Search

CN-122019148-A - Task processing method and device based on robot, computer equipment and medium

CN122019148ACN 122019148 ACN122019148 ACN 122019148ACN-122019148-A

Abstract

The application belongs to the technical field of artificial intelligence, and relates to a task processing method, a device, computer equipment and a medium based on a robot, wherein the task processing method comprises the steps of carrying out demand analysis on task instructions to obtain task demand data; the method comprises the steps of analyzing task demand data, collected hardware state data and environment characteristic data of a robot to generate a demand instruction, carrying out module splitting and priority ordering processing on a core algorithm of the robot based on the demand instruction to obtain a modularized algorithm structure and priority data corresponding to sub-modules, optimizing the modularized algorithm structure based on the priority data and a strategy library to obtain a target modularized algorithm structure, generating a resource allocation scheme based on the target modularized algorithm structure and hardware resource conditions, and carrying out execution processing on the task based on the target modularized algorithm structure and the resource allocation scheme. The application can be applied to task processing scenes in the fields of financial science and technology and digital medical treatment, and can improve the accuracy of task processing.

Inventors

  • WANG JIANZONG
  • SUN AOLAN

Assignees

  • 平安科技（深圳）有限公司

Dates

Publication Date
20260512
Application Date
20260109

Claims (10)

  1. 1. The task processing method based on the robot is characterized by comprising the following steps of: receiving a task instruction corresponding to a task to be processed, and carrying out demand analysis on the task instruction to obtain corresponding task demand data; collecting hardware state data of a robot and collecting current environmental characteristic data; Analyzing and processing the task demand data, the hardware state data and the environment characteristic data based on a preset demand modeling engine to generate a corresponding demand instruction; Performing module splitting and priority ordering processing on a core algorithm of the robot based on the demand instruction to obtain a modularized algorithm structure and priority data corresponding to sub-modules, wherein the number of the sub-modules comprises a plurality of sub-modules; Optimizing the modular algorithm structure based on the priority data and a preset strategy library to obtain a corresponding target modular algorithm structure; generating a corresponding resource allocation scheme based on the target modularized algorithm structure and a preset hardware resource condition; And executing the task based on the target modularized algorithm structure and the resource allocation scheme to obtain a corresponding task processing result.
  2. 2. The method for processing tasks based on robots according to claim 1, wherein the step of performing module splitting and prioritization processing on the core algorithm of the robots based on the demand instruction to obtain modular algorithm structures and priority data corresponding to sub-modules specifically comprises: Performing functional decoupling on a core algorithm of the robot to obtain a plurality of corresponding sub-modules in a splitting way; finishing all the submodules to generate a corresponding modularized algorithm structure; Acquiring a preset priority evaluation strategy; And carrying out evaluation processing on each sub-module based on the priority evaluation strategy to obtain priority data respectively corresponding to each sub-module.
  3. 3. The robot-based task processing method according to claim 1, wherein the step of optimizing the modular algorithm structure based on the priority data and a preset policy library to obtain a corresponding target modular algorithm structure specifically comprises: Calling a preset strategy library; inquiring an optimization strategy matched with the priority data from the strategy library; dynamically adjusting the optimization strategy to obtain a corresponding target optimization strategy; Optimizing the submodules contained in the modularized algorithm structure based on the target optimization strategy to obtain corresponding target submodules; Finishing the target sub-module to obtain finished structural data; and taking the tidied structure data as the target modularized algorithm structure.
  4. 4. The robot-based task processing method according to claim 3, wherein the step of dynamically adjusting the optimization strategy based on a preset adjustment rule to obtain a corresponding target optimization strategy specifically includes: calling a preset monitoring algorithm; Monitoring current hardware load information in real time based on the monitoring algorithm; based on a preset parameter adjustment rule, dynamically adjusting the strategy parameters of the optimization strategy according to the hardware load information to obtain an adjusted optimization strategy; And taking the adjusted optimization strategy as the target optimization strategy.
  5. 5. The method for processing a task based on a robot according to claim 1, wherein the step of performing the processing on the task based on the target modular algorithm structure and the resource allocation scheme to obtain a corresponding task processing result specifically includes: Obtaining target sub-modules corresponding to the target modularized algorithm structure, wherein the number of the target sub-modules comprises a plurality of target sub-modules; Constructing a corresponding module dependency relationship map based on the target sub-module; determining the execution sequence of each target sub-module based on the module dependency graph, and distributing corresponding hardware resources for each target sub-module based on the resource distribution scheme to obtain corresponding appointed sub-modules; And executing the task on the basis of the appointed submodule to obtain a corresponding task processing result.
  6. 6. The robot-based task processing method according to claim 5, wherein the step of constructing a corresponding module dependency graph based on the target sub-module specifically includes: performing input and output analysis on all the target submodules to obtain corresponding input and output relations; Calling a preset graphical tool; performing map drawing processing on the input-output relationship based on the graphical tool to obtain a corresponding generated map; and taking the generated map as the module dependency relation map.
  7. 7. The robot-based task processing method according to claim 1, further comprising, after the step of performing execution processing on the task based on the target modular algorithm structure and the resource allocation scheme to obtain a corresponding task processing result: In the processing process of the task, collecting performance feedback data based on a preset performance monitoring strategy; collecting hardware feedback data based on a preset hardware monitoring tool; collecting environmental feedback data based on a preset sensor; performing integration processing based on time stamps on the performance feedback data, the hardware feedback data and the environment feedback data to obtain corresponding target feedback data; and storing the target feedback data.
  8. 8. A robot-based task processing device, comprising: The receiving module is used for receiving a task instruction corresponding to a task to be processed, and carrying out demand analysis on the task instruction to obtain corresponding task demand data; The first acquisition module is used for acquiring hardware state data of the robot and acquiring current environment characteristic data; The analysis module is used for analyzing and processing the task demand data, the hardware state data and the environment characteristic data based on a preset demand modeling engine to generate a corresponding demand instruction; The processing module is used for carrying out module splitting and priority ordering processing on the core algorithm of the robot based on the demand instruction to obtain modularized algorithm structures and priority data corresponding to the submodules, wherein the number of the submodules comprises a plurality of submodules; The optimization module is used for optimizing the modular algorithm structure based on the priority data and a preset strategy library to obtain a corresponding target modular algorithm structure; the generating module is used for generating a corresponding resource allocation scheme based on the target modularized algorithm structure and a preset hardware resource condition; and the execution module is used for executing the task based on the target modularized algorithm structure and the resource allocation scheme to obtain a corresponding task processing result.
  9. 9. A computer device comprising a memory and a processor, the memory having stored therein computer readable instructions which when executed by the processor implement the steps of the robot-based task processing method of any of claims 1 to 7.
  10. 10. A computer readable storage medium, characterized in that it has stored thereon computer readable instructions, which when executed by a processor, implement the steps of the robot-based task processing method according to any of claims 1 to 7.

Description

Task processing method and device based on robot, computer equipment and medium Technical Field The application relates to the technical field of artificial intelligence, which can be applied to the fields of financial science and technology, digital medical treatment and the like, in particular to a task processing method, a task processing device, computer equipment and a storage medium based on a robot. Background In the existing robot algorithm deployment technology, the algorithm is lighter than the algorithm which is completed at one time before the robot deployment, and the algorithm parameters are usually fixed by means of model pruning, quantization and the like. Although the method for fixing the parameters can reduce the algorithm complexity to a