transformer) can be used to map the first and second matrices into a K*C-dimensional matrix Q1 and an N*C-dimensional matrix Q2L, respectively. The information content in the same direction in matrix Q1 and the N*C-dimensional matrix Q2 belongs to the same dimension, which can be C. Optionally, after obtaining the first matrix and the second matrix, you can obtain the first matrix and the second matrix. As an optional implementation, determining simulation information based on the first noise information includes: utilizing the first noise information to determine first-type features from a first matrix and / or second-type features from a second matrix, wherein the first-type features represent characteristics of at least one agent, and the second-type features represent characteristics of map information; and determining the simulation information based on the first-type features and / or the second-type features. In this embodiment, after determining the first noise information, the first-type features can be determined from the first matrix using the first noise information, wherein the first-type features can be depth features related to the agent. Similarly, the first-type features can be determined from the second matrix using the first noise information, wherein the second-type features can be depth features related to the map. For example, the first matrix, the second matrix, and the first noise information can be input into a Transformer encoder. After calculations through multiple self-attention layers, a (1+N'+K')*C-dimensional matrix is ​​obtained, where K represents the first-type features determined in the first matrix, i.e., deep features related to the agent, and N represents the second-type features determined in the second matrix, i.e., deep features related to the map. Optionally, after determining the first-type features and / or the second-type features, simulation information can be determined based on the first-type features and / or the second-type features. As an optional implementation, determining the simulation information based on the first-type features and / or the second-type features includes: aligning the first-type features with historical behavior information in the historical state information to generate simulation information corresponding to the historical behavior information; and / or aligning the second-type features with description information corresponding to the simulation rule information to generate simulation information corresponding to the simulation rule information, where the description information is used to describe the simulation rule information. In this embodiment, since the historical behavior information in the historical state information represents the agent's behavior at historical moments, the historical behavior information in the historical state information can be used as a reference to align the first-type features with the historical behavior information in the historical state information to improve the accuracy of the agent's behavior simulation. Alternatively, since the simulation rule information indicates the behavioral rules that the agent must follow when operating within a region, the descriptive information corresponding to the simulation rule information can be text describing the behavioral rules that the agent must follow within the region, such as "turn left at the intersection" or "go straight at the intersection." The second-type features indicate map-related depth features. Since the map corresponds to the region, the second-type features indicate graph features related to the map corresponding to the region.Simulation rule information is textual information describing the behavioral rules that an agent must follow when operating within a region. In other words, simulation rule information is a textual feature. Based on this, the second-type features and the simulation rule information can be input into a graph-text correlation matching model for feature alignment. Here, gradient descent can also be used to optimize noise, minimizing the distance between the second-type features and the simulation rule information in semantic space, thereby improving the accuracy of the simulation information corresponding to the generated simulation rule information. Optionally, after generating at least one simulation result, to ensure that simulation results that meet expectations are prioritized, a target simulation result that matches the actual situation within the region can be determined from the at least one simulation result. The process of determining the target simulation result that matches the actual situation within the region from the at least one simulation result will be further described below. As an optional implementation, the agent processing method further includes: obtaining a distance loss function corresponding to the description information in a semantic space; determining second noise information for the region based on the description information and the distance loss function, wherein the second noise information is used to ensure that the distance between the semantics of the description information and the semantics of the map information of the region in the semantic space is less than a distance threshold; and determining at least one target simulation result that matches the description information in at least one simulation result based on the second noise information. In this embodiment, the description information is used to describe the congestion conditions of road sections and intersections within the region, as well as the state information of the agent, and the semantic space may be a CLIP semantic space. After determining the description information of the road sections and intersections within the region, a corresponding distance loss function in the semantic space may be obtained based on the description information, and then determining second noise information for the region based on the description information and the distance loss function. The second noise information is used to ensure that the distance between the semantics of the description information and the semantics of the map information of the region in the semantic space is less than a distance threshold, that is, to ensure that the distance between the semantic attributes of the description information and the semantic attributes of the map information of the region in the CLIP semantic space is less than the distance threshold. For example, the second noise information for the region can be determined using a gradient descent method by combining the high-definition semantic map corresponding to the description information, the multi-agent state information, and all network parameters used in determining the simulation results of the agents. Optionally, after obtaining the second noise information, at least one target simulation result matching the description information can be determined based on the second noise information in at least one simulation result. The target simulation result indicates a preferred trajectory among the simulation trajectories of the at least one agent.As an optional implementation, determining at least one target simulation result that matches the description information from at least one simulation result based on the second noise information includes: sorting the at least one simulation result according to the description information; selecting at least one simulation result with a higher ranking from the sorted at least one simulation result, wherein a higher ranking simulation result has a greater correlation with the description information; and determining the target simulation result that matches the description information from the at least one simulation result with a higher ranking based on the second noise information. In this embodiment, when determining at least one target simulation result that matches the description information from at least one simulation result based on the second noise information, to ensure that simulation results that meet expectations are preferentially selected, the at least one simulation result may be sorted by solution relevance using an algorithm based on the description information. Since higher ranking simulation results have a greater correlation with the description information, after sorting the at least one simulation result with a higher ranking, the at least one simulation result with a higher ranking may be selected, and then, based on the second noise information, the target simulation result that matches the description information may be determined from the selected at least one simulation result. Optionally, when sorting the at least one simulation result, the at least one simulation result may be sorted by rationality to obtain a ranking result of the simulation results. The at least one simulation result with a higher ranking is then selected, and based on the second noise information, a target simulation result that matches the description information is determined from the at least one simulation result with a higher ranking. As an optional implementation, the intelligent agent processing method further includes: determining a simulation index for the target simulation result, where the simulation index indicates the reliability of the target simulation result; and sending target simulation results whose simulation index exceeds an index threshold to an information monitoring platform. In this embodiment, after determining the at least one target simulation result, the target simulation result may be further subjected to automated and / or manual multiple simulation evaluations to determine the simulation index of the target simulation result. The simulation index of the at least one target simulation result is then compared with the index threshold, and the target simulation result whose simulation index exceeds the index threshold is sent to the information monitoring platform, which is used for application-level