Distributed real-time control method and device for service robot

Abstract

The invention provides a service robot distributed real-time control method and device, and the method comprises the steps: an edge calculation module of a base station receives environment data of a service robot, transmits the environment data to a local data center, triggers a highest priority instruction when the environment data exceeds a first threshold condition, and transmits the highest priority instruction to the service robot; the local data center carries out advanced analysis and future state prediction on the environment data to obtain an analysis result, generates a control command when the analysis result exceeds a second threshold condition, carries out priority judgment on the control command and then sends the control command to the service robot for execution according to the priority; the core cloud gathers the environment data and the analysis result from the local data center, analyzes and solves the environment data and the analysis result to obtain an optimal resource allocation and strategy scheme so as to generate a corresponding global optimization strategy, and sends the global optimization strategy to the 5G core network; and the 5G core network analyzes and executes the global optimization strategy, converts the global optimization strategy into a corresponding control command and sends the control command to the service robot. The method is high in control efficiency and high in reliability in a complex and dynamic environment.

Description

Technical Field

[0001] The present invention relates to the field of control, and in particular to the control of a service robot. Background Art

[0002] Service robots are service robots that can perform tasks autonomously, and are usually used in specific service scenarios such as cleaning, patrolling, delivery, etc. They integrate a variety of sensors and communication modules to achieve environmental perception and remote control.

[0003] In modern service robot applications, 5G networks, with their high bandwidth, low latency and large-scale connectivity, provide important support for service robots to achieve more intelligent and automated operations. Service robots are usually used in industrial, medical, agricultural and other scenarios, and are required to be able to quickly respond to external changes and perform precise tasks in complex and changing environments.

[0004] Traditional service robot communications rely on Wi-Fi, 4G or dedicated wired connections, which have limited bandwidth, high latency and poor mobility, making it difficult to meet real-time and reliability requirements. 5G networks, through their enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC) and massive machine-type communications (mMTC) features, provide service robots with a wider range of communication capabilities and resource management methods, especially for real-time adjustment and optimization of service robot operations in dynamic environments.

[0005] There are several obvious deficiencies in the existing technology in the combination of service robots and 5G networks: First, data processing is highly dependent on the centralized processing mode of the core network, resulting in high latency problems in tasks that require real-time response, affecting the timeliness of operations. Second, the existing resource allocation strategy is prone to uneven resource allocation in multi-task and complex environments, and cannot always ensure that key tasks receive priority support. In addition, most of these technologies rely on static rules and preset parameters, lack the ability to adapt to environmental changes, making it difficult for service robots to make rapid adjustments in dynamic and emergency situations.

[0006] Therefore, there is an urgent need for a service robot control method and device that can solve the above problems. Summary of the invention

[0007] The purpose of the present invention is to provide a distributed real-time control method and device for a service robot, which respectively performs the highest priority, advanced analysis and processing, and global control of the service robot in the edge computing module, local data center, and core cloud, taking into account the timeliness and comprehensiveness of the control, and the control is flexible, with high control efficiency and reliability in complex and dynamic environments.

[0008] In order to achieve the above-mentioned objectives, the present invention provides a distributed real-time control method for a service robot, comprising: an edge computing module of a base station receives environmental data collected by a service robot, sends the environmental data to a local data center, and triggers a highest priority instruction when the environmental data exceeds a first threshold condition, and sends the highest priority instruction to the service robot for execution; after the local data center obtains the environmental data sent by the edge computing module, it performs advanced analysis and future state prediction on the environmental data to obtain an analysis result, generates a corresponding control command when the analysis result exceeds a second threshold condition, determines the priority of the control command based on the analysis result, and then sends the control command to the service robot for execution according to the priority; the core cloud aggregates the environmental data and analysis results from several local data centers, analyzes and solves the environmental data and analysis results to obtain the best resource allocation and strategy scheme, generates a corresponding global optimization strategy based on the best resource allocation and strategy scheme and sends it to the 5G core network; the 5G core network parses and executes the resource allocation in the global optimization strategy, converts the strategy scheme of the global optimization strategy into a corresponding control command and sends it to the corresponding service robot for execution.

