An environment perception system based on multi-agent interaction

Abstract

The present invention discloses an environment perception system based on multi-agent interaction, comprising: an environment perception data acquisition unit for acquiring multi-agent environment perception data; a data feature vector construction unit for acquiring environment perception standard data, performing noise elimination processing on the perception standard data, constructing a data feature vector, and acquiring standard association data from the environment perception standard data; a perception analysis model construction unit for clustering the data feature vector according to the standard association data to obtain a to-be-trained data set, and performing multi-agent interaction training on the initial model with the to-be-trained data set to obtain a multi-agent perception analysis model; and a perception analysis unit for performing hierarchical processing on the environment perception data to obtain perception feedback data, and inputting the perception feedback data into the multi-agent perception analysis model to obtain a dynamic perception result. The survival ability of the agent in a complex environment is improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular, to an environment perception system based on multi-agent interaction. Background Art

[0002] In current environment perception technologies, mainly rely on single or limited number of sensor devices to obtain and process physical parameters in the environment. These systems usually use fixed perception points, such as cameras, temperature sensors, humidity sensors, etc., to collect data and analyze it through preset algorithms. However, this traditional method has significant limitations when facing complex and dynamic environments.

[0003] Due to the lack of intelligent processing capabilities in traditional perception systems, they cannot effectively respond to and adjust to the changing factors in the environment. This defect is particularly obvious in complex environments with multiple targets and variables. For example, in disaster monitoring, intelligent traffic management, or complex industrial production environments, environmental conditions often change rapidly, and single or preset perception means are difficult to cope with diverse scenario changes, easily leading to decision-making errors or slow responses.

[0004] The deficiencies of the prior art in terms of adaptability, dynamics, and intelligent processing of environment perception. As an emerging technology that simulates complex social behaviors and adaptive systems, multi-agent systems provide the possibility to solve this problem. However, current multi-agent systems mainly focus on the fields of simulation and emulation, lacking an organic combination with actual environment perception. In the prior art, there is no mature solution that can fully utilize the collaboration and interaction of multi-agents to construct a dynamic and efficient environment perception system that can adapt to complex environments.

[0005] Therefore, there is an urgent need for an environment perception system based on multi-agent interaction. Summary of the Invention

[0006] The present invention provides an environment perception system based on multi-agent interaction to solve the above problems existing in the prior art.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] An environment perception system based on multi-agent interaction, comprising:

[0009] An environment perception data acquisition unit, configured to acquire multi-agent environment perception data, where the environment perception data includes environmental characteristic parameters and agent perception parameters;

[0010] A data feature vector construction unit, configured to acquire environment perception standard data, perform noise elimination processing on the perception standard data, construct a data feature vector, and obtain standard association data from the environment perception standard data;

[0011] Construct a perception analysis model unit, which is used to cluster data feature vectors according to standard associated data to obtain a data group to be trained, and after multi-agent interactive training of the data group to be trained on an initial model, obtain a multi-agent perception analysis model;

[0012] A perception analysis unit, which is used to perform hierarchical processing on environmental perception data to obtain perception feedback data, and input the perception feedback data into the multi-agent perception analysis model to obtain a dynamic perception result.

[0013] Among them, the environmental perception data acquisition unit includes:

[0014] An obtained sensing result module, which is used to control multiple agents to simultaneously start sensing operations in a preset environment, output initial environmental perception data, and determine the sensing results of each agent regarding the current environmental perception data. The sensing results include data acquisition accuracy, acquisition frequency, and corresponding environmental feature parameters;

[0015] An environmental perception data acquisition module, which is used to match and analyze the sensing results of each agent with a preset perception model, determine new environmental perception tasks and allocate them to the agents, and generate new environmental perception data;

[0016] A feature extraction module, which is used to extract features from the new environmental perception data to obtain new environmental feature parameters and agent perception parameters, and determine multiple new environmental perception tasks until the change rate of the environmental feature parameters is less than a preset threshold, or determine that the current perception task is the last task;

[0017] A completed perception task module, which is used to control multiple agents to perform sensing operations in another preset environment, and repeatedly execute the steps from the output of the initial environmental perception data to the change rate of the current environmental feature parameters being less than the preset threshold until all preset environmental perception tasks are completed.

[0018] Among them, the data feature vector construction unit includes:

[0019] A preprocessing operation module, which is used to preprocess the obtained environmental perception standard data to eliminate incomplete or abnormal data, input the preprocessed environmental perception standard data into a noise elimination module, and perform noise elimination processing on the environmental perception standard data to obtain denoised data;

[0020] An initial data feature vector construction module, which is used to extract feature parameters from the denoised data according to preset rules and construct an initial data feature vector;

[0021] Construct a second data feature vector module, which is used to input the initial data feature vector into the feature screening module, screen the initial data feature vector according to the set screening logic, and obtain the optimized data feature vector;

[0022] Construct a third data feature vector module, which is used to obtain correlation parameters from the optimized data feature vector, obtain standard associated data associated with the optimized data feature vector in combination with the set logical relationship, input the standard associated data into the feature construction module, and further process the standard associated data to generate the final data feature vector.

[0023] Among them, obtaining the multi-agent perception analysis model includes:

[0024] Perform clustering processing on the data feature vector to generate multiple clustering results;

[0025] Determine the data group to be trained based on the clustering results and output the data group to be trained;

[0026] Input the data group to be trained into the initial model to perform multi-agent interactive training;

[0027] During the multi-agent interactive training process, obtain the perception data of each agent and perform real-time analysis on the perception data;

[0028] Based on the perception data of each agent and the feedback data in the interactive training, update the parameters of the initial model and output the updated model parameters;

[0029] Based on the updated model parameters, generate and output the multi-agent perception analysis model;

[0030] Verify the multi-agent perception analysis model and adjust the parameters of the multi-agent perception analysis model based on the verification results;

[0031] In the adjusted multi-agent perception analysis model, perform multi-agent interactive training again until the preset convergence condition is reached, and output the final multi-agent perception analysis model.