management and control of downstream smart cities. Optionally, after receiving the target simulation result, the information monitoring platform may construct a smart city based on the target simulation result.As an optional implementation, determining first-type features from a first matrix and second-type features from a second matrix using first noise information includes: invoking a first encoder in a simulation model to perform attention calculations on the first noise information, the first matrix, and the second matrix to obtain first-type features and second-type features. The simulation model is obtained by training a neural network model and is used to simulate the behavior of at least one agent within a region using simulation information. In this embodiment, the simulation model includes a first encoder, which may be a Transformer encoder and capable of performing attention calculations. Based on this, when determining first-type features from the first matrix and second-type features from the second matrix using the first noise information, the first encoder in the simulation model may be invoked, and the first noise information, the first matrix, and the second matrix may be input to the first encoder to perform multi-layer attention calculations to obtain first-type features and second-type features. For example, as described above, the first matrix can be matrix Q1', the second matrix can be matrix Q2', and the first noise information can be Gaussian noise of length 1*C. Based on this, matrix Q1' and matrix Q2' can be input to the first encoder. After multi-layer attention calculations, a matrix of (1+N'+K')*C dimensions is obtained, where K' can be used to indicate the first type of features, and N' can be used to indicate the second type of features. As an optional implementation, converting historical state information into the first matrix and converting map information into the second matrix includes: invoking a second encoder in the simulation model to map the historical state information into the first matrix; and invoking a third encoder in the simulation model to map the map information into the second matrix. In this embodiment, the second and third encoders in the simulation model can be PointTransformer encoders, configured to map matrices into matrices of target dimensions. Therefore, when converting the historical state information into the first matrix, the second encoder in the simulation model can be invoked to map the historical state information into the first matrix. Similarly, when converting the map information into the second matrix, the third encoder in the simulation model can be called to map the map information into the second matrix. The first matrix and the second matrix have the same dimensions in the same direction. As an optional implementation, obtaining map information for a region includes: collecting image information for the region; performing semantic segmentation on the image information; and calibrating the segmented image information using road information in the region to obtain map information.In this embodiment, when obtaining map information for a region, cameras within the region can capture video data of the region and obtain surveying and mapping data for the region. The video data and surveying and mapping data are then used as image information for the region. Semantic segmentation is then performed on the image information using a semantic segmentation network to obtain segmented image information. Subsequently, the segmented image information is calibrated using road information in the region, such as lane markings, road signs, and traffic lights, to obtain map information. In the above steps, the simulation of the agent's behavior within the region is performed using the agent's historical behavior information within the region and the simulation rules for simulating the agent within the region. This ensures that the simulation results are strongly correlated with the agent's historical behavior information and simulation rules, improving the accuracy of the agent's simulation results. After obtaining at least one simulation result, the simulation results can be evaluated to select those that meet expected conditions, providing data reference for the construction of smart cities and addressing the technical issue of low accuracy of generated simulation results. Within the aforementioned operating environment, the present disclosure also provides an agent processing method as shown in Figure 3. Figure 3 is a flowchart of another agent processing method according to an embodiment of the present disclosure. As shown in Figure 3, the method may include the following steps: Step S301: Determine an urban area to be monitored. In the technical solution provided in step S301 of the present disclosure, the urban area to be monitored may be determined in response to a simulation request from an agent within the urban area to be monitored. The agents may be pedestrians, motor vehicles, non-motor vehicles, or other traffic participants within the urban area to be monitored. Step S302: Obtain map information of the urban area and at least one agent within the urban area. In the technical solution provided in step S302 of the present disclosure, after determining the urban area to be monitored in step S301, further map information of the urban area and at least one agent within the urban area may be obtained. The map information of the urban area may include a global map, a physical map, a global semantic map, a high-definition semantic map, etc. of the urban area, without specific limitation herein. The map information of the urban area may include information such as the geographical location, topography, roads, building distribution, and natural resource distribution of the area. At least one intelligent agent is used to indicate an entity within the urban area that can act autonomously and make decisions. For example, it can be a pedestrian, a motor vehicle, a non-motor vehicle, etc., which is not specifically limited here. Step S303: Simulation information associated with the map information and the at least one intelligent agent is obtained.In the technical solution provided in step S303 of the present disclosure, after determining the map information of the urban area and at least one agent located within the urban area in step S302, simulation information associated with the map information and the at least one agent can be further obtained. The simulation information represents the agent's historical behavior information within the urban area corresponding to the map information and / or the simulation rules used by the agent during simulation within the urban area. In this embodiment, the agent's historical behavior information within the urban area indicates the agent's historical trajectory and behavior within the urban area. The historical trajectory and behavior of the at least one agent within the urban area can be obtained using camera video data within the urban area. The simulation rules used by the agent during simulation within the urban area indicate the behavioral rules that the agent must follow when operating within the urban area corresponding to the map information. For example, these rules include the safety and traffic regulations that autonomous entities such as pedestrians, motor vehicles, and non-motor vehicles must follow when operating on roads. The simulation rule information used by the agent during simulation within the area can be obtained from the road traffic guide published by the traffic management department. This is merely an example and does not limit the specific method for obtaining the simulation rule information. Optionally, when obtaining simulation information associated with map information and at least one agent, first noise information can be determined based on the map information and the at least one agent. The first noise information indicates the type of simulation result, and the simulation information is determined based on the first noise information. For example, the first noise information can be Gaussian noise, which can be used to learn the diversity of the simulation and simulate the noise and uncertainty in a real environment. Based on this, after determining the map information and the at least one agent, the first noise information can be determined, and then the simulation information is determined based on the first noise information. In step S304, the traffic behavior of the at least one agent within the urban area is simulated using the simulation information to obtain at least one simulation result. In the technical solution provided in step S304 of the present disclosure, after obtaining simulation information associated with the map information and at least one agent in step S303, the simulation information can be used to simulate the behavior of the at least one agent within the urban area, thereby obtaining at least one simulation result. The simulation result can at least indicate a simulated trajectory of the agent within the urban area, and the simulation trajectory can be used to simulate the dynamic behavior of the agent within the urban area.In this embodiment, after obtaining simulation information, the simulation information associated with the map information and at least one agent can be input into a simulation generation component. Based on the historical behavior information of the agent within the area and / or the simulation rules used by the agent during the simulation within the city area, the simulation generation component can simulate the behavior and interactions of the at least one agent within the city area, generating at least one simulation result. In step S305, the simulation result is transmitted to the city information monitoring platform. In the technical solution provided in step S305 of the present disclosure, after obtaining at least one simulation result according to step S304, the simulation result can be transmitted to the