[0009] Preferably, the environmental data exceeding the first threshold condition includes: the battery charge of the service robot is less than a preset value, the battery temperature is greater than a preset value, the distance between the camera and the obstacle is less than a preset value, and the object width of the obstacle in front of the camera is greater than a preset value, one or more combinations thereof; the highest priority instruction generated by the edge computing module includes one or more combinations of entering power saving mode, returning to the charging station immediately, reducing the operating speed, shutting down, sending an alarm, sending an obstacle avoidance instruction, and detouring.

[0010] Preferably, the conditions when the analysis result exceeds the second threshold value specifically include: an obstacle appears on the preset path on the front side, the service robot currently deviates from the preset path, the service robot will deviate from the preset path in the future, the battery life of the service robot is less than a preset value, the battery life of the service robot is less than the time it takes for the service robot to complete the current task, the battery power of the service robot is less than a preset value and the distance to charge is greater than a preset value; the control commands generated by the local data center include entering power saving mode, returning to the charging station immediately, reducing the operating speed, shutting down, sending an alarm, sending obstacle avoidance instructions, detouring, path adjustment commands, returning to the preset path, direction adjustment commands, status update commands, sensor data verification commands, and storing one or more of preset key data.

[0011] Preferably, the local data center performs advanced analysis, future state prediction, and priority determination on the environmental data through a random forest algorithm.

[0012] Specifically, the local data center also stores the environmental data and control commands to generate historical data. The local data center performs advanced analysis, future state prediction, and priority determination on the environmental data through a random forest algorithm in combination with the historical data.

[0013] Preferably, the control command issued by the 5G core network includes one or more combinations of mode adjustment commands and path adjustment commands.

[0014] Preferably, the edge computing module of the base station obtains the encrypted environmental data sent by the service robot. After the edge computing module of the base station, the local data center and the core cloud obtain the environmental data, they decrypt and verify the environmental data and perform processing and analysis based on the decrypted environmental data.

[0015] Preferably, the environmental data includes camera data, GPS positioning data and status information.

[0016] Preferably, the core cloud obtains the best resource allocation and strategy solution based on the environmental data analysis, specifically including: establishing a linear programming optimization model, defining the objective function and constraints, using an optimization solver to solve the model, and obtaining the best resource allocation and strategy solution.

[0017] The present invention also provides a distributed real-time control device for a service robot, comprising an edge computing module, a local data center, a core cloud and a 5G core network, wherein the edge computing module is an edge computing module of a base station; the edge computing module, the local data center, the core cloud and the 5G core network are respectively provided with a processor and a memory, wherein the memory stores operation instructions, and the operation instructions enable the processor to execute the distributed real-time control method for the service robot as described above.

[0018] Compared with the prior art, the present invention performs the highest priority, advanced analysis and processing, and global control on the service robot in the edge computing module, local data center, and core cloud, respectively. The edge computing module of the base station greatly reduces the task processing delay, improves the robot's responsiveness in a dynamic environment, and reduces the dependence on the core network. Through the global control of the core cloud, intelligent resource allocation is carried out to achieve more efficient resource utilization and more stable system performance. Through the predictive processing of the local data center, the operation strategy of the service robot can be adjusted in real time, so that it can adapt to environmental changes in a timely manner to ensure efficient completion of tasks. Through the distributed computing and multi-node collaboration of the edge computing module, local data center, and core cloud, the stability and fault tolerance of the system are effectively improved. In summary, the present invention dynamically adjusts resources to ensure the priority of key tasks and avoids task delays caused by resource competition through an intelligent resource allocation mechanism. By using machine learning and adaptive strategies, the robot can automatically adjust the operation strategy according to real-time environmental changes, improve the efficiency and reliability of task execution in complex and dynamic environments, and ultimately achieve more intelligent and efficient robot operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of the distributed real-time control method of the service robot of the present invention. DETAILED DESCRIPTION

[0020] In order to explain the technical content, structural features, achieved objectives and effects of the present invention in detail, the following is a detailed description in conjunction with the implementation methods and the accompanying drawings.

[0021] refer to Figure 1 The present invention discloses a distributed real-time control method for a service robot, comprising steps S1 to S5.

[0022] S1, the service robot collects environmental data through its own sensors and sends the environmental data to the base station.