[0032] Among them, obtaining the dynamic perception result includes:

[0033] Perform hierarchical processing on the perception data to generate multiple perception feature sets at different levels;

[0034] Based on the preset feature extraction algorithm, extract multiple key features from the perception feature set and generate perception feedback data;

[0035] Input the perception feedback data into the multi-agent perception analysis model to trigger the multi-agent perception analysis model to process the perception feedback data;

[0036] During the processing, the analysis results of each agent are obtained in real time, and multiple sub-level perception results are generated based on the analysis results;

[0037] Integrate the sub-level perception results to generate a preliminary dynamic perception result;

[0038] Based on the preset dynamic adjustment rules, dynamically adjust the preliminary dynamic perception result to obtain the final dynamic perception result.

[0039] Among them, determining a new environmental perception task and assigning it to an agent includes:

[0040] Determine multiple second sensing results generated by the agent historically based on the sensing result database;

[0041] Extract multiple first perception features of the second sensing results;

[0042] Obtain the preset perception model library, match the first perception features with the second perception features in the perception model library. If the match is successful, regard the successfully matched second sensing result as the third sensing result. At the same time, obtain the perception task corresponding to the successfully matched second perception feature and associate it with the corresponding agent;

[0043] Obtain the historical environmental item corresponding to the third sensing result, and at the same time, obtain the sensing generation moment corresponding to the third sensing result;

[0044] Determine multiple fourth sensing results in the sensing result database that correspond to the historical environmental item and are generated within a preset time period before and after the generation moment;

[0045] Input all the fourth sensing results into the preset task analysis model to obtain at least one new perception task output by the task analysis model. At the same time, accumulate and calculate the perception tasks associated with the agent to obtain the task total;

[0046] If the perception task is the first task and / or the perception task is the fourth task other than the third task in the second task and / or the task total is greater than or equal to the preset first task total threshold, the agent perception task assignment fails;

[0047] Otherwise, obtain the preset environmental perception task process record library, and determine multiple environmental perception task process records corresponding to the perception item from the environmental perception task process record library;

[0048] Input all the environmental perception task process records into the preset importance analysis model to obtain the importance degree output by the importance analysis model;

[0049] Obtain the second task total threshold corresponding to the importance degree;

[0050] If the total of the tasks is greater than or equal to the second task total threshold, the agent perceives that the task assignment has failed;

[0051] Otherwise, the agent perceives that the task assignment is successful.

[0052] The acquired environmental perception standard data is preprocessed, including:

[0053] Determine a preliminary processing result from the environmental perception standard data, and extract multiple data features from the preliminary processing result;

[0054] Determine whether the data features meet the preset integrity standards. If so, execute: further process the preliminary processing results, remove redundant data, and output the processed environmental perception standard data;

[0055] If not, the incomplete or abnormal data features are marked and removed from the environmental perception standard data;

[0056] It is determined again whether the remaining data features in the environmental perception standard data after the elimination meet the preset integrity standard. If so, further processing is performed on the data features and the final processing result is output;

[0057] If not, repeat the process: remove incomplete or abnormal data from the environment perception standard data set until the data features meet the preset integrity standards or the size of the environment perception standard data reaches the preset minimum threshold.

[0058] Among them, multiple clustering results are generated, including:

[0059] Obtaining a data feature vector set, where the data feature vector set includes multiple data feature vectors;

[0060] Performing a preprocessing step on the data feature vector set, the preprocessing step including normalizing the data feature vector;

[0061] According to a preset clustering algorithm, the similarity between each data feature vector and other data feature vectors is determined, and according to a similarity matrix, the data feature vector set is divided into a plurality of preliminary clustering results;

[0062] According to the preset optimization strategy, the preliminary clustering results are optimized, and the optimization includes readjusting the clustering boundaries to improve the overall consistency of the clustering results and generate the final clustering results;

[0063] Determine whether the final clustering result meets the preset clustering evaluation criteria. If so, output the final clustering result.

[0064] Among them, multiple perceptual feature sets of different levels are generated, including:

[0065] Obtain a set of perception data, where the set of perception data includes multiple perception data points;

[0066] According to a preset hierarchical processing rule, divide the set of perception data into several preliminary perception levels, and the hierarchical processing rule is determined based on the feature dimension, data density, and correlation of the perception data;

[0067] Perform hierarchical optimization on the preliminary perception levels. The hierarchical optimization includes adjusting the hierarchical boundaries and merging similar levels to ensure the rationality of the hierarchical division and generate multiple optimized perception levels;

[0068] Based on the data characteristics of each perception level, extract the feature set of the perception level respectively, and the feature set is generated through a preset feature extraction algorithm;

[0069] Combine the extracted perception feature sets in the hierarchical order to generate multiple perception feature sets at different levels.

[0070] Among them, obtaining the set of perception data includes:

[0071] Obtain the request information of the perception data;

[0072] According to the perception data acquisition mechanism, identify whether the perception data meets the preset conditions, where the perception data acquisition mechanism is used to configure the perception data acquisition strategy;

[0073] When the perception data meets the preset conditions, based on the data association analysis mechanism, determine the correlation of the current perception data point;

[0074] When the perception data points are relevant, summarize the perception data points that meet the conditions to form a set of perception data.

[0075] Compared with the prior art, the present invention has the following advantages:

[0076] An environment perception system based on multi-agent interaction, comprising: an environment perception data acquisition unit for acquiring multi-agent environment perception data, where the environment perception data includes environmental characteristic parameters and agent perception parameters; a data feature vector construction unit for acquiring standard environment perception data, performing noise elimination processing on the standard perception data, constructing a data feature vector, and obtaining standard associated data from the standard environment perception data; a perception analysis model construction unit for clustering the data feature vector according to the standard associated data to obtain a to-be-trained data set, and performing multi-agent interaction training on the initial model with the to-be-trained data set to obtain a multi-agent perception analysis model; a perception analysis unit for performing hierarchical processing on the environment perception data to obtain perception feedback data, and inputting the perception feedback data into the multi-agent perception analysis model to obtain a dynamic perception result. It not only improves the survival ability of agents in complex environments, but also enhances the security and operation efficiency of the overall system.