city information monitoring platform to assist the city information monitoring platform in creating a smart city. The city information monitoring platform is used for downstream application-level management and control of the smart city. In the technical solution provided in steps S301 to S305 of the present disclosure, historical behavior information of at least one agent within the urban area and simulation rule information for simulating at least one agent within the urban area are obtained based on map information of the urban area to be monitored. Since the historical behavior information of the agent within the urban area can be used to indicate the agent's behavioral habits, and the simulation rule information for simulating the agent within the urban area can be used to indicate the behavioral rules that the agent must comply with within the urban area, simulating the agent's behavior within the urban area using the historical behavior information of the agent within the urban area and the simulation rule information for simulating the agent within the urban area can ensure a strong correlation between the simulation results and the agent's historical behavior information and simulation rule information, thereby improving the accuracy of the agent's simulation results and obtaining multi-agent simulation results with controllable behaviors and plans, thus resolving the technical problem of being unable to effectively simulate the agent's behavior. Within the aforementioned operating environment, the present disclosure also provides an agent processing method as shown in Figure 4. Figure 4 is a flow chart of another agent processing method according to an embodiment of the present disclosure. As shown in Figure 4, the method may include the following steps: Step S401: In response to an input instruction applied to an operation interface, displaying an area to be monitored on the operation interface. In the technical solution provided in step S401 of the present disclosure, the operation interface includes an input box in which a user can enter an instruction. The input instruction may be a simulation request for an agent within the area to be monitored. In response to the input operation in the input box, i.e., in response to the input instruction applied to the operation interface, the area to be detected may be retrieved from a map based on the identification information of the area to be monitored included in the input instruction, and the area to be monitored may be displayed on the operation interface.Step S402: In response to a simulation instruction applied to the operation interface, at least one simulation result of at least one agent within the area is displayed on the operation interface. In the technical solution provided in step S402 of the present disclosure, the operation interface includes a simulation control that is used to trigger the simulation instruction. In response to a selection operation of the simulation control on the operation interface, i.e., in response to the simulation instruction applied to the operation interface, the simulation instruction is used to display at least one simulation result of at least one agent within the area on the operation interface. The simulation result is obtained by simulating the behavior of at least one agent within the area using simulation information. The simulation information is associated with the map information of the area and the at least one agent, and is used to represent the historical behavior information of the agent within the area corresponding to the map information and / or the simulation rule information used when simulating the agent within the area. The process of simulating the behavior of at least one agent within the area using the simulation information can be referred to the description of steps S202 to S204 above and will not be repeated here. In steps S401 and S402 above, in response to input commands on the operation interface, the area to be monitored is displayed on the operation interface. In response to simulation commands on the operation interface, at least one simulation result of at least one agent within the area is displayed on the operation interface. The simulation result is obtained by simulating the behavior of at least one agent within the area using simulation information. The simulation information is associated with the map information of the area and the at least one agent, and represents the agent's historical behavior within the area corresponding to the map information and / or the simulation rules used when simulating the agent within the area. In other words, in embodiments of the present disclosure, in response to input commands and simulation commands, the simulation results of the area to be monitored and the agent can be displayed on the operation interface. This helps users promptly understand the simulation results of the monitored area and the agent within the area, allowing them to intuitively determine whether the agent's simulation results meet expectations, thereby improving the user experience. Within the aforementioned operating environment, the present disclosure also provides an agent processing method as shown in Figure 5. FIG5 is a flowchart of another agent processing method according to an embodiment of the present disclosure. As shown in FIG5 , the method may include the following steps: Step S501: Obtaining map information of the area to be monitored by invoking a first interface. In the technical solution provided in step S501 of the present disclosure, the first interface may be an interface for data exchange between a server and a client. After the client generates a simulation request for an agent within the area to be monitored, the server may obtain the simulation request from the agent within the area to be monitored by invoking the first interface, wherein the simulation request includes the map information of the area to be monitored.Step S502: Simulation information associated with the map information and at least one agent located within the area is obtained. In the technical solution provided in step S502 above, the map information can be a global map, physical map, global semantic map, high-definition semantic map, etc. corresponding to the area, without specific limitation. The map information for the area may include information such as the area's geographic location, topography, transportation routes, building distribution, and natural resource distribution. The at least one agent represents an entity capable of autonomous action and decision-making, such as a pedestrian, motor vehicle, or non-motor vehicle, without specific limitation. In this embodiment, after obtaining the map information for the area and at least one agent located within the area, simulation information associated with the map information and the at least one agent can be further obtained. The simulation information represents the agent's historical behavior within the area corresponding to the map information and / or the simulation rules used by the agent when performing simulation within the area. The steps of obtaining simulation information associated with the map information and at least one agent within the area can be referred to above in steps S202 to S203 and will not be further described here. In step S503, the behavior of the at least one agent within the area is simulated using the simulation information to obtain at least one simulation result. In the technical solution provided in step S503 of the present disclosure, after obtaining the simulation information associated with the map information and at least one agent in step S502, the behavior of the at least one agent within the area can be simulated using the simulation information to obtain at least one simulation result. The simulation result can at least indicate a simulation trajectory of the agent within the area, which is used to simulate the dynamic behavior of the agent within the area. The process of simulating the behavior of the at least one agent within the area using the simulation information can be referred to above in the description of step S204 and will not be further described here. In step S504, the simulation result is output by invoking the second interface. In the technical solution provided in step S504 of the present disclosure, the second interface may be an interface for data interaction between the server and the client. After obtaining at least one simulation result, the server may output the simulation result to the client by calling the second interface.Based on steps S501 to S504 of the above embodiment, historical behavior information of at least one agent within the area and simulation rule information for simulating at least one agent within the area are obtained based on map information of the area to be monitored. Since the historical behavior information of the agent within the area can be used to indicate the agent's behavioral habits, the simulation rule information for simulating the agent within the area can be used to indicate the behavioral rules that the agent must comply with when simulating the agent within the area. Therefore, by utilizing the historical behavior information of the agent within the area and the simulation rule information for simulating the agent within the area, the behavior of the agent within the area is simulated. This ensures that the simulation results are strongly correlated with the agent's historical behavior information and simulation rule information, improving the accuracy of the agent simulation results. This results in a multi-agent simulation result with controllable behavior and solutions, resolving the technical issue of effectively simulating agent behavior. The following further illustrates the technical solutions of the disclosed embodiments with reference to preferred embodiments, specifically using smart city application scenarios as an example. Currently, with the accelerated urbanization process, the scale