Terminal equipment control method and device, terminal equipment and storage medium
By acquiring sensing parameters from other terminal devices and performing agent reasoning to generate decision instructions, the problem of lack of deep collaboration among terminal devices in multi-agent systems is solved, and highly intelligent and robust collaborative decision-making in dynamic environments is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-15
- Publication Date
- 2026-03-10
AI Technical Summary
In existing technologies, terminal devices in multi-agent systems lack deep collaborative capabilities and are unable to make comprehensive judgments and autonomous decisions in dynamic environments, resulting in insufficient intelligence.
After receiving control commands, the system sends parameter acquisition requests to other terminal devices to obtain sensing parameters and inputs them into a pre-installed intelligent agent for reasoning. It then generates decision commands and optimizes behavioral strategies by combining reinforcement learning mechanisms, thereby enabling collaborative reasoning and autonomous decision-making by the terminal devices.
It improves the intelligence and robustness of terminal devices, enabling deep collaboration in dynamic environments, adapting to complex task requirements, and enhancing the scalability and efficiency of the system.
Smart Images
Figure CN121645159A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application belongs to the technical field of communication, and particularly relates to a terminal device control method and device, a terminal device, and a storage medium. BACKGROUND
[0002] With the rapid development of artificial intelligence and Internet of Things technology, multi-agent systems are increasingly widely used in cross-terminal device collaboration. Related technologies mainly achieve instruction transmission and response through a preset message passing mechanism. For example, in a smart home scenario, control instructions are unidirectionally transmitted and executed between terminal devices through a wireless network. However, this approach can only complete simple instruction responses and cannot support multi-agents to make collaborative reasoning and autonomous decisions based on dynamic environments and overall goals, resulting in a lack of deep collaboration capabilities between terminal devices in multi-agent systems and difficulty in dealing with tasks that require comprehensive judgment of dynamic factors. SUMMARY
[0003] Embodiments of the present application provide a terminal device control method and device, a terminal device, and a storage medium to address the technical problem of a lack of deep collaboration capabilities between terminal devices in related technologies and low intelligence.
[0004] In a first aspect, embodiments of the present application provide a terminal device control method. A multi-agent system includes a plurality of communication-connected terminal devices. The terminal device control method is applied to a first terminal device in the plurality of terminal devices. The terminal device control method includes: receiving a first control instruction and sending a parameter acquisition request to at least one second terminal device according to the first control instruction, the second terminal device being a terminal device other than the first terminal device in the plurality of terminal devices; receiving sensing parameters fed back by the second terminal device according to the parameter acquisition request, the sensing parameters being parameters generated by a sensor of the second terminal device; calling a pre-installed agent and inputting the sensing parameters fed back by the at least one second terminal device and the first control instruction into the pre-installed agent for reasoning to generate a decision instruction; running according to the decision instruction.
[0005] In some embodiments, calling the pre-installed agent and inputting the sensing parameters fed back by the at least one second terminal device and the first control instruction into the pre-installed agent for reasoning to generate a decision instruction includes: receiving second working state parameters fed back by the second terminal device according to the parameter acquisition request; calling the pre-installed agent and inputting the sensing parameters fed back by the at least one second terminal device, the first control instruction, and the second working state parameters into the pre-installed agent for reasoning to generate a decision instruction.
[0006] In some embodiments, after running according to the decision instruction, the method comprises: generating a first working state parameter according to the working state of the first terminal device, and sending the first working state parameter to the second terminal device, so that the second terminal device adjusts the running state of the second terminal device according to the first working state parameter.
[0007] In some embodiments, after running according to the decision instruction, the method comprises: generating a first working state parameter according to the working state of the first terminal device; calculating a reward value according to the first working state parameter and a preset reward function; adjusting the calculation parameters of the agent according to the reward value.
[0008] In some embodiments, the control method of the terminal device further comprises: receiving a third working state parameter sent by the second terminal device, the third working state parameter being used to represent the working state of the second terminal device; calling a pre-installed agent and inputting the third working state parameter into the pre-installed agent for reasoning to generate an adjustment instruction; running according to the adjustment instruction.
[0009] In some embodiments, the control method of the terminal device further comprises: receiving an identification and a sensing parameter type broadcasted by the second terminal device, the sensing parameter type being generated by the second terminal device according to data collected by a sensor arranged in the second terminal device; updating the cooperative rule database according to the identification and the sensing parameter type; receiving a first control instruction and generating a parameter acquisition request according to the first control instruction, comprising: receiving a first control instruction and generating a necessary parameter type according to the first control instruction; determining a sensing parameter type matched with the necessary parameter type in the cooperative rule database, and determining an identification corresponding to the matched sensing parameter type; sending a parameter acquisition request to the second terminal device corresponding to the matched sensing parameter type.