[0077] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention.

[0078] The technical solutions of the present invention will be further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification, and are used to explain the present invention together with the embodiments of the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0080] Figure 1 is a structural diagram of an environment perception system based on multi-agent interaction in an embodiment of the present invention;

[0081] Figure 2 is a structural diagram of the environment perception data acquisition unit in an embodiment of the present invention;

[0082] Figure 3 is a structural diagram of the data feature vector construction unit in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0083] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0084] The embodiment of the present invention provides an environment perception system based on multi-agent interaction, comprising:

[0085] An environment perception data acquisition unit for acquiring multi-agent environment perception data, where the environment perception data includes environmental characteristic parameters and agent perception parameters;

[0086] Build a data feature vector unit for obtaining environmental perception standard data, performing noise elimination processing on the perception standard data, constructing a data feature vector, and obtaining standard associated data from the environmental perception standard data;

[0087] Build a perception analysis model unit for clustering the data feature vector according to the standard associated data to obtain a data group to be trained, and performing multi-agent interaction training on the initial model with the data group to be trained to obtain a multi-agent perception analysis model;

[0088] A perception analysis unit for performing hierarchical processing on environmental perception data to obtain perception feedback data, and inputting the perception feedback data into the multi-agent perception analysis model to obtain a dynamic perception result.

[0089] The working principle of the above technical solution is as follows: First, through the environmental perception data acquisition unit, the characteristic parameters of the environment where the multi-agent is located and the perception parameters of the agent itself are collected. These data include the physical characteristics of the environment (such as temperature, humidity, light, etc.) and the state information of the agent (such as position, speed, direction, etc.). These information are the basis for subsequent analysis and processing.

[0090] After data collection, the system standardizes the collected environmental perception data through the data feature vector construction unit. This process includes noise elimination to ensure the accuracy and consistency of the data. On this basis, the system generates a data feature vector and extracts standard associated data related to the environment to provide support for subsequent clustering and model training.

[0091] Next, the system uses the standard associated data to perform clustering analysis on the data feature vector, extracts the data groups to be trained from it, and these data groups are used to train a multi-agent perception analysis model. Through multiple trainings, the model can learn and simulate the interaction behavior between agents and the interaction between agents and the environment, thereby improving the system's perception and understanding ability of the environment.

[0092] Finally, the system performs hierarchical processing on the real-time environmental perception data through the perception analysis unit to obtain perception feedback data. Then, the system inputs these feedback data into the trained multi-agent perception analysis model to obtain a dynamic perception result of the current environment. These results can be used to guide the further actions of the agent or provide decision-making support.

[0093] Suppose in an intelligent transportation system, multiple autonomous vehicles act as agents and drive on the road. The system first collects the environmental information around the vehicles (such as road conditions, weather conditions, etc.) and the driving states of the vehicles (such as speed, direction). After noise processing, these data are used to train a perception analysis model. During the actual operation of the system, when the road conditions or weather change, the system can perceive and analyze these changes in real time, provide accurate perception results to each vehicle, and thus help the vehicle make safe driving decisions. For example, when driving in foggy weather, the system can, through real-time perception analysis, suggest that the vehicle reduce its speed and maintain a safe distance to prevent traffic accidents from occurring.

[0094] The beneficial effects of the above technical solution are as follows: Through noise elimination and normalization processing, the system can ensure the accuracy of the data, thereby improving the reliability of the perception and analysis results; the data feature vectors and clustering analysis enable the system to effectively process a large amount of environmental data and quickly construct an efficient perception analysis model; the multi-agent interactive training process enables the system to have the ability to adapt to different environmental changes, and can maintain a high perception accuracy even in complex or dynamic environments.

[0095] In another embodiment, the environmental perception data acquisition unit includes:

[0096] A sensing result acquisition module, which is used to control multiple agents to start sensing operations simultaneously in a preset environment, output initial environmental perception data, determine the sensing results of each agent regarding the current environmental perception data, and the sensing results include data acquisition accuracy, acquisition frequency, and corresponding environmental feature parameters;

[0097] An environmental perception data acquisition module, which is used to match and analyze the sensing results of each agent with a preset perception model, determine new environmental perception tasks and assign them to the agents, and generate new environmental perception data;

[0098] A feature extraction module, which is used to extract features from the new environmental perception data, obtain new environmental feature parameters and agent perception parameters, and determine multiple new environmental perception tasks until the change rate of the environmental feature parameters is less than a preset threshold, or determine that the current perception task is the last task;

[0099] A perception task completion module, which is used to control multiple agents to perform sensing operations in another preset environment, and repeat the steps from the output of the initial environmental perception data to the change rate of the current environmental feature parameters being less than the preset threshold until all the perception tasks in the preset environments are completed.

[0100] The working principle of the above technical solution is as follows: The module for obtaining sensing results starts first. It commands multiple agents to start performing sensing operations in a preset environment. These agents are equipped with various sensors, such as cameras, microphones, temperature sensors, etc. They collect information in the environment according to predetermined parameters (such as data acquisition accuracy, acquisition frequency); the data collected by each agent is sent back to the module for obtaining sensing results, and this module processes and analyzes this data to identify environmental characteristic parameters, such as temperature, humidity, light intensity, etc.; then the module for obtaining environmental sensing data intervenes. It compares and matches the data collected by the agents with a preset sensing model. This step is to determine whether the environment has changed or whether new sensing tasks need to be performed by the agents. If the environment has changed or there are new sensing requirements, the module will determine new environmental sensing tasks and assign them to the corresponding agents.