of cities is rapidly expanding, and the complexity of urban management and planning is also increasing. The concept of smart cities has emerged. Smart cities aim to leverage information and communication technologies to optimize urban functions, enhance the efficiency of urban operations, improve residents' quality of life, and achieve sustainable development. Digital twin technology has become a crucial tool in the development of smart cities. By creating a virtual city model, it can reflect and simulate various dynamic urban processes in real time. Multi-agent simulation, a component of digital twin technology, involves simulating the behavior and interactions of various agents in a city (e.g., pedestrians, motor vehicles, and non-motor vehicles). Agents can be people, vehicles, or other entities capable of autonomous action and decision-making. In multi-agent simulation, each agent follows specific behavioral rules, and their interactions form a complex system that represents the entire dynamic city. In urban planning and traffic management, multi-agent simulation can simulate the interactions among different traffic participants and provide solutions for optimizing traffic flow, reducing congestion, and improving air quality. In urban facility design, multi-agent simulation can assess the impact of new facilities on the urban environment, including their impact on pedestrians and traffic flow, helping to design more humane and efficient urban spaces. In one implementation, multi-agent collaborative simulation can be achieved to a certain extent based on multi-agent behavior prediction neural networks and trajectory prediction neural networks. However, this mainly targets short-term behavior simulation for autonomous driving, that is, predicting the behavior and trajectory of vehicles in urban road traffic. It cannot provide multi-agent behavior simulation results and cannot be applied in smart city construction.However, the present disclosure provides a multi-agent simulation generation method. By building a comprehensive solution for city-level controllable multi-agent simulation generation, encompassing global map perception, multi-agent collaborative simulation generation, controllable simulation optimization, and simulation evaluation, this method provides a comprehensive solution for city-level controllable multi-agent simulation generation. First, city-level camera perception data is used to perform automated or semi-automated global precision semantic calibration, providing physical map information for simulation generation. Second, the simulation generation component utilizes a deep multi-agent behavior prediction network to receive historical trajectories or simulation plans of multiple agents (e.g., pedestrians, vehicles, and non-motorized vehicles) and generate multi-agent simulation results. To ensure that expected simulation results are prioritized among the numerous generated simulation results, simulation results can be ranked according to various expectations. After the algorithm automatically generates recommended simulation results, the simulation results undergo multiple automated and manual evaluations, generating multiple evaluation opinions for downstream smart city application-level management and control. That is, in the present disclosure, after generating numerous simulation results, the simulation results can be sorted according to expectations to select simulation results that meet the requirements, providing a reference for smart city management. This achieves the technical effect of accurately predicting the behavior of intelligent agents on urban traffic roads, thereby resolving the technical problem of being unable to accurately predict the behavior of intelligent agents on urban traffic roads. Next, the multi-agent simulation generation method according to an embodiment of the present disclosure is further described. Figure 5 is a schematic diagram of a multi-agent simulation generation process according to an embodiment of the present disclosure. As shown in Figure 5, the multi-agent simulation generation process primarily includes the following steps: city-level map construction, simulation generation, solution optimization, and simulation evaluation. During city-level map construction, global traffic perception and precise semantic calibration are performed on the entire city map to produce a high-definition semantic map. During simulation generation, simulation trajectories are primarily generated by perceiving the historical trajectories of intelligent agents such as pedestrians, motor vehicles, and non-motor vehicles. After the simulation is generated, the solution optimization phase can begin. During this optimization phase, trajectories that meet certain criteria can be selected from the generated simulation trajectories based on preset simulation trajectory selection principles. These selection principles may include prioritizing clear roads, minimizing accidents, and prioritizing solution relevance. These are merely illustrative examples and do not limit the selection principles. After selecting a trajectory that meets certain criteria, a simulation evaluation can be performed on the selected trajectory. For example, the spatiotemporal and behavioral consistency of the selected trajectory can be evaluated through multi-perspective automatic evaluation and manual feedback. Figure 6 is a flowchart of an intelligent agent processing method according to an embodiment of the present disclosure. The intelligent agent processing method includes the following steps: Step S601: Obtain a global semantic map of a city-level map.In this embodiment, a SAM semantic segmentation network can be used to perform global semantic segmentation on city-level camera video data and surveying and modeling data. Combined with manual lane line calibration, global map information can be generated as one of the inputs to the simulation generation model. Alternatively, the global semantic map can be viewed as a map composed of multiple polygons with semantic attributes. This can be vectorized into an N*M*d matrix, which corresponds to the semantic map lines in Figure 6. N represents the number of polygons, M represents the maximum number of edges per polygon, and d represents certain characteristics of the polygon edges (including line position, length, direction, and semantic attributes). For example, Figure 7 is a schematic diagram of a high-definition semantic map matrixed into semantic map lines according to an embodiment of the present disclosure. Vectorizing the high-definition semantic map yields a semantic map line represented by the N*M*d matrix. Step S602: Obtain the historical states of multiple agents in the global semantic map and generate simulation trajectories for the multiple agents. In this embodiment, the historical states of multiple agents (e.g., pedestrians, motor vehicles, and non-motor vehicles) on urban traffic roads at a historical time T can be obtained and vectorized into a K*T*dl matrix. dl contains the multiple historical states of the multiple agents, including the agent's position, size, speed, and type. The multiple historical states can be denoted as the matrix QL. Alternatively, FIG8 is a schematic diagram of collaboratively generating map-related semantic features and multi-agent simulation trajectories according to an embodiment of the present disclosure. As shown in FIG8, a structure similar to the Point Transformer is used to map the matrix Q1 and the semantic map line matrix Q2 in step S501 to a K*C dimensional matrix Q1 and an N*C dimensional matrix Q2L, respectively. The matrix Q1 and the matrix Q2 are then fed into a Transformer encoder along with a Gaussian noise Z of length 1*C. After computations are performed in multiple self-attention layers, the (1+N,+K?) matrix Q1 is obtained. *C-dimensional matrix, where N corresponds to map-related deep features and K corresponds to agent-related deep features. Optionally, during the training phase, the map deep features can be aligned with some descriptive text in the real data (for example, whether a road is congested, whether a certain agent is present within a road section, etc.) in the CLIP semantic space. In other words, the trajectory data can be aligned with the real historical trajectories of multiple agents. Step S603: Select trajectories that meet the requirements from the simulated trajectories of the multiple agents.In this embodiment, for a given urban road condition, such as congested traffic and text descriptions of certain road sections / intersections (e.g., congestion at the road section / intersection or a large number of vehicles making left turns at the road section / intersection), the corresponding noise z is first decalculated based on the text description and the distance loss function in the CLIP semantic space. For example, the high-definition map, multi-agent states, and network parameters of the current model input into the distance loss function in the CLIP semantic space can be fixed (or frozen). Gradient descent is then used to optimize the noise z to minimize the distance between the text description and the map semantics in the CLIP space. Optionally, after decalculating the noise z, it can be used as input to obtain a generative trajectory generated by simulation. This trajectory is then the selected trajectory. Step S604 evaluates the authenticity of the selected trajectory. In this embodiment, after selecting a trajectory that meets the requirements, certain authenticity indicators of the selected trajectory can be automatically evaluated. Authenticity indicators may include whether there is a collision, whether the agent's movement involves significant turns, and whether the agent interacts unreasonably with the surrounding semantic map. The distances shown here are merely exemplary and do not limit the specific content of the authenticity indicators. Optionally, after evaluating the authenticity indicators of the selected trajectory, these indicators can be provided to downstream digital twin applications, enabling more efficient management and construction of smart cities. In steps S601 through S604, the automated process of semantic map construction, simulation generation, simulation optimization, and simulation evaluation provides automated, controllable multi-agent simulation capabilities for smart city construction in the digital twin field. This significantly reduces the need for manual intervention, achieves the technical effect of accurately predicting the behavior of agents on urban roads, and addresses the technical issue of being unable to accurately predict the behavior of agents on urban roads. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.It should be noted that, for simplicity of description, the aforementioned method embodiments are described as a series of combined actions. However, those skilled in the art should be aware that the present disclosure is not limited by the order of the actions described, as certain steps may be performed in a different order or simultaneously, according to the present disclosure. Furthermore, those skilled in the art should also be aware that the embodiments described in this specification are preferred embodiments, and the actions and components involved are not necessarily required for the present disclosure. Through the above description of the embodiments, those skilled in the art will clearly understand that the methods according to the aforementioned embodiments can be implemented using software and a required general-purpose hardware platform, or alternatively, hardware. Based on this understanding, the technical solution of the present disclosure, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (e.g., ROM / RAM, a magnetic disk, or an optical disk) and includes instructions for enabling a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present disclosure. According to an embodiment of the present disclosure, an agent processing device for implementing the above-mentioned agent processing method is also provided. FIG9 is a schematic diagram of an agent processing device according to an embodiment of the present disclosure. As shown in FIG9 , the device includes a first determination component 901, a first acquisition component 902, a second acquisition component 903, and a first simulation component 904. The first determination component 901 is configured to determine an area to be monitored. The first acquisition component 902 is configured to acquire map information of the area and at least one agent located within the area. The second acquisition component 903 is configured to acquire simulation information associated with the map information and the at least one agent. The simulation information represents historical behavior information of the agent within the area corresponding to the map information and / or simulation rule information used when simulating the agent within the area. The first simulation component 904 is configured to use the simulation information to simulate the behavior of the at least one agent within the area to obtain at least one simulation result. It should be noted that the first determination component 901, the first acquisition component 902, the second acquisition component 903, and the first simulation component 904 correspond to steps S201 to S204 in the aforementioned embodiment. The four components and corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the aforementioned embodiment 1. It should be noted that the above-mentioned components or assemblies may be hardware components or software components stored in a memory and processed by one or more processors. The above-mentioned components may also be part of an apparatus and run in the server 10 provided in the embodiment.According to an embodiment of the present disclosure, an agent processing device for implementing the above-mentioned agent processing method is also provided. FIG10 is a schematic diagram of another agent processing device according to an embodiment of the present disclosure. As shown in FIG10 , the device includes: a second determination component 1001, a third acquisition component 1002, a fourth acquisition component 1003, a second simulation component 1004, and a sending component 1005. The second determination component 1001 is configured to determine an urban area to be monitored. The third acquisition component 1002 is configured to acquire map information of the urban area and at least one agent located within the urban area. The fourth acquisition component 1003 is configured to acquire simulation information associated with the map information and the at least one agent. The simulation information represents historical traffic behavior information of the agent within the urban area corresponding to the map information and / or simulation rule information used when simulating the agent within the area. The second simulation component 1004 is configured to use the simulation information to simulate the traffic behavior of the at least one agent within the urban area to obtain at least one simulation result. The sending component 1005 is used to send the simulation results to the city information monitoring platform. It should be noted that the second determination component 1001, the third acquisition component 1002, the fourth acquisition component 1003, the second simulation component 1004, and the sending component 1005 correspond to steps S301 to S305 in the aforementioned embodiment. The examples and application scenarios implemented by these five components and the corresponding steps are the same, but are not limited to the content disclosed in the aforementioned embodiment 1. It should be noted that the above-mentioned components or assemblies can be hardware components or software components stored in a memory and processed by one or more processors. The above-mentioned components can also be part of a device and run on the server 10 provided in the embodiment. According to an embodiment of the present disclosure, an agent processing device for implementing the above-mentioned agent simulation method is also provided. Figure 11 is a schematic diagram of another agent processing device according to an embodiment of the present disclosure. As shown in Figure 11, the device includes a first display component 1101 and a second display component 1102. The first display component 1101 is configured to display the area to be monitored on the operation interface in response to input instructions applied to the operation interface.The second display component 1102 is configured to respond to a simulation instruction applied to the operation interface and display at least one simulation result of at least one agent within the area on the operation interface. The simulation result is obtained by simulating the behavior of the at least one agent within the area using simulation information. The simulation information is associated with the map information of the area and the at least one agent, and is used to represent the agent's historical behavior within the area corresponding to the map information and / or the simulation rules used when simulating the agent within the area. It should be noted that the first display component 1101 and the second display component 1102 described above correspond to steps S401 and S402 in the aforementioned embodiment. The examples and application scenarios implemented by these two components and corresponding steps are the same, but are not limited to the content disclosed in the first embodiment. It should be noted that the above-mentioned components or assemblies can be hardware components or software components stored in memory and processed by one or more processors. These components can also be part of a device and run on the server 10 provided in the embodiment. According to an embodiment of the present disclosure, an agent processing device for implementing the above-mentioned agent processing method is also provided. FIG12 is a schematic diagram of another agent processing device according to an embodiment of the present disclosure. As shown in FIG12 , the device includes: a fifth acquisition component 1201, a sixth acquisition component 1202, a third simulation component 1203, and an output component 1204. The fifth acquisition component 1201 is configured to acquire map information of the area to be monitored by calling a first interface, wherein the first interface includes a first parameter whose parameter value includes the map information. The sixth acquisition component 1202 is configured to acquire simulation information associated with the map information and at least one agent located within the area. The simulation information represents the historical behavior of the agent within the area corresponding to the map information and / or simulation rule information used when simulating the agent within the area. The third simulation component 1203 is configured to use the simulation information to simulate the behavior of the at least one agent within the area to obtain at least one simulation result. Output component 1204 is configured to output simulation results by calling a second interface, where the second interface includes a second parameter, and the parameter value of the second parameter includes the simulation result. It should be noted that the fifth acquisition component 1201, the sixth acquisition component 1202, the third simulation component 1203, and the output component 1204 correspond to steps S501 to S504 in the aforementioned embodiment. The examples and application scenarios implemented by these four components and the corresponding steps are the same, but are not limited to the content disclosed in the aforementioned embodiment 1.It should be noted that the above-mentioned components or assemblies may be hardware components or software components stored in a memory and processed by one or more processors. These components may also be part of an apparatus and run