[0010] In some embodiments, when the type of the first sensing parameter collected by the sensor of the at least one first terminal device matches at least one necessary parameter type, the pre-installed agent is called and the sensing parameter fed back by the at least one second terminal device and the first control instruction are input into the pre-installed agent for reasoning to generate a decision instruction, comprising: The cosine similarity between the second sensing parameters fed back by multiple second terminal devices and the first sensing parameters is calculated. The sensing parameters fed back by the second terminal devices include the second sensing parameters, and the type of the second sensing parameters is the same as that of the first sensing parameters. Based on the cosine similarity, assign corresponding weights to the second sensing parameter; The pre-installed intelligent agent is invoked, and the sensing parameters and corresponding weights fed back by the second terminal device, the sensing parameters collected by the sensors of the first terminal device, and the first control command are input into the pre-installed intelligent agent for inference to generate decision commands.
[0011] Secondly, embodiments of this application provide a control device, the control device comprising: The sending module is used to receive a first control command and send a parameter acquisition request to at least one second terminal device according to the first control command. The second terminal device is a terminal device other than the first terminal device among multiple signal-connected terminal devices in the multi-agent system. The receiving module is used to receive the sensing parameters requested and fed back by the second terminal device based on the parameters. The sensing parameters are the parameters collected and generated by the sensors of the second terminal device. The generation module is used to call the pre-installed intelligent agent and input the sensing parameters fed back by at least one second terminal device and the first control command into the pre-installed intelligent agent for inference, and generate decision commands; The execution module is used to run according to decision instructions.
[0012] Thirdly, embodiments of this application provide a terminal device, which includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the control method of the terminal device as described above.
[0013] Fourthly, embodiments of this application provide a computer storage medium storing computer program instructions, which, when executed by a processor, implement the control method for the terminal device as described above.
[0014] The control method, apparatus, terminal device, and storage medium for the terminal device provided in this application embodiment obtain sensor parameters collected by sensors of other terminal devices by sending a parameter acquisition request to at least one second terminal device according to a first control instruction; generating decision instructions by invoking a pre-installed intelligent agent and inputting the sensor parameters fed back by at least one second terminal device and the first control instruction into the pre-installed intelligent agent for reasoning; and operating according to the decision instructions, so that the operation of the first terminal device not only refers to the first control instruction but also to the sensor parameters collected by the second terminal devices, so that the operation generated by the pre-installed intelligent agent coordinates with other terminal devices, thereby improving the intelligence of the operation of the first terminal device. Attached Figure Description
[0015] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] Figure 1 This is a schematic flowchart of a control method for a terminal device provided in an embodiment of this application; Figure 2 This is a schematic diagram of the architecture of a terminal device provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a control device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the hardware structure of a terminal device provided in an embodiment of this application. Detailed Implementation
[0017] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples of this application.
[0018] It should be noted that, in this document, relational terms such as "second" and "third" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.
[0019] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The embodiments will now be described in detail with reference to the accompanying drawings.
[0020] To address the problems of the prior art, embodiments of this application provide a control method, apparatus, terminal device, and storage medium for a terminal device. The control method for the terminal device provided in this application embodiment will be described first below.
[0021] Figure 1 A flowchart illustrating a control method for a terminal device according to an embodiment of this application is shown. The multi-agent system includes multiple communicatively connected terminal devices, and the control method for the terminal devices is applied to a first terminal device among the multiple terminal devices; the control method for the terminal devices includes: S110, receive the first control instruction, and send a parameter acquisition request to at least one second terminal device according to the first control instruction, wherein the second terminal device is a terminal device other than the first terminal device among multiple terminal devices; S120, Receive sensing parameters from the second terminal device according to the parameter acquisition request feedback, the sensing parameters are the parameters collected and generated by the sensor of the second terminal device; S130, invoke the pre-installed intelligent agent and input the sensing parameters fed back by at least one second terminal device and the first control command into the pre-installed intelligent agent for reasoning, and generate decision commands; S140 operates according to the decision-making instructions.
[0022] The control method for the terminal device provided in this application can be applied to a first terminal device. The first terminal device and at least one second terminal device are communicatively connected to form a multi-agent system, and intelligent agents are deployed in both the first and second terminal devices. The terminal device can specifically be a smart home appliance, vehicle, mobile terminal, desktop computer, laptop computer, etc. Smart home appliances can be smart TVs, smart air conditioners, smart refrigerators, smart curtain motors, smart speakers, etc. Multiple terminal devices in the same intelligent agent system can establish connections based on the same communication protocol or based on different communication protocols, including but not limited to HTTP, TCP / IP, Bluetooth, WiFi Direct, etc.