[0101] The feature extraction module will deeply process the new environmental sensing data to extract key features. These features include not only environmental parameters but also the sensing performance parameters of the agents. The module continuously monitors the change rate of the environmental characteristic parameters. Once the change rate is less than a preset threshold, indicating that the environment is stable or the expected sensing depth has been reached, the module will determine that the current sensing task is the last task.

[0102] Finally, the module for completing sensing tasks will guide the agents to repeat the above process in another preset environment. This process will loop until all the sensing tasks in all preset environments are completed.

[0103] The beneficial effects of the above technical solution are as follows: The agents can efficiently collect environmental data according to preset parameters, ensuring the accuracy and timeliness of the data. Through the preset sensing model, the agents can adapt to environmental changes and dynamically adjust sensing tasks, thereby improving the flexibility and accuracy of environmental monitoring. The feature extraction module can effectively extract key information from a large amount of sensing data, reduce the amount of data processing, and speed up the identification speed of environmental changes. The module for completing sensing tasks ensures that the agents can systematically and comprehensively cover all preset sensing tasks, avoiding sensing blind spots and ensuring the integrity of environmental monitoring.

[0104] In another embodiment, the unit for constructing a data feature vector includes:

[0105] A preprocessing operation module, which is used to preprocess the obtained standard environmental sensing data to eliminate incomplete or abnormal data, and input the preprocessed standard environmental sensing data into a noise elimination module to perform noise elimination processing on the standard environmental sensing data to obtain denoised data;

[0106] A module for constructing an initial data feature vector, which is used to extract feature parameters from the denoised data according to preset rules and construct an initial data feature vector;

[0107] Construct a second data feature vector module for inputting the initial data feature vector into a feature screening module, screening the initial data feature vector according to the set screening logic, and obtaining an optimized data feature vector;

[0108] Construct a third data feature vector module for obtaining correlation parameters from the optimized data feature vector, obtaining standard correlation data associated with the optimized data feature vector in combination with the set logical relationship, inputting the standard correlation data into a feature construction module, and further processing the standard correlation data to generate a final data feature vector.

[0109] The working principle of the above technical solution is as follows: The main purpose of the preprocessing operation module is to ensure the quality and integrity of the input data. It identifies and eliminates incomplete or abnormal data through a series of algorithms and rules. This process may include steps such as data cleaning, data verification, and data filling. The preprocessed data is considered "clean" and can be used for subsequent processing flows.

[0110] The role of the noise elimination module is to perform noise elimination processing on the preprocessed data to improve the credibility and accuracy of the data. Noise comes from various reasons, such as sensor errors, data transmission errors, etc. By using filtering algorithms, smoothing algorithms, etc., the denoising module can reduce the impact of these noises and generate smoother and more reliable data.

[0111] The goal of constructing the initial data feature vector module is to extract useful information from the denoised data and convert it into a feature vector. These feature parameters may include environmental variables such as temperature, humidity, light intensity, etc. Preset rules guide how to extract these feature parameters from the data and construct the feature vector.

[0112] The construction of the second data feature vector module and the third data feature vector module further process the initial feature vector. The feature screening module optimizes the initial data feature vector according to the set screening logic, only retaining the most important features. This helps to reduce the complexity of the data and may improve the efficiency of subsequent analysis and processing. Finally, by combining the optimized data feature vector and a specific logical relationship, the standard correlation data is extracted and further processed in the feature construction module to generate a final data feature vector.

[0113] The beneficial effects of the above technical solution are as follows: It ensures the quality and accuracy of data. The preprocessing operation module eliminates incomplete or abnormal data, thus reducing the impact of incorrect data on the final analysis results. The denoising module improves the credibility of the data, making the decisions based on this data more reliable. By constructing feature vectors, the data is converted into a format that is easier to process and analyze. The optimization process of the feature screening module helps to extract the most critical information, which may improve the analysis efficiency and reduce the consumption of computing resources. Finally, by generating the final data feature vectors, the entire system can process and interpret the data more effectively, providing valuable information and insights to users.

[0114] In another embodiment, obtaining a multi-agent perception analysis model includes:

[0115] Performing clustering processing on the data feature vectors to generate multiple clustering results;

[0116] Determining a training data group based on the clustering results and outputting the training data group;

[0117] Inputting the training data group into an initial model to perform multi-agent interactive training;

[0118] During the multi-agent interactive training process, obtaining the perception data of each agent and performing real-time analysis on the perception data;

[0119] Updating the parameters of the initial model based on the perception data of each agent and the feedback data in the interactive training, and outputting the updated model parameters;

[0120] Generating and outputting a multi-agent perception analysis model based on the updated model parameters;

[0121] Verifying the multi-agent perception analysis model and adjusting the parameters of the multi-agent perception analysis model based on the verification results;

[0122] In the adjusted multi-agent perception analysis model, performing multi-agent interactive training again until a preset convergence condition is reached, and outputting the final multi-agent perception analysis model.

[0123] The working principle of the above technical solution is as follows: Clustering the data feature vectors is to group similar data points together to discover the natural clusters or patterns in the data. This step is usually implemented using algorithms such as K-means, DBSCAN, etc. The clustering results generate multiple cluster centers, and each center represents a data group. Based on the clustering results, the data groups to be trained can be determined. These data groups are selected from the clusters and they contain more similar features and behavior patterns, so they are selected as important data for training the model. The data groups to be trained are then input into the initial model. In the multi-agent interactive training stage, the models learn how to better understand and predict the data through simulated interactions. This training method can improve the generalization ability and interactivity of the models. During the training process, each agent generates perception data. These data reflect the performance and behavior of the model during training. Analyzing these perception data in real time can help monitor the training progress and performance of the model. According to the perception data of each agent and the feedback data in the interactive training, the parameters of the model are updated, and this process is to optimize the model performance by adjusting the internal parameters of the model. The updated model parameters generate a new model version.