in the server 10 provided in the embodiment. It should be noted that the preferred implementation schemes involved in the above-mentioned embodiments of the present disclosure are the same as the solutions, application scenarios, and implementation processes provided in the embodiments, but are not limited to the solutions provided in the embodiments. The embodiments of the present disclosure may provide an electronic device, which may be any electronic device in a group of electronic devices. Optionally, in this embodiment, the above-mentioned electronic device may be replaced by a terminal device such as a mobile terminal. Optionally, in this embodiment, the above-mentioned electronic device may be located in at least one of multiple network devices in a computer network. In this embodiment, the computer terminal may execute the program code in the method. Optionally, Figure 13 is a block diagram of the structure of an electronic device according to an embodiment of the present disclosure. As shown in Figure 13, the electronic device A may include: one or more (only one shown) processors 1302, a memory 1304, a storage controller, and a peripheral interface, wherein the peripheral interface is connected to a radio frequency component, an audio component, and a display. The memory can be used to store software programs and components, such as program instructions / components corresponding to the methods and apparatuses in the embodiments of the present disclosure. The processor executes the software programs and components stored in the memory to perform various functional applications and data processing, thereby implementing the methods in the aforementioned embodiments. The memory can include high-speed random access memory (RAM) and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory can further include memory remotely located from the processor, and such remote memory can be connected to electronic device A via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof. The processor can call upon information and applications stored in the memory through the transmission device to execute the following steps: determining an area to be monitored; obtaining map information of the area and at least one intelligent agent located within the area; obtaining simulation information associated with the map information and the at least one intelligent agent, wherein the simulation information is used to represent historical behavior information of the intelligent agent within the area corresponding to the map information and / or simulation rule information used when simulating the intelligent agent within the area; and using the simulation information, simulating the behavior of the at least one intelligent agent within the area to obtain at least one simulation result.Optionally, the processor may further execute program code for the following steps: determining first noise information based on map information and at least one agent, wherein the first noise information is used to indicate the type of simulation results; and determining simulation information based on the first noise information. Optionally, the processor may further execute program code for the following steps: obtaining historical state information of at least one agent in an area; determining first noise information based on the amount of information in the historical state information and the amount of information in the map information. Optionally, the processor may further execute program code for the following steps: converting the historical state information into a first matrix and converting the map information into a second matrix, wherein the amount of information in the same direction in the first and second matrices belongs to the same dimension; and obtaining first noise information that matches the dimensions in the first and second matrices. Optionally, the processor may further execute program code for the following steps: using the first noise information, determining first-type features from the first matrix and / or second-type features from the second matrix, wherein the first-type features are used to indicate characteristics of the at least one agent and the second-type features are used to indicate characteristics of the map information; and determining simulation information based on the first-type features and / or the second-type features. Optionally, the processor may further execute program code for the following steps: aligning the first type of feature with historical behavior information in the historical state information to generate simulation information corresponding to the historical behavior information; and / or aligning the second type of feature with description information corresponding to the simulation rule information to generate simulation information corresponding to the simulation rule information, wherein the description information is used to describe the simulation rule information. Optionally, the processor may further execute program code for the following steps: obtaining a distance loss function corresponding to the description information in a semantic space; determining second noise information for the region based on the description information and the distance loss function, wherein the second noise information is used to ensure that the distance between the semantics of the description information and the semantics of the map information of the region in the semantic space is less than a distance threshold; and determining, based on the second noise information, at least one target simulation result in at least one simulation result that matches the description information. Optionally, the processor may further execute program code for the following steps: sorting at least one simulation result according to the description information; selecting at least one simulation result with a higher ranking among the at least one sorted simulation results, wherein the simulation result with a higher ranking has a greater correlation with the description information; and determining, based on the second noise information, a target simulation result that matches the description information among the at least one simulation result with a higher ranking.Optionally, the processor may further execute program code for the following steps: determining a simulation index for a target simulation result, wherein the simulation index is used to indicate the reliability of the target simulation result; and sending target simulation results whose simulation index exceeds a threshold to an information monitoring platform. Optionally, the processor may further execute program code for the following steps: invoking a first encoder in a simulation model to perform attention calculation on the first noise information, the first matrix, and the second matrix to obtain first-type features and second-type features. The simulation model is obtained by training a neural network model and is used to simulate the behavior of at least one intelligent agent within an area using the simulation information. Optionally, the processor may further execute program code for the following steps: invoking a second encoder in the simulation model to map historical state information into a first matrix; invoking a third encoder in the simulation model to map map information into a second matrix. Optionally, the processor may further execute program code for the following steps: collecting image information of the area; performing semantic segmentation on the image information; and calibrating the segmented image information using road information in the area to obtain map information. In the embodiments of the present disclosure, based on map information of the area to be monitored, historical behavior information of at least one agent within the area and simulation rule information for simulating at least one agent within the area are obtained. Because the historical behavior information of the agent within the area can be used to indicate the agent's behavioral habits, the simulation rule information for simulating the agent within the area can be used to indicate the behavioral rules that the agent must comply with within the area. Therefore, by utilizing the historical behavior information of the agent within the area and the simulation rule information for simulating the agent within the area, the behavior of the agent within the area is simulated. This ensures that the simulation results are strongly correlated with the agent's historical behavior information and simulation rule information, improving the accuracy of the agent simulation results. This results in a multi-agent simulation result with controllable behavior and solutions, thus resolving the technical issue of being unable to effectively simulate the behavior of the agents. Those skilled in the art will appreciate that the structure shown in the figure is merely illustrative, and the electronic device may also be a smartphone (e.g., an Android phone, an iOS phone, etc.), a tablet computer, a PDA, or a terminal device such as a mobile internet device (MID), or a PAD. This figure does not limit the structure of the aforementioned electronic devices. For example, electronic device A may further include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a configuration different from that shown in the figure.Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by a program instructing the hardware associated with the terminal device. The program can be stored in a computer-readable storage medium, which may include a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. The embodiments of the present disclosure also provide a computer-readable storage medium. Optionally, in this embodiment, the computer-readable storage medium can be used to store program code executed by the methods provided in the above embodiments. Optionally, in this embodiment, the storage medium can be located in any electronic device in a group of electronic devices in a computer network, or in any mobile terminal in a group of mobile terminals. Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for executing the following steps: determining an area to be monitored; obtaining map information for the area and at least one agent located within the area; obtaining simulation information associated with the map information and the at least one agent, wherein the simulation information represents historical behavior information of the agent within the area corresponding to the map information and / or information about simulation rules used when simulating the agent within the area; and using the simulation information to simulate the behavior of the at least one agent within the area to obtain at least one simulation result. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: determining first noise information based on the map information and the at least one agent, wherein the first noise information represents a type of simulation result; and determining simulation information based on the first noise information. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: obtaining historical state information of the at least one agent within the area; and determining first noise information based on the amount of information in the historical state information and the amount of information in the map information. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: converting the historical state information into a first matrix, and converting the map information into a second matrix, wherein the information amount in the same direction of the first matrix and the second matrix belongs to the same dimension; and obtaining first noise information that matches the dimensions in the first matrix and the second matrix.Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: using the first noise information, determining first-type features from the first matrix, and / or determining second-type features from the second matrix, wherein the first-type features are used to represent characteristics of at least one agent, and the second-type features are used to represent characteristics of map information; and determining simulation information based on the first-type features and / or the second-type features. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: aligning the first-type features with historical behavior information in the historical state information to generate simulation information corresponding to the historical behavior information; and / or aligning the second-type features with description information corresponding to the simulation rule information to generate simulation information corresponding to the simulation rule information, wherein the description information is used to describe the simulation rule information. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: obtaining a distance loss function corresponding to the description information in a semantic space; determining second noise information for the region based on the description information and the distance loss function, wherein the second noise information is used to ensure that the distance between the semantics of the description information and the semantics of the map information of the region in the semantic space is less than a distance threshold; and determining at least one target simulation result that matches the description information from at least one simulation result based on the second noise information. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: sorting the at least one simulation result according to the description information; selecting at least one simulation result with a higher ranking from the sorted at least one simulation result, wherein a higher ranking simulation result has a greater correlation with the description information; and determining, based on the second noise information, a target simulation result that matches the description information from the at least one simulation result with a higher ranking. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: determining a simulation index for a target simulation result, wherein the simulation index is used to indicate the reliability of the target simulation result; and transmitting target simulation results whose simulation index exceeds an index threshold to an information monitoring platform. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: invoking a first encoder in a simulation model to perform attention calculation on the first noise information, the first matrix, and the second matrix to obtain first-type features and second-type features, wherein the simulation model is obtained by training a neural network model and is used to simulate the behavior of at least one agent within a region using simulation information.Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: invoking a second encoder in the simulation model to map historical state information into a first matrix; invoking a third encoder in the simulation model to map map information into a second matrix. Optionally, the computer-readable storage medium is further configured to store program code for executing the following steps: acquiring image information of a region; performing semantic segmentation on the image information; and calibrating the segmented image information using road information in the region to obtain map information. Embodiments of the present disclosure also provide a computer program product. Optionally, in this embodiment, the computer program product may include a computer program that, when executed by a processor, implements the agent processing method provided in the above embodiment. Embodiments of the present disclosure also provide a computer program product. Optionally, the computer program product may include a non-volatile computer-readable storage medium that may store a computer program that, when executed by a processor, implements the agent processing method provided in the above embodiment. Embodiments of the present disclosure also provide a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, it implements the agent processing method provided in the above embodiment. The serial numbers of the above embodiments of the present disclosure are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. In the above embodiments of the present disclosure, the descriptions of each embodiment are given with emphasis. For portions not described in detail in a particular embodiment, reference should be made to the relevant descriptions of other embodiments. It should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of components is merely a logical functional division. In actual implementation, other divisions may be used, such as combining or integrating multiple components or components into another system, or omitting or not implementing certain features. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through interfaces, or indirect coupling or communication connection between components or components may be electrical or other forms. The components described as separate parts may or may not be physically separate, and the components shown as components may or may not be physical components, that is, they may be located in one place or distributed across multiple network components. Some or all of these components may be selected based on actual needs to achieve the objectives of the present embodiment. Furthermore, the functional components in various embodiments of the present disclosure may be integrated into a single processing component, each component may exist physically separately, or two or more components may be integrated into a single component.The aforementioned integrated components can be implemented in either hardware or software functional components. If implemented as software functional components and sold or used as independent products, the integrated components can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, or the portion that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes instructions for enabling a computer device (such as a personal computer, server, or network device) to perform all or part of the steps of the methods described in various embodiments of this disclosure. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, ROM, RAM, removable hard drives, magnetic disks, or optical disks. The above description is merely a preferred embodiment of the present disclosure. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this disclosure, and such improvements and modifications should also be considered within the scope of protection of this disclosure. Industrial Applicability The solution provided by the embodiments of the present disclosure can be applied to the simulation processing process of an intelligent agent. Based on the map information of the area to be monitored, the historical behavior information of at least one intelligent agent in the area and the simulation rule information for simulating at least one intelligent agent in the area corresponding to the map information are obtained. Since the historical behavior information of the intelligent agent in the area can be used to indicate the behavioral habits of the intelligent agent, the simulation rule information for simulating the intelligent agent in the area can be used to indicate the behavioral rules that the intelligent agent must comply with when behaving in the area. Based on this, the historical behavior information of the intelligent agent in the area and the simulation rule information for simulating the intelligent agent in the area are used to simulate the behavior of the intelligent agent in the area. Based on this, the simulation result can be strongly correlated with the historical behavior information and simulation rule information of the intelligent agent, thereby improving the accuracy of the simulation result generation of the intelligent agent, and thus obtaining a multi-agent simulation result with controllable behavior and scheme.