[0023] The environmental data includes camera data, GPS positioning data and status information. The status information includes battery power, battery temperature, distance to the obstacle ahead, obstacle width, obstacle height, etc.

[0024] The service robot collects environmental data through integrated sensors and performs preliminary processing on the environmental data: data format conversion processing, data compression processing, and data encryption processing.

[0025] Specifically, the service robot converts image data into JPEG format, converts sensor data into JSON format, and then compresses the data. Image data is lossily compressed using the JPEG algorithm. Other sensor data is losslessly compressed using the LZW algorithm. Finally, the AES-256 algorithm is used for data encryption. After that, the built-in 5G module searches and connects to nearby base stations, uses the TCP protocol for data transmission, and transmits environmental data to the base station.

[0026] S2, the edge computing module of the base station receives the environmental data sent by the service robot, sends the environmental data to the local data center, and triggers the highest priority instruction when the environmental data exceeds the first threshold condition, and sends the highest priority instruction to the service robot for execution.

[0027] Among them, the edge computing module of the base station obtains the encrypted environmental data sent by the service robot, and the edge computing module of the base station performs decryption processing and preliminary data cleaning on the environmental data received, and performs anomaly detection on the environmental data after preliminary processing to determine whether the environmental data exceeds the first threshold condition.

[0028] Specifically, the edge computing module includes a data decryption and verification submodule, a data cleaning submodule, an anomaly detection submodule and a data forwarding submodule. The data decryption and verification submodule decrypts and verifies the environmental data sent by the service robot to verify whether the environmental data is legal data. Subsequently, the data cleaning submodule performs data cleaning on the decrypted and verified environmental data to remove duplicate data packets. Then the anomaly detection submodule performs anomaly detection through a decision tree model to determine whether the environmental data is abnormal: set thresholds and conditions, which are called first threshold conditions, and determine whether the environmental data exceeds the first threshold condition (normal range). If so, the highest priority instruction is immediately triggered, and the data forwarding submodule forwards the highest priority instruction to the service robot to adjust the real-time operation of the service robot. The data forwarding submodule also forwards the environmental data sent by the service robot to the local data center via optical fiber.

[0029] In this embodiment, the environmental data exceeds the first threshold condition includes: the battery power of the service robot is less than the preset value, the battery temperature is greater than the preset value, the distance between the camera and the obstacle is less than the preset value, and the width of the obstacle in front of the camera is greater than the preset value. One or more combinations; the highest priority instruction includes entering power saving mode, returning to the charging station immediately, reducing the running speed, stopping, sending an alarm, sending an obstacle avoidance instruction, and detouring. The form of the highest priority instruction and the corresponding first condition threshold is as follows: Table 1:

[0030]

[0031]

[0032] Table 1

[0033] S3, after the local data center obtains the environmental data sent by the edge computing module, it performs advanced analysis and future state prediction on the environmental data to obtain an analysis result, generates a corresponding control command when the analysis result exceeds a second threshold condition, determines the priority of the control command based on the analysis result, and sends the control command to the service robot for execution according to the priority.

[0034] The local data center performs advanced analysis, future state prediction, and priority determination on the environmental data through a random forest algorithm. The local data center also stores the environmental data and control commands to generate historical data, and performs advanced analysis, future state prediction, and priority determination on the environmental data through a random forest algorithm combined with historical data.

[0035] The analysis result exceeds the second threshold condition specifically including: an obstacle appears on the preset path on the front side, the service robot currently deviates from the preset path, the service robot will deviate from the preset path in the future, the service robot's battery life is less than a preset value, the service robot's battery life is less than the time it takes for the service robot to complete the current task, the service robot's battery power is less than a preset value and the distance to charge is greater than a preset value; the control command generated by the local data center includes one or more combinations of entering power saving mode, returning to the charging station immediately, reducing the running speed, shutting down, sending an alarm, sending an obstacle avoidance instruction, detouring, path adjustment command, returning to the preset path, direction adjustment command, status update command, sensor data verification command, and storing preset key data. For example: when the battery power is low, adjust the current path and plan the charging time, such as returning to charge when the battery power is low and the distance to the charging pile is too far, such as returning to charge when the difference between the current battery power and the time to return to the charging pile is within the preset value.