[0023] The first control command can be sent by the user to the first electronic device through operating the electronic device they hold, or it can be generated by the user directly operating buttons, touch screens, etc. on the first electronic device.
[0024] For example, the first terminal device is a smart desk lamp. The user issues a voice command to the smart speaker to "turn on the desk lamp". The smart speaker issues a first control command to the smart desk lamp to instruct the smart desk lamp to turn on, based on the voice command.
[0025] The first terminal device can send a parameter acquisition request to one second terminal device in the same multi-agent system according to the first control command, or it can send a parameter acquisition request to multiple second terminal devices in the same multi-agent system according to the first control command. The parameters to be sent to which second terminal devices can be determined based on the distance, topological relationship, and scene linkage relationship between each second terminal device and the first terminal device in the multi-agent system.
[0026] After receiving a parameter acquisition request, the second terminal device feeds back the sensor parameters it has collected to the first terminal device according to the request. The second terminal device is equipped with one or more sensors that collect and generate sensor parameters. These sensors include, but are not limited to, temperature sensors, humidity sensors, light sensors, proximity sensors, cameras, gyroscopes, and position sensors. Sensor parameters include, but are not limited to, temperature, humidity, light intensity, distance, image, attitude, heading, and geographical location.
[0027] The pre-installed intelligent agent is an agent pre-configured in the first terminal device. This agent possesses a knowledge base, inference parameters, etc., and can perform function calculations based on input information and output information. In this embodiment, sensing parameters and a first control command fed back from at least one second terminal device are input into the pre-installed intelligent agent for inference, and the pre-installed intelligent agent outputs a decision command. The quality of this decision is obtained based on the sensing parameters and the first control command fed back from the second terminal device. The pre-installed intelligent agent can be an agent based on algorithms such as Q-learning and Deep Q Network. Agents deployed in different intelligent terminals can be trained independently and can use different algorithm models.
[0028] For example, the first terminal device is a smart desk lamp. The user issues a voice command to the smart speaker to "turn on the desk lamp." The smart speaker then issues a first control command to the smart desk lamp to instruct it to turn on, based on the voice command. The smart desk lamp sends parameter acquisition requests to the smart TV and the smart curtain motor. The smart TV feeds back the light intensity generated by its light sensor to the smart desk lamp, and the smart curtain motor feeds back the light intensity generated by its light sensor to the smart desk lamp. The smart desk lamp inputs the first control command and the two light intensities into a preset intelligent agent. The preset intelligent agent outputs a decision command instructing the smart desk lamp to turn on and emit light with a color temperature of 3000K. The smart desk lamp then turns on and emits light with a color temperature of 3000K based on the decision command.
[0029] In the control method for the terminal device provided in this application, a parameter acquisition request is sent to at least one second terminal device according to a first control instruction to obtain sensing parameters collected by the sensors of other terminal devices; a decision instruction is generated by invoking a pre-installed intelligent agent and inputting the sensing parameters fed back by at least one second terminal device and the first control instruction into the pre-installed intelligent agent for reasoning; and the device operates according to the decision instruction, so that the operation of the first terminal device not only refers to the first control instruction but also to the sensing parameters collected by the second terminal devices, so that the operation generated by the pre-installed intelligent agent coordinates with other terminal devices, thereby improving the intelligence of the operation of the first terminal device.
[0030] In some embodiments, when a terminal device joins a multi-agent system for the first time, the user needs to assist in configuring the terminal device to enable communication connections between the terminal device and other terminal devices, establish topology relationships, and set geographical locations.
[0031] When the operating system of a terminal device starts, the intelligent agent reads a preset configuration file from the terminal device's memory. This file includes the agent's initial knowledge, inference rules, and communication protocol parameters. The agent initializes its own knowledge base according to the preset configuration file, loading the initial knowledge into the memory of the first terminal device; it starts the inference engine, sets inference parameters, including but not limited to inference depth and inference time limits; and it opens the communication interface, establishing connections with intelligent agents on other terminal devices according to a preset communication protocol to ensure data transmission and information exchange.
[0032] One or more sensors in the terminal device acquire environmental information, location information, attitude information, and other sensory information. The collected sensory information is preprocessed, including but not limited to data filtering and normalization, to improve data quality and usability. The preprocessed sensory information is stored in the agent's cache for subsequent collaborative inference.
[0033] Optionally, S140 also includes: The pre-installed intelligent agent is invoked, and the sensing parameters generated by the sensors of the first terminal device, the sensing parameters fed back by at least one second terminal device, and the first control command are input into the pre-installed intelligent agent for inference to generate decision commands.