Claims

Claims 1. A method for processing an intelligent agent, comprising: Identify areas to be monitored; Acquire map information of the area and at least one intelligent agent located in the area; Acquiring simulation information associated with the map information and at least one of the agents, wherein the simulation information is used to represent historical behavior information of the agent within the area corresponding to the map information and / or simulation rule information used when simulating the agent within the area; The simulation information is used to simulate the behavior of at least one of the agents in the area to obtain at least one simulation result.

2. The method according to claim 1, wherein: Acquiring simulation information associated with the map information and at least one of the agents includes: determining first noise information based on the map information and the at least one of the agents, wherein the first noise information is used to indicate a type of the simulation result; and determining the simulation information based on the first noise information.

3. The method according to claim 2, wherein: Determining first noise information based on the map information and at least one of the agents includes: obtaining historical state information of at least one of the agents in the area; and determining the first noise information based on an amount of information in the historical state information and an amount of information in the map information.

4. The method according to claim 3, wherein: Determining the first noise information based on the information volume of the historical state information and the information volume of the map information includes: converting the historical state information into a first matrix and converting the map information into a second matrix, wherein the information volume of the first matrix and the second matrix in the same direction belongs to the same dimension; and obtaining the first noise information matching the dimensions in the first matrix and the second matrix.

5. The method according to claim 4, wherein Determining the simulation information based on the first noise information includes: using the first noise information to determine a first type of feature from the first matrix, and / or determining a second type of feature from the second matrix, wherein the first type of feature is used to represent a feature of at least one of the intelligent agents, and the second type of feature is used to represent a feature of the map information; and determining the simulation information based on the first type of feature and / or the second type of feature.

6. The method according to claim 5, wherein Determining the simulation information based on the first type feature and / or the second type feature includes: aligning the first type feature with the historical behavior information in the historical state information, 28 generating the simulation information corresponding to the historical behavior information; and / or aligning the second type feature with the description information corresponding to the simulation rule information to generate the simulation information corresponding to the simulation rule information, wherein the description information is used to describe the simulation rule information.

7. The method according to claim 6, wherein: The method further includes: obtaining a distance loss function corresponding to the description information in a semantic space; determining second noise information of the area based on the description information and the distance loss function, wherein the second noise information is used to make the distance between the semantics of the description information and the semantics of the map information of the area in the semantic space less than a distance threshold; and determining at least one target simulation result that matches the description information in the at least one simulation result based on the second noise information.

8. The method according to claim 7, wherein: Determining, based on the second noise information, at least one target simulation result that matches the description information in the at least one simulation result, includes: sorting the at least one simulation result according to the description information; selecting at least one simulation result with a higher ranking among the at least one sorted simulation results, wherein the higher the ranking of the simulation result, the greater the correlation between the simulation result and the description information; and determining, based on the second noise information, the target simulation result that matches the description information in the at least one simulation result with a higher ranking.

9. The method according to claim 7, wherein: The method further includes: determining a simulation index of the target simulation result, wherein the simulation index is used to represent the reliability of the target simulation result; and sending the target simulation result whose simulation index is greater than an index threshold to an information monitoring platform.

10. The method according to claim 5, wherein: Utilizing the first noise information, determining a first type of feature from the first matrix, and determining a second type of feature from the second matrix, includes: calling a first encoder in a simulation model to perform attention calculation on the first noise information, the first matrix, and the second matrix to obtain the first type of feature and the second type of feature, wherein the simulation model is obtained by training a neural network model and is used to simulate the behavior of at least one of the intelligent agents within the area using the simulation information.

11. The method according to claim 10, wherein: Converting the historical state information into a first matrix and converting the map information into a second matrix includes: calling a second encoder in the simulation model to map the historical state information into the first matrix; and calling a third encoder in the simulation model to map the map information into the second matrix.

12. The method according to any one of claims 1 to 11, wherein Get map information of the area, The method comprises: collecting image information of the area; performing semantic segmentation on the image information; and calibrating the segmented image information using road information of the area to obtain the map information.

13. A method for processing an intelligent agent, comprising: Identify urban areas to be monitored; Acquire map information of the urban area and at least one intelligent agent located in the urban area; Acquiring simulation information associated with the map information and at least one of the agents, wherein the simulation information is used to represent historical traffic behavior information of the agent within the urban area corresponding to the map information and / or simulation rule information used when simulating the agent within the area; Using the simulation information, simulate the traffic behavior of at least one of the intelligent agents in the urban area to obtain at least one simulation result; and send the simulation result to the urban information monitoring platform.

14. The method according to claim 13, wherein Acquiring simulation information associated with the map information and at least one of the agents includes: determining first noise information based on the map information and the at least one of the agents, wherein the first noise information is used to indicate a type of the simulation result; and determining the simulation information based on the first noise information.

15. A method for processing an intelligent agent, comprising: Acquire map information of an area to be monitored by calling a first interface, wherein the first interface includes a first parameter, and the parameter value of the first parameter includes the map information; acquire simulation information associated with the map information and at least one intelligent agent located in the area, wherein the simulation information is used to represent historical behavior information of the intelligent agent in the area corresponding to the map information, and / or simulation rule information used when simulating the intelligent agent in the area; use the simulation information to simulate the behavior of at least one intelligent agent in the area to obtain at least one simulation result; and output the simulation result by calling a second interface, wherein the second interface includes a second parameter, and the parameter value of the second parameter includes the simulation result.

16. An electronic device, comprising: a memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 15 when running.

17. A computer-readable storage medium, wherein: The computer readable storage medium includes a stored executable program The program is configured to control the device where the computer-readable storage medium is located to execute the method according to any one of claims 1 to 15 when the executable program is running.

18. A computer program product, wherein: The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 15.

19. A computer program product, wherein: The method comprises a non-volatile computer-readable storage medium storing a computer program, wherein the computer program implements the method according to any one of claims 1 to 15 when executed by a processor.

20. A computer program, wherein When the computer program is executed by a processor, the method according to any one of claims 1 to 15 is implemented.

Images