[0036] Among them, the local data center decrypts and verifies the environmental data sent by the base station, standardizes all the data, and then uses the random forest algorithm to perform advanced analysis and future state prediction on the standardized environmental data to obtain the analysis results, and determines whether the analysis results exceed the second threshold condition. If so, the corresponding control command is generated. If there are multiple control commands, the control commands are prioritized and then sent to the service robot for execution according to their priority levels. For example, if the current battery power is low and the distance to the charging station is far, and there are obstacles on the path ahead, relevant control commands such as obstacle avoidance instructions, return charging instructions, and path adjustment instructions are generated. Through the analysis of the random forest algorithm, the obstacle avoidance instruction is determined to be the highest priority, so the highest priority control command of the obstacle avoidance instruction is immediately generated and sent to the service robot; the return charging instruction generated because of the low power is determined to be the second highest priority, and the second priority control command for path adjustment is generated, which is then sent to the service robot; the others are downgraded to low priority and delayed.

[0037] S4, the core cloud aggregates the environmental data and analysis results from several local data centers, analyzes and solves the optimal resource allocation and strategy solution based on the environmental data and analysis results, generates a corresponding global optimization strategy based on the optimal resource allocation and strategy solution, and sends it to the 5G core network via optical fiber.

[0038] The strategy planning instruction includes one or more combinations of a mode adjustment command and a path adjustment command.

[0039] Among them, the core cloud analyzes and solves the best resource allocation and strategy plan based on the environmental data and analysis results, specifically including: establishing a linear programming optimization model, defining objective functions and constraints, using an optimization solver to solve the model, and obtaining the best resource allocation and strategy plan.

[0040] Resource allocation refers to the core cloud allocating global resources (including network bandwidth, computing power, and multi-robot task coordination) by integrating data uploaded by multiple local data centers (environmental data and analysis results), ensuring that high-priority instructions (such as emergency obstacle avoidance or rescue tasks) receive more resource support, while balancing the remaining resources to maintain the efficiency and stability of the overall system. Prioritize low-latency communication resources for robots performing emergency tasks, while optimizing bandwidth utilization for non-emergency tasks.

[0041] S5, the 5G core network parses and executes the resource allocation in the global optimization strategy, converts the strategy scheme of the global optimization strategy into corresponding control commands and sends them to the corresponding service robot for execution.

[0042] After the 5G core network receives the global optimization strategy of the core cloud computing. The first step is user access and authentication. AMF processes user access requests, performs identity authentication, and interacts with UDM to obtain user data; NSSF selects the appropriate network slice. The second step is session management and resource allocation. SMF establishes and manages user sessions, including IP address allocation and resource allocation; PCF formulates session QoS and policies. The third step is data forwarding and traffic management. UPF is responsible for data packet forwarding and traffic management, and optimizes data transmission according to PCF policies. The fourth step is policy control and execution. PCF dynamically generates network policies and sends them to SMF and UPF to ensure continuous optimization of resources and service quality. The fifth step is service discovery and network element collaboration. NRF manages network element registration and service discovery to ensure efficient collaboration between network elements.

[0043] After receiving the control command sent by the 5G core network, the service robot will parse it, identify the type and content of the command, and confirm the validity and executability of the command. Adjust the operation mode (automatic mode or manual mode) according to the command of the core network. If a path adjustment command is received, the implementation path is updated according to the built-in Dijkstra algorithm. Finally, continue to monitor according to the previous process, and regularly feedback environmental data and abnormal reports.

[0044] Among them, after the service robot receives the corresponding commands sent by the base station, local data center and 5G core network, it will identify the priority of the command and execute the corresponding instructions according to its priority.

[0045] refer to Figure 1 The present invention also discloses a distributed real-time control device for a service robot, including an edge computing module, a local data center, a core cloud and a 5G core network, wherein the edge computing module is an edge computing module of a base station; the edge computing module, the local data center, the core cloud and the 5G core network are respectively provided with a processor and a memory, wherein the memory stores operation instructions, and the operation instructions enable the processor to execute the distributed real-time control method for the service robot as described above.

[0046] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of rights of the present invention. Therefore, equivalent changes made according to the scope of the patent application of the present invention are still within the scope covered by the present invention.