[0034] In some embodiments, S140 includes: S210, Receive the second working status parameter requested by the second terminal device based on the parameter acquisition request; S220, invoke the pre-installed intelligent agent, input the sensing parameters fed back by at least one second terminal device, the first control command, and the second working state parameters into the pre-installed intelligent agent for inference, and generate decision commands.
[0035] The second terminal device not only receives feedback sensing parameters based on parameter requests, but also feeds back its own second operating status parameters. When the second terminal device is an air conditioner, the second operating status parameters include, but are not limited to, power on / off status, operating mode (cooling / heating / fan), fan speed, target temperature, and timer power on / off settings. When the second terminal device is a vehicle, the second operating status parameters include, but are not limited to, vehicle start / stop status, driving speed, gear information, engine speed, and remaining range. When the second terminal device is a smartwatch, the second operating status parameters include, but are not limited to, screen on / off status, current application process, sports mode, remaining battery power, storage space usage, Bluetooth connection status, alarm settings, health monitoring switch, and message notification methods. When the second terminal device is a smart machine tool, the second operating status parameters include, but are not limited to, start / stop status, spindle speed, machining progress, current machining instruction code, tool wear value, lubricating oil level, motor current and voltage, and downtime due to faults.
[0036] The first terminal device inputs the sensing parameters, the first control command, and the second working state parameters into the pre-installed intelligent agent for reasoning, so that the generated decision command takes into account the sensing parameters, the first control command, and the second working state parameters. This enables multiple intelligent agents to conduct collaborative reasoning and autonomous decision-making based on the dynamic environment and the overall goal, thereby enabling deep collaboration among the terminal devices and improving the intelligence of the terminal devices in comprehensively judging dynamic factors.
[0037] In this embodiment, the agents on each terminal device communicate and perform collaborative reasoning directly, eliminating the need for a central node. Each agent can interact with other agents based on the sensing parameters generated by its own terminal device to make reasoning decisions, thus improving the robustness and scalability of the system.
[0038] In some embodiments, after S140, the following is included: S310, generate a first working status parameter based on the working status of the first terminal device, and send the first working status parameter to the second terminal device so that the second terminal device can adjust its operating status according to the first working status parameter.
[0039] The first operating status parameter is generated by the first terminal device after executing the decision instruction. This parameter represents the state of the first terminal device after executing the instruction. The first operating status parameter is sent to the second terminal device so that they can subsequently coordinate adjustments. The second terminal device can also adjust its operating state based on the first operating status parameter to adapt its operating state to that of the first terminal device.
[0040] In some embodiments, after S140, the following is included: S410, Generate first working status parameters based on the working status of the first terminal device; S420, calculate the reward value based on the first working state parameters and the preset reward function; S430 adjusts the agent's calculation parameters based on the reward value.
[0041] This embodiment introduces a reinforcement learning mechanism, enabling each agent in the multi-agent system to continuously learn and optimize its own behavioral strategies during cross-terminal communication and collaboration.
[0042] Intelligent agents can model the collaborative environment based on parameters collected by terminal devices and user input, thus obtaining an environment model. For example, in an intelligent logistics scenario, the agent models the collaborative environment based on a state space, an action space, and a reward function. The state space can include information such as the location of goods, the status of transport vehicles, and traffic conditions; the action space can include operations such as vehicle direction and speed adjustments; the reward function can be defined based on factors such as the efficiency and cost of goods transportation, such as a positive reward for timely delivery and a negative reward for delays or accidents.
[0043] The agent stores the environment model in the memory of the terminal device for use in the subsequent reinforcement learning process.
[0044] Each time, an intelligent agent selects a decision instruction from the action space based on input data and random or empirical strategies. This instruction is then sent to other intelligent agents via a communication interface, allowing them to understand the agent's intentions and coordinate adjustments. For example, a vehicle agent might select its direction and speed based on current traffic conditions and cargo transport needs.
[0045] After the intelligent terminal executes a decision instruction, the agent calculates and stores the reward value according to a predefined reward function for subsequent policy updates. After executing a preset number of decision instructions or reaching a preset time period, the agent updates its computational parameters based on a reinforcement learning algorithm and the reward value. Through continuous iterative learning, the agent's computational parameters are gradually optimized, enabling it to make more reasonable decisions in cross-device communication and collaboration environments, thus improving the overall performance and efficiency of the system.
[0046] For example, the agent is an agent using the Q-learning algorithm. The agent updates the Q-value of the current state-action pair based on the current reward value and the maximum Q-value of the next state.