[0124] The generated multi-agent perception analysis model needs to be verified to ensure that its performance meets the expectations. The verification may include using a data set that did not participate in the training to test the accuracy and generalization ability of the model. According to the verification results, the model parameters may need to be further adjusted to improve the quality of the model. Repeat the multi-agent interactive training in the adjusted model until the model reaches the preset convergence condition. Convergence means that the performance of the model no longer improves significantly, or the changes in the model parameters tend to be stable. The final multi-agent perception analysis model is output for actual perception and analysis tasks.

[0125] The beneficial effects of the above technical solution are as follows: It can improve the learning efficiency and accuracy of the model through the interaction of multiple agents. The clustering process helps identify the key features in the data, while the multi-agent interactive training promotes the adaptability and learning ability of the model. Analyzing the perception data in real time ensures the monitorability and transparency of the training process. By continuously updating the model parameters and verifying the model performance, the final multi-agent perception analysis model has higher accuracy and better generalization ability. This method is particularly suitable for complex and dynamic environments because it can self-adjust and improve to adapt to the changing data and conditions.

[0126] In another embodiment, obtaining the dynamic perception result includes:

[0127] Performing hierarchical processing on the perception data to generate multiple perception feature sets at different levels;

[0128] Extract multiple key features from the set of perceptual features based on a preset feature extraction algorithm, and generate perceptual feedback data;

[0129] Input the perceptual feedback data into the multi-agent perception analysis model, triggering the multi-agent perception analysis model to process the perceptual feedback data;

[0130] During the processing, obtain the analysis results of each agent in real time, and generate multiple sub-level perceptual results based on the analysis results;

[0131] Integrate the sub-level perceptual results to generate a preliminary dynamic perceptual result;

[0132] Based on a preset dynamic adjustment rule, dynamically adjust the preliminary dynamic perceptual result to obtain the final dynamic perceptual result.

[0133] The working principle of the above technical solution is as follows: Hierarchical processing of perceptual data is to understand and utilize data at different abstraction levels. This step usually involves decomposing data into multiple levels, each level capturing different aspects or details of the data. For example, in image processing, hierarchical processing may include from raw pixel data to edge detection, and then to higher-level feature recognition, etc. After hierarchical processing, a preset feature extraction algorithm is applied to the set of perceptual features at each level to identify and extract key features. These key features are considered crucial for understanding the data and building the model. Perceptual feedback data is a collection of these extracted features, which represents important information about the environment or scene.

[0134] The perceptual feedback data is then input into the multi-agent perception analysis model. This model consists of multiple agents, each agent is responsible for processing a part of the data, and shares information and experience through interaction. The model processes the perceptual feedback data to identify patterns, make predictions or generate decisions. During the processing, each agent generates its analysis result in real time. These results reflect the processing and interpretation of the model on the perceptual feedback data. Based on these analysis results, multiple sub-level perceptual results can be generated, which are more specific and refined, providing in-depth insights into the original data.

[0135] The sub-level perceptual results are then integrated to generate a preliminary dynamic perceptual result. These results synthesize the analysis and interpretation of multiple agents, providing a basis for subsequent decisions and behaviors. Finally, based on a preset dynamic adjustment rule, the preliminary dynamic perceptual result is adjusted. These rules involve weight assignment, threshold setting or feedback mechanism, ensuring the accuracy and adaptability of the perceptual result. After dynamic adjustment, the final dynamic perceptual result is obtained, which can be used to guide the behavior of the agent or further decision-making process.

[0136] The beneficial effects of the above technical solution are as follows: The accuracy and real-time performance of perception analysis are improved through the collaborative work of multiple agents. The hierarchical processing and feature extraction algorithms help identify key information from complex data, while the multi-agent perception analysis model enhances its analysis ability through interaction and information sharing. The analysis results of each agent are obtained in real time and sub-level perception results are generated, enabling deeper mining and understanding of the data. The final dynamic perception results generated by integrating and dynamically adjusting these results provide accurate perception information for the agents, enabling them to better adapt to the environment and execute tasks.

[0137] In another embodiment, determining a new environmental perception task and assigning it to an agent includes:

[0138] Determining multiple second sensing results generated by the agent historically based on the sensing result database;

[0139] Extracting multiple first perception features of the second sensing results;

[0140] Obtaining a preset perception model library, matching the first perception features with the second perception features in the perception model library. If the match is successful, taking the successfully matched second sensing result as the third sensing result. At the same time, obtaining the perception task corresponding to the successfully matched second perception feature and associating it with the corresponding agent;

[0141] Obtaining the historical environmental item corresponding to the third sensing result, and at the same time, obtaining the sensing generation moment corresponding to the third sensing result;

[0142] Determining multiple fourth sensing results corresponding to the historical environmental item in the sensing result database and generated within a preset time period before and after the generation moment;

[0143] Inputting all the fourth sensing results into a preset task analysis model, obtaining at least one new perception task output by the task analysis model, and at the same time, cumulatively calculating the perception tasks associated with the agent to obtain the task total;

[0144] If the perception task is the first task and / or the perception task is the fourth task other than the third task in the second task and / or the task total is greater than or equal to the preset first task total threshold, the agent perception task assignment fails;

[0145] Otherwise, obtaining a preset environmental perception task process record library, and determining multiple environmental perception task process records corresponding to the perception item from the environmental perception task process record library;

[0146] Inputting all the environmental perception task process records into a preset importance analysis model, and obtaining the importance degree output by the importance analysis model;

[0147] Obtain a second task sum threshold corresponding to the importance;

[0148] If the total of the tasks is greater than or equal to the second task total threshold, the agent perceives that the task assignment has failed;

[0149] Otherwise, the agent perceives that the task assignment is successful.

[0150] The working principle of the above technical solution is as follows: historical data analysis, obtaining the second sensory result generated by the agent in history from the sensory result database, extracting the first sensory feature of the second sensory result, this step uses historical data to provide a basis for subsequent analysis. Model matching, matching the first sensory feature with the second sensory feature in the sensory model library, and determining the third sensory result and the corresponding sensory task if the match is successful. This step determines the appropriate sensory task through model matching. Environmental analysis, obtaining the historical environmental item and generation time corresponding to the third sensory result, determining the fourth sensory result in the relevant time period, this step analyzes the environmental background of the sensory result. Task analysis, inputting the fourth sensory result into the task analysis model, obtaining a new sensory task, and accumulating and calculating the sum of sensory tasks associated with the agent. This step generates a new task based on the environmental analysis results. Task screening, preliminary screening is performed based on the task type and the task sum, this step filters out unsuitable tasks. Importance analysis, inputting the environmental perception task process record into the importance analysis model, obtaining the importance and the corresponding task sum threshold, this step further analyzes the importance of the task. Final judgment, comparing the task sum with the threshold, deciding whether to pass the task allocation, this step makes the final task allocation decision.

[0151] The beneficial effects of the above technical solution are: automatically assigning tasks using historical data and multiple models; ensuring that tasks match the capabilities of the agent through multiple screenings; dynamically adjusting task assignments based on the environment and historical records; preventing the agent from having too many tasks through task sum control; considering the importance of tasks and giving priority to important tasks; adjusting various thresholds and models as needed; and recording detailed task assignment processes for subsequent analysis.

[0152] In another embodiment, preprocessing the acquired environment perception standard data includes:

[0153] Determine a preliminary processing result from the environmental perception standard data, and extract multiple data features from the preliminary processing result;

[0154] Determine whether the data features meet the preset integrity standards. If so, execute: further process the preliminary processing results, remove redundant data, and output the processed environmental perception standard data;

[0155] If not, the incomplete or abnormal data features are marked and removed from the environmental perception standard data;

[0156] Judge again whether the remaining data features in the environment perception standard data after exclusion meet the preset integrity standard. If they meet, perform further processing on the data features and output the final processing result.

[0157] If they do not meet, repeat the following: exclude incomplete or abnormal data from the environment perception standard data set until the data features meet the preset integrity standard or the size of the environment perception standard data reaches the preset minimum threshold.

[0158] The working principle of the above technical solution is as follows: determine the preliminary processing result from the environment perception standard data, and extract multiple data features from the preliminary processing result; judge whether the data features meet the preset integrity standard; judge whether the data features meet the preset integrity standard. If they meet, perform the following: perform further processing on the preliminary processing result, remove redundant data, and output the processed environment perception standard data; if they do not meet, mark the incomplete or abnormal data features and exclude the incomplete or abnormal data features from the environment perception standard data; judge again whether the remaining data features in the environment perception standard data after exclusion meet the preset integrity standard. If they meet, perform further processing on the data features and output the final processing result; if they do not meet, repeat the following: exclude incomplete or abnormal data from the environment perception standard data set until the data features meet the preset integrity standard or the size of the environment perception standard data reaches the preset minimum threshold.

[0159] The beneficial effects of the above technical solution are as follows: ensure the quality of the finally output data through multiple integrity checks and data cleaning. Automatically select different processing paths according to the integrity of the data features; can identify and exclude incomplete or abnormal data features, improving data reliability; remove redundant data when meeting the integrity standard, improving data efficiency; continuously optimize the data set through loop processing until the expected standard is reached; can adjust the integrity standard and the minimum data threshold as needed. While ensuring data quality, control the retention of sufficient data volume through the minimum threshold.

[0160] In another embodiment, generate multiple clustering results, including:

[0161] Obtain a data feature vector set, and the data feature vector set contains multiple data feature vectors;

[0162] Perform a preprocessing step on the data feature vector set, and the preprocessing step includes normalizing the data feature vectors.

[0163] According to a preset clustering algorithm, determine the similarity between each data feature vector and other data feature vectors, and divide the data feature vector set into multiple preliminary clustering results according to the similarity matrix;

[0164] According to a preset optimization strategy, perform optimization processing on the preliminary clustering results. The optimization processing includes readjusting the clustering boundary to improve the overall consistency of the clustering results and generate the final clustering results;

[0165] Judge whether the final clustering result meets the preset clustering evaluation criteria. If it meets, output the final clustering result.

[0166] The working principle of the above technical solution is as follows: Obtain a data feature vector set, which contains multiple data feature vectors; perform preprocessing on the data feature vector set, mainly including normalization processing. This step is very important as it can eliminate the dimensional differences between different features and make subsequent clustering more accurate; use a preset clustering algorithm to calculate the similarity between data feature vectors, and divide the data feature vector set into multiple preliminary clustering results according to the obtained similarity matrix. According to a preset optimization strategy, optimize the preliminary clustering results, which includes readjusting the clustering boundary to improve the overall consistency of the clustering results and finally generate the optimized clustering results.

[0167] Judge whether the final clustering result meets the preset clustering evaluation criteria. If it meets, output the final clustering result.

[0168] The beneficial effects of the above technical solution are as follows: Through normalization processing, ensure the comparability between different features; different clustering algorithms and optimization strategies can be selected according to needs; improve the quality of the clustering results through optimization processing; use preset evaluation criteria to ensure the reliability of the clustering results; can be applied to various types of data sets and clustering requirements; more preprocessing steps or optimization strategies can be added according to needs.

[0169] In another embodiment, generate multiple different-level perception feature sets, including:

[0170] Obtain a perception data set, which includes multiple perception data points;

[0171] According to a preset hierarchical processing rule, divide the perception data set into several preliminary perception levels. The hierarchical processing rule is determined based on the feature dimension, data density, and correlation of the perception data;

[0172] Perform hierarchical optimization on the preliminary perception levels. The hierarchical optimization includes adjusting the hierarchical boundary and merging similar levels to ensure the rationality of the hierarchical division and generate multiple optimized perception levels;

[0173] Based on the data characteristics of each perception level, a feature set of the perception level is extracted respectively, and the feature set is generated by a preset feature extraction algorithm;

[0174] The extracted perception feature sets are combined in hierarchical order to generate multiple perception feature sets of different levels.

[0175] The working principle of the above technical solution is as follows: First, collect a data set containing multiple perception data points; according to the preset hierarchical processing rules, divide the entire perception data set into several preliminary perception levels. These rules usually consider the feature dimensions of the perception data (such as color, shape, size, etc.), data density (the distribution of data points in space or time), and correlation (the degree of association between data points).

[0176] Optimize the preliminary perception levels, which includes adjusting the boundaries of the levels and merging similar levels. The purpose of optimization is to ensure the rationality of the level division, so that each level can more accurately reflect the internal structure and characteristics of the data. The optimized perception levels should better reflect the essential differences in the data. Based on each optimized perception level, use a preset feature extraction algorithm to extract the feature set of that level. These feature sets contain the key information of the data at that level and can represent the data characteristics of that level.

[0177] Combine the feature sets extracted from each level in hierarchical order to form multiple perception feature sets of different levels. These sets can be used for subsequent data analysis, pattern recognition, or decision-making.

[0178] Among them, according to the preset hierarchical processing rules, the perception data set is divided into several preliminary perception levels, including:

[0179] Classify the perception data into multiple dimension sets according to the feature dimensions;

[0180] Obtain the data density information in the perception data set, classify each dimension set according to the data density information, and use the classified data set as the preliminary perception level data;

[0181] According to the correlation of the preliminary perception level data, further subdivide the data, and store the subdivided data as the final perception level data;

[0182] Correspond each dimension set in each perception level data with the classified data density information one by one to generate the corresponding data of the perception level, and construct a hierarchical processing structure.

[0183] The beneficial effects of the above technical solution are as follows: Extract useful features from complex perception data, and these features are used for higher-level tasks such as classification, prediction, or control. Through layering and optimization, the efficiency and accuracy of feature extraction can be improved.

[0184] According to the preset hierarchical processing rules, divide the perception data set into several preliminary perception levels, including:

[0185] In another embodiment, obtaining the perception data set includes:

[0186] Obtain the request information of the perception data;

[0187] According to the perception data acquisition mechanism, identify whether the perception data meets the preset conditions, where the perception data acquisition mechanism is used to configure the perception data acquisition strategy;

[0188] When the perception data meets the preset conditions, based on the data association analysis mechanism, determine the relevance of the current perception data point;

[0189] When the perception data points are relevant, summarize the perception data points that meet the conditions to form a perception data set.

[0190] The working principle of the above technical solution is as follows: The system receives a request to obtain perception data. This usually involves specifying parameters such as data type, source, time range, etc. The system identifies whether the perception data meets the requirements according to the preset perception data acquisition strategy. This includes filtering out data that does not meet the quality standards, collecting only specific types of data, or restricting data collection according to time.

[0191] For the perception data that meets the preset conditions, the system uses the data association analysis mechanism to judge the relevance between these data points. This involves time series analysis, spatial analysis, or pattern recognition algorithms.

[0192] Through association analysis, the system determines which perception data points are relevant, that is, there is some connection or pattern between them; when it is determined that there is a correlation between the perception data points, the system aggregates these data points to form a perception data set, which can be used for further analysis or decision-making.

[0193] The beneficial effects of the above technical solution are as follows: Extract valuable information from a large amount of perception data. In this way, the system can process and utilize perception data more efficiently, thereby improving the quality and speed of decision-making.

[0194] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the same technology, the present invention is also intended to include these modifications and variations.

Claims

1. An environment perception system based on multi-agent interaction, characterized in that, Including: An environmental perception data acquisition unit, configured to acquire multi-agent environmental perception data, where the environmental perception data includes environmental characteristic parameters and agent perception parameters; A data feature vector construction unit, configured to acquire standard environmental perception data, perform noise elimination processing on the standard perception data, construct a data feature vector, and acquire standard association data from the standard environmental perception data; A perception analysis model construction unit, configured to cluster the data feature vector according to the standard association data to obtain a data group to be trained, and perform multi-agent interaction training on the initial model with the data group to be trained to obtain a multi-agent perception analysis model; A perception analysis unit, configured to perform hierarchical processing on the environmental perception data to obtain perception feedback data, and input the perception feedback data into the multi-agent perception analysis model to obtain a dynamic perception result; Obtaining the dynamic perception result includes: Performing hierarchical processing on the perception data to generate multiple perception feature sets at different levels; Extracting multiple key features from the perception feature sets based on a preset feature extraction algorithm, and generating perception feedback data; Inputting the perception feedback data into the multi-agent perception analysis model to trigger the multi-agent perception analysis model to process the perception feedback data; During the processing, analyzing results of each agent are obtained in real time, and multiple sub-level perception results are generated based on the analyzing results; Integrating the sub-level perception results to generate a preliminary dynamic perception result; Dynamically adjusting the preliminary dynamic perception result based on a preset dynamic adjustment rule to obtain a final dynamic perception result; Generating multiple perception feature sets at different levels includes: Obtaining a perception data set, where the perception data set includes multiple perception data points; Dividing the perception data set into several preliminary perception levels according to a preset hierarchical processing rule, where the hierarchical processing rule is determined based on the feature dimension, data density, and correlation of the perception data; Performing level optimization on the preliminary perception levels, where the level optimization includes adjusting level boundaries and merging similar levels to ensure the rationality of the level division, and generating multiple optimized perception levels; Respectively extracting a feature set of each perception level based on the data features of each perception level, where the feature set is generated by a preset feature extraction algorithm; Combining the extracted perception feature sets in the order of levels to generate multiple perception feature sets at different levels; Obtaining the perception data set includes: Obtaining request information of the perception data; Identifying whether the perception data meets a preset condition according to a perception data acquisition mechanism, where the perception data acquisition mechanism is used to configure a perception data acquisition strategy; When the perception data meets the preset condition, determining the correlation of the current perception data point based on a data association analysis mechanism; When the perception data point is relevant, summarizing the eligible perception data points to form a perception data set.

2. The environmental perception system based on multi-agent interaction according to claim 1, characterized in that, The environmental perception data acquisition unit includes: The sensing result acquisition module is used to control multiple agents to simultaneously start sensing operations in a preset environment, output initial environmental sensing data, and determine the sensing results of each agent regarding the current environmental sensing data. The sensing results include data collection accuracy, collection frequency, and corresponding environmental characteristic parameters. The environment perception data acquisition module is used to match and analyze the sensing results of each intelligent agent with the preset perception model, determine new environment perception tasks and assign them to the intelligent agent, and generate new environment perception data; A feature extraction module is used to extract features from new environmental perception data, obtain new environmental feature parameters and agent perception parameters, and determine multiple new environmental perception tasks until the rate of change of the environmental feature parameters is less than a preset threshold, or the current perception task is determined to be the last task; The perception task completion module is used to control multiple intelligent agents to perform perception operations in another preset environment, and repeatedly execute the steps from the initial environment perception data output to the current environment feature parameter change rate being less than the preset threshold until the perception tasks of all preset environments are completed.

3. An environment perception system based on multi-agent interaction according to claim 1, characterized in that, The unit for constructing data feature vector includes: A preprocessing operation module is used to preprocess the acquired environmental perception standard data to eliminate incomplete or abnormal data, input the preprocessed environmental perception standard data into the noise elimination module, perform noise elimination processing on the environmental perception standard data, and obtain denoised data; An initial data feature vector construction module is used to extract feature parameters from the denoised data according to preset rules to construct an initial data feature vector; Constructing a second data feature vector module, which is used to input the initial data feature vector into the feature screening module, and screen the initial data feature vector according to the set screening logic to obtain an optimized data feature vector; Construct a third data feature vector module, which is used to obtain associated parameters from the optimized data feature vector, obtain standard associated data associated with the optimized data feature vector in combination with the set logical relationship, input the standard associated data into the feature construction module, further process the standard associated data, and generate the final data feature vector.

4. The environmental perception system based on multi-agent interaction according to claim 1, characterized in that, Get multi-agent perception analysis models, including: Performing clustering processing on the data feature vectors to generate multiple clustering results; Determine a data group to be trained based on the clustering result, and output the data group to be trained; Input the data set to be trained into the initial model and perform multi-agent interactive training; During the multi-agent interactive training process, the perception data of each agent is obtained and analyzed in real time; Based on the perception data of each agent and the feedback data in the interactive training, update the parameters of the initial model and output the updated model parameters; Based on the updated model parameters, a multi-agent perception analysis model is generated and output; Verify the multi-agent perception analysis model and adjust the parameters of the multi-agent perception analysis model based on the verification results; In the adjusted multi-agent perception analysis model, multi-agent interaction training is performed again until the preset convergence conditions are reached, and the final multi-agent perception analysis model is output.

5. An environment perception system based on multi-agent interaction according to claim 2, characterized in that, Identify new environmental perception tasks and assign them to agents, including: Determine, based on the sensing result database, a plurality of second sensing results historically generated by the agent; extracting a plurality of first perception features of the second sensing result; Obtain a preset perception model library, match the first perception feature with the second perception feature in the perception model library, and if the match is consistent, use the matched second sensing result as the third sensing result, and at the same time, obtain the perception task corresponding to the matched second perception feature and associate it with the corresponding intelligent agent; Acquire the historical environment item corresponding to the third sensing result, and at the same time, acquire the sensing generation time corresponding to the third sensing result; Determine from the sensing result database a plurality of fourth sensing results corresponding to the historical environmental item and generated within a preset time period before and after the generation time; Inputting all fourth sensing results into a preset task analysis model, obtaining at least one new perception task output by the task analysis model, and accumulating and calculating the perception tasks associated with the intelligent agent to obtain a task sum; If the perception task is the first task and / or the perception task is the fourth task in the second task excluding the third task and / or the total of the tasks is greater than or equal to the preset total threshold of the first task, the agent perception task allocation fails; Otherwise, obtaining a preset environment perception task process record library, and determining a plurality of environment perception task process records corresponding to the perception task from the environment perception task process record library; Input all environmental perception task process records into a preset importance analysis model to obtain the importance of the importance analysis model output; Obtain a second task sum threshold corresponding to the importance; If the total of the tasks is greater than or equal to the second task total threshold, the agent perceives that the task assignment has failed; Otherwise, the agent perceives that the task assignment is successful.

6. An environment perception system based on multi-agent interaction according to claim 3, characterized in that Preprocess the acquired environmental perception standard data, including: Determine a preliminary processing result from the environmental perception standard data, and extract multiple data features from the preliminary processing result; Determine whether the data features meet the preset integrity standards. If so, execute: further process the preliminary processing results, remove redundant data, and output the processed environmental perception standard data; If not, the incomplete or abnormal data features are marked and removed from the environmental perception standard data; It is determined again whether the remaining data features in the environmental perception standard data after the elimination meet the preset integrity standard. If so, further processing is performed on the data features and the final processing result is output; If not, repeat the process: remove incomplete or abnormal data from the environment perception standard data set until the data features meet the preset integrity standards or the size of the environment perception standard data reaches the preset minimum threshold.

7. An environment perception system based on multi-agent interaction according to claim 4, characterized in that, Generates multiple clustering results, including: Obtaining a data feature vector set, where the data feature vector set includes multiple data feature vectors; Performing a preprocessing step on the data feature vector set, the preprocessing step including normalizing the data feature vector; According to a preset clustering algorithm, the similarity between each data feature vector and other data feature vectors is determined, and according to a similarity matrix, the data feature vector set is divided into a plurality of preliminary clustering results; According to the preset optimization strategy, the preliminary clustering result is optimized. The optimization process includes readjusting the clustering boundary to improve the overall consistency of the clustering result and generating the final clustering result; Determine whether the final clustering result meets the preset clustering evaluation criteria. If it meets, output the final clustering result.

Images