Claims

1. A distributed real-time control method for a service robot, characterized in that: include: The edge computing module of the base station receives the environmental data collected by the service robot, sends the environmental data to the local data center, triggers the highest priority instruction when the environmental data exceeds the first threshold condition, and sends the highest priority instruction to the service robot for execution; After the local data center obtains the environmental data sent by the edge computing module, it performs advanced analysis and future state prediction on the environmental data to obtain an analysis result, generates a corresponding control command when the analysis result exceeds a second threshold condition, determines the priority of the control command according to the analysis result, and then sends the control command to the service robot for execution according to the priority of the control command; The core cloud aggregates the environmental data and analysis results from the plurality of local data centers, analyzes and solves the optimal resource allocation and strategy scheme based on the environmental data and analysis results, generates a corresponding global optimization strategy based on the optimal resource allocation and strategy scheme, and sends the strategy to the 5G core network; The 5G core network parses and executes the resource allocation in the global optimization strategy, converts the strategy scheme of the global optimization strategy into corresponding control commands and sends them to the corresponding service robot for execution.

2. The distributed real-time control method for a service robot according to claim 1, characterized in that: The condition that the environmental data exceeds the first threshold value includes: the battery power of the service robot is less than a preset value, the battery temperature is greater than a preset value, the distance between the camera and the obstacle is less than a preset value, and the width of the obstacle in front of the camera is greater than a preset value. One or more combinations of the above; The highest priority instructions generated by the edge computing module include a combination of one or more of entering power saving mode, returning to the charging station immediately, reducing the operating speed, shutting down, sending an alarm, sending obstacle avoidance instructions, and detouring.

3. The distributed real-time control method for a service robot according to claim 1, characterized in that: The condition that the analysis result exceeds the second threshold value specifically includes: an obstacle appears on the preset path on the front side, the service robot currently deviates from the preset path, the service robot will deviate from the preset path in the future, the service robot's endurance time is less than a preset value, the service robot's endurance time is less than the time it takes for the service robot to complete the current task, the service robot's battery power is less than a preset value and the distance to charge is greater than a preset value, or one or more combinations thereof; The control commands generated by the local data center include entering power saving mode, returning to the charging station immediately, reducing the operating speed, shutting down, sending an alarm, sending an obstacle avoidance instruction, detouring, path adjustment command, returning to a preset path, direction adjustment command, status update command, sensor data verification command, and storing one or more combinations of preset key data.

4. The distributed real-time control method for a service robot according to claim 1, characterized in that: The local data center performs advanced analysis, future state prediction, and priority determination on the environmental data through a random forest algorithm.

5. The distributed real-time control method for a service robot according to claim 4, characterized in that: The local data center also stores the environmental data and control commands to generate historical data. The local data center performs advanced analysis, future state prediction, and priority determination on the environmental data through a random forest algorithm combined with historical data.

6. The distributed real-time control method for a service robot according to claim 1, characterized in that: The control command issued by the 5G core network includes one or more combinations of mode adjustment commands and path adjustment commands.

7. The distributed real-time control method for a service robot according to claim 1, characterized in that: The edge computing module of the base station obtains the encrypted environmental data sent by the service robot. After the edge computing module of the base station, the local data center and the core cloud obtain the environmental data, they decrypt and verify the environmental data and perform processing and analysis based on the decrypted environmental data.

8. The distributed real-time control method for a service robot according to claim 1, characterized in that: The environmental data includes camera data, GPS positioning data and status information.

9. The distributed real-time control method for a service robot according to claim 1, characterized in that: The core cloud analyzes and solves the best resource allocation and strategy plan based on the environmental data and analysis results, specifically including: establishing a linear programming optimization model, defining the objective function and constraints, using the optimization solver to solve the model, and obtaining the best resource allocation and strategy plan.

10. A distributed real-time control device for a service robot, characterized in that: It includes an edge computing module, a local data center, a core cloud and a 5G core network, wherein the edge computing module is an edge computing module of a base station; the edge computing module, the local data center, the core cloud and the 5G core network are respectively provided with a processor and a memory, wherein the memory stores operation instructions, and the operation instructions enable the processor to execute the distributed real-time control method of the service robot as described in any one of claims 1 to 9.

Images