[0047] In some embodiments, the control method for the terminal device further includes: S510, receive the third working status parameter sent by the second terminal device, the third working status parameter is used to characterize the working status of the second terminal device; S520 calls the pre-installed intelligent agent and inputs the third working state parameter into the pre-installed intelligent agent for inference, generating adjustment instructions; S530 operates according to the adjustment instructions.
[0048] Because the terminal devices in a multi-agent system work collaboratively, the second terminal device can send a third working status parameter to the first terminal device when its working status parameters change or at preset intervals. The first terminal device then invokes a pre-installed agent and inputs the third working status parameter into the pre-installed agent for inference, generates an adjustment command, and runs according to the adjustment command to adapt the working status of the first terminal device to that of the second terminal device.
[0049] For example, the first terminal device is a smart TV, and the second terminal device is a smart curtain motor. After the second terminal device closes the curtains, it sends a third working state parameter indicating that the curtains are closed. The first terminal device then invokes a pre-installed intelligent agent and inputs the third working state parameter into the pre-installed intelligent agent for inference, generating an adjustment command to adjust the brightness. The first terminal device then adjusts the display brightness according to the adjustment command.
[0050] During operation, the intelligent agent detects updates to the working status of other intelligent terminals through a communication interface. When a change in the working status of other intelligent terminals is detected, the agent adjusts its collaborative behavior. Through continuous collaborative adjustments, the multi-agent system forms a coordinated and efficient cross-terminal communication and collaboration system to adapt to dynamically changing environments and task requirements.
[0051] In some embodiments, the control method for the terminal device further includes: S610, receive the identification mark and sensing parameter type broadcast by the second terminal device, wherein the sensing parameter type is generated by the second terminal device based on the data collected by the sensors installed in the second terminal device; S620 updates the collaborative rule database based on the identification identifier and sensing parameter type; S110 includes: S630 receives the first control command and generates the required parameter types according to the first control command; S640, determine the sensing parameter type that matches the required parameter type in the collaborative rule database, and determine the identification identifier corresponding to the matched sensing parameter type; S650, a parameter acquisition request is sent to the second terminal device corresponding to the identification tag that matches the sensing parameter type.
[0052] When a second terminal device joins a multi-agent system, it can broadcast its own identification identifier and sensing parameter type. The sensing parameter type reveals the types of data that the second terminal device's sensors can collect. The first terminal device updates the collaboration rule database based on the identification identifier and sensing parameter type, i.e., it stores the identification identifier and sensing parameter type in the collaboration rule database.
[0053] The system receives a first control command and determines the required parameter types to be input into the intelligent agent based on the command. The collaborative rule database stores the identification identifiers of each terminal device and their corresponding sensing parameter types. It queries the collaborative rule database for sensing parameter types that match the required parameter types and determines the identification identifiers corresponding to the matching sensing parameter types. The query results may include multiple identification identifiers, multiple sensing parameter types corresponding to one identification identifier, or the same sensing parameter type corresponding to multiple identification identifiers.
[0054] When a matching identifier and one or more sensing parameter types corresponding to that identifier are obtained, a parameter acquisition request is sent to the second terminal device corresponding to that identifier. When multiple matching identifiers and one or more sensing parameter types corresponding to that identifier are obtained, parameter acquisition requests are sent to the second terminal devices corresponding to each of the multiple identifiers to obtain sensing parameters fed back by the multiple second terminal devices. The type of the sensing parameter is adapted to the required parameter type.
[0055] In some embodiments, in S640, if a sensing parameter type matching the required parameter type is found in the collaborative rule database, the pre-installed agent is invoked and the first control instruction is input into the pre-installed agent for inference to generate a decision instruction.
[0056] In some embodiments, if the type of the first sensing parameter collected by the sensor of at least one first terminal device matches at least one required parameter type, S130 includes: S710, calculate the cosine similarity between the second sensing parameters fed back by multiple second terminal devices and the first sensing parameters respectively. The sensing parameters fed back by the second terminal devices include the second sensing parameters, and the type of the second sensing parameters is the same as the type of the first sensing parameters. S720, based on cosine similarity, assigns corresponding weights to the second sensing parameter; S730 invokes the pre-installed intelligent agent, inputs the sensing parameters and corresponding weights fed back by the second terminal device, the sensing parameters collected by the sensors of the first terminal device, and the first control command into the pre-installed intelligent agent for inference, and generates decision commands.
[0057] When the type of the first sensing parameter collected by the sensor of at least one first terminal device matches the type of at least one required parameter, it indicates that the first terminal device itself has the ability to collect parameters that match the required parameter type. Even without obtaining the sensing parameters fed back by the second terminal device, the first terminal device can also generate a decision command based on the first control command and the first sensing parameter.
[0058] A first terminal device sends parameter acquisition requests to multiple second terminal devices. The second terminal devices then return sensing parameters, including second sensing parameters of the same type as the first sensing parameters. When acquiring multiple second sensing parameters and first sensing parameters of the same type, the first terminal device can measure their similarity by calculating the cosine of the angle between the first and second sensing parameters; that is, by calculating cosine similarity. Then, a weight is assigned to each second sensing parameter according to the rule that higher similarity corresponds to higher weight. By adjusting the sensing parameters input to the intelligent agent with different weights, the influence of the same type of sensing parameters collected by different intelligent devices on the decision-making instructions can be adjusted.
[0059] Please see Figure 2 In some embodiments, the multi-agent system can adopt a layered architecture, including a perception layer, a communication layer, a collaborative reasoning layer, and an execution layer. The perception layer is responsible for collecting environmental information, the communication layer enables information transmission between agents, the collaborative reasoning layer performs collaborative decision-making, and the execution layer is responsible for executing decision-making operations. The layers interact with each other through interfaces.
[0060] The perception layer comprises sensors distributed across various terminal devices. These sensors are responsible for collecting sensing parameters and operational status parameters. The sensors convert the collected physical signals into electrical signals, which are then converted into digital signals via an analog-to-digital converter. Preliminary processing, such as filtering and amplification, is performed on the digital signals to improve their quality. The processed sensing and operational status parameters are then encapsulated according to a preset format and sent to the communication layer. Data transmission between the perception and communication layers can be achieved via a bus deployed within the terminal devices.
[0061] The communication layer is responsible for information transmission between intelligent agents. It supports multiple communication protocols, such as WiFi, Bluetooth, and 4G / 5G, to adapt to different network environments and communication needs. The communication layer also handles message encapsulation, decapsulation, routing, and forwarding. When the communication layer receives data from the perception layer, it encapsulates the data according to a preset message format, adds a message header (which may include message type, sender identifier, and receiver identifier), and selects the appropriate communication protocol and routing path to send the message based on the receiver identifier. When receiving messages from other terminal devices, it decapsulates the messages, extracts the data content, and sends it to the collaborative inference layer. The communication layer and the collaborative inference layer interact through a software interface. Communication layers on different terminal devices are connected via a communication network, enabling cross-device communication between intelligent agents on different terminal devices.
[0062] The collaborative reasoning layer is responsible for collaborative reasoning and decision-making based on information transmitted from the communication layer. It includes a knowledge base, an inference engine, and a collaborative rule base. The knowledge base stores the agents' domain knowledge and experience; the inference engine uses inference algorithms to analyze and process information; and the collaborative rule base defines the collaborative methods and rules between agents. After receiving information from the communication layer, the collaborative reasoning layer stores it in the terminal device's cache. The inference engine reads relevant knowledge from the knowledge base and, combined with rules from the collaborative rule base, analyzes and infers the information in the cache. During the inference process, multiple iterative calculations can be performed, continuously updating the inference results. Finally, a decision suggestion is output and sent to the execution layer. The collaborative reasoning layer and the execution layer are connected through a control interface.
[0063] The execution layer is responsible for executing corresponding operations based on the decision suggestions sent by the collaborative inference layer. It controls and operates the device by calling the terminal device's control interface. After receiving the decision suggestions from the collaborative inference layer, the execution layer parses them, extracting the operation instructions and parameters. Based on the operation instructions, it calls the terminal device's control interface, sending corresponding control signals to implement the device's operation. Simultaneously, the execution layer monitors the terminal device's operating status in real time and feeds it back to the collaborative inference layer for subsequent collaborative adjustments. The execution layer connects to the terminal device through hardware interfaces such as relay interfaces and PWM interfaces.
[0064] Based on the control method for the terminal device provided in the above embodiments, this application also provides specific implementations of the control device. Please refer to the following embodiments.
[0065] See Figure 3 This application provides a control device 100, which includes: The sending module 11 is used to receive a first control command and send a parameter acquisition request to at least one second terminal device according to the first control command. The second terminal device is a terminal device other than the first terminal device among multiple signal-connected terminal devices in the multi-agent system. The receiving module 12 is used to receive the sensing parameters requested and fed back by the second terminal device based on the parameters. The sensing parameters are the parameters collected and generated by the sensors of the second terminal device. The generation module 13 is used to call the pre-installed intelligent agent and input the sensing parameters and first control instructions fed back by at least one second terminal device into the pre-installed intelligent agent for reasoning, and generate decision instructions; Module 14 is used to run according to decision instructions.
[0066] In some embodiments, the generation module 13 is further configured to: Receive the second working status parameters requested and fed back by the second terminal device based on the parameters; The pre-installed intelligent agent is invoked, and the sensing parameters fed back by at least one second terminal device, the first control command, and the second working state parameters are input into the pre-installed intelligent agent for inference to generate decision commands.
[0067] In some embodiments, the sending module 11 is further configured to: A first operating status parameter is generated based on the operating status of the first terminal device, and the first operating status parameter is sent to the second terminal device so that the second terminal device adjusts its operating status according to the first operating status parameter.
[0068] In some embodiments, the control device 100 further includes an iteration module, which is configured to: generate a first working state parameter based on the working state of the first terminal device; The reward value is calculated based on the first working state parameters and the preset reward function; The agent's calculation parameters are adjusted based on the reward value.
[0069] In some embodiments, the receiving module 12 is further configured to receive a third working status parameter sent by the second terminal device, the third working status parameter being used to characterize the working status of the second terminal device. The generation module 13 is also used to call the pre-installed intelligent agent and input the third working state parameter into the pre-installed intelligent agent for inference, and generate adjustment instructions; The running module 14 is also used to run according to the adjustment instructions.
[0070] In some embodiments, the receiving module 12 is further configured to receive the identification mark and sensing parameter type broadcast by the second terminal device, wherein the sensing parameter type is generated by the second terminal device based on the data collected by the sensors installed in the second terminal device; The iteration module is also used to update the collaborative rule database based on the identification identifier and sensing parameter type; The sending module 11 is also used for: Receive the first control command and generate the required parameter types according to the first control command; In the collaborative rule database, identify the sensing parameter type that matches the required parameter type, and determine the identification identifier corresponding to the matched sensing parameter type; Send a parameter acquisition request to the second terminal device corresponding to the identification identifier that matches the sensing parameter type.
[0071] In some embodiments, where the type of the first sensing parameter collected by the sensor of at least one first terminal device matches at least one required parameter type, the generation module 13 is further configured to: The cosine similarity between the second sensing parameters fed back by multiple second terminal devices and the first sensing parameters is calculated. The sensing parameters fed back by the second terminal devices include the second sensing parameters, and the type of the second sensing parameters is the same as that of the first sensing parameters. Based on the cosine similarity, assign corresponding weights to the second sensing parameter; The pre-installed intelligent agent is invoked, and the sensing parameters and corresponding weights fed back by the second terminal device, the sensing parameters collected by the sensors of the first terminal device, and the first control command are input into the pre-installed intelligent agent for inference to generate decision commands.
[0072] The control device 100 provided in this embodiment of the invention can implement the various steps in the above method embodiments, and will not be described again here to avoid repetition.
[0073] Figure 4 A schematic diagram of the hardware structure of the terminal device provided in an embodiment of this application is shown.
[0074] The terminal device may include a processor 1001 and a memory 1002 storing computer program instructions.
[0075] Specifically, the processor 1001 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.
[0076] Memory 1002 may include mass storage for data or instructions. For example, and not limitingly, memory 1002 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 1002 may include removable or non-removable (or fixed) media. Where appropriate, memory 1002 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 1002 is non-volatile solid-state memory.
[0077] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the methods according to one aspect of this disclosure.
[0078] The processor 1001 reads and executes computer program instructions stored in the memory 1002 to implement any of the terminal device control methods in the above embodiments.
[0079] In one example, the terminal device may also include a communication interface 1003 and a bus 1010. The processor 1001, memory 1002, and communication interface 1003 are connected via the bus 1010 and communicate with each other.
[0080] The communication interface 1003 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.
[0081] Bus 1010 includes hardware, software, or both, that couples components of an end device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 1010 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.
[0082] The terminal device can be based on the above embodiments to implement the above-described terminal device control method and apparatus.
[0083] Furthermore, in conjunction with the control methods for terminal devices in the above embodiments, this application embodiment can provide a computer storage medium for implementation. This computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the control methods for terminal devices in the above embodiments and achieve the same technical effect. To avoid repetition, further details are omitted here. The aforementioned computer-readable storage medium may include non-transitory computer-readable storage media, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, etc., and is not limited thereto.
[0084] In addition, this application also provides a computer program product, including computer program instructions, which, when executed by a processor, can implement the steps and corresponding content of the aforementioned method embodiments.
[0085] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.
[0086] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.
[0087] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.
[0088] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.
[0089] The above are merely specific embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.
Claims
1. A control method of a terminal device, characterized by, The multi-agent system comprises a plurality of terminal devices connected in communication, and a control method of the terminal device is applied to a first terminal device of the plurality of terminal devices; The control method of the terminal device comprises: receiving a first control instruction, and sending a parameter acquisition request to at least one second terminal device according to the first control instruction, the second terminal device being a terminal device other than the first terminal device in the plurality of terminal devices; receiving a sensing parameter fed back by the second terminal device according to the parameter acquisition request, the sensing parameter being a parameter generated by a sensor of the second terminal device; calling a pre-installed agent and inputting the sensing parameter fed back by the at least one second terminal device and the first control instruction into the pre-installed agent for reasoning to generate a decision instruction; running according to the decision instruction.
2. The control method of the terminal device according to claim 1, characterized by, The calling of the pre-installed agent and the inputting of the sensing parameter fed back by the at least one second terminal device and the first control instruction into the pre-installed agent for reasoning to generate a decision instruction comprises: receiving a second working state parameter fed back by the second terminal device according to the parameter acquisition request; calling a pre-installed agent and inputting the sensing parameter fed back by the at least one second terminal device, the first control instruction and the second working state parameter into the pre-installed agent for reasoning to generate a decision instruction.
3. The control method of the terminal device according to claim 1, characterized by, After the running according to the decision instruction, the method comprises: generating a first working state parameter according to a working state of the first terminal device, and sending the first working state parameter to the second terminal device, so that the second terminal device adjusts a running state of the second terminal device according to the first working state parameter.
4. The control method of the terminal device according to claim 1, characterized by, After the running according to the decision instruction, the method comprises: generating a first working state parameter according to a working state of the first terminal device; calculating a reward value according to the first working state parameter and a preset reward function; adjusting a calculation parameter of the agent according to the reward value.
5. The control method of the terminal device according to claim 1, characterized by, The control method of the terminal device further comprises: receiving a third working state parameter sent by the second terminal device, the third working state parameter being used to represent a working state of the second terminal device; calling a pre-installed agent and inputting the third working state parameter into the pre-installed agent for reasoning to generate an adjustment instruction; running according to the adjustment instruction.
6. The control method of the terminal device according to claim 1, characterized by The control method of the terminal device further comprises: receiving an identification and a sensing parameter type broadcast by the second terminal device, the sensing parameter type being generated by the second terminal device according to data collected by a sensor arranged in the second terminal device; updating a cooperative rule database according to the identification and the sensing parameter type; The receiving of the first control instruction and the sending of the parameter acquisition request to at least one second terminal device according to the first control instruction comprises: receiving a first control instruction, and generating a necessary parameter type according to the first control instruction; determining the sensing parameter type matched with the necessary parameter type in the cooperative rule database, and determining an identification corresponding to the matched sensing parameter type; The second terminal device corresponding to the identification corresponding to the matched sensing parameter type is sent a parameter acquisition request.
7. The control method of the terminal device according to claim 6, characterized by, In a case where a type of a first sensing parameter collected by a sensor of the at least one first terminal device matches the at least one necessary parameter type, the pre-installed agent is invoked, and the sensing parameter fed back by the at least one second terminal device and the first control instruction are input into the pre-installed agent for reasoning to generate a decision instruction, including: A cosine similarity is calculated between a second sensing parameter fed back by each of the second terminal devices and the first sensing parameter, the sensing parameter fed back by the second terminal device including the second sensing parameter, and a type of the second sensing parameter being the same as that of the first sensing parameter; According to the cosine similarity, a corresponding weight is configured for the second sensing parameter; The pre-installed agent is invoked, and the sensing parameter fed back by the second terminal device and the corresponding weight, the sensing parameter collected by the sensor of the first terminal device, and the first control instruction are input into the pre-installed agent for reasoning to generate a decision instruction.
8. A control device characterized by comprising: The control device includes: The sending module is configured to receive a first control instruction and send a parameter acquisition request to at least one second terminal device according to the first control instruction, the second terminal device being a terminal device other than the first terminal device among a plurality of signal-connected terminal devices in a multi-agent system; The receiving module is configured to receive a sensing parameter fed back by the second terminal device according to the parameter acquisition request, the sensing parameter being a parameter generated by a sensor of the second terminal device; The generating module is configured to invoke a pre-installed agent and input the sensing parameter fed back by the at least one second terminal device and the first control instruction into the pre-installed agent for reasoning to generate a decision instruction; The running module is configured to run according to the decision instruction.
9. A terminal device, comprising: The terminal device includes a processor and a memory storing computer program instructions; The processor implements the control method of the terminal device according to any one of claims 1-7 when executing the computer program instructions.
10. A computer storage medium, characterized in that The computer storage medium stores computer program instructions, and the computer program instructions are executed by the processor to implement the control method of the terminal device according to any one of claims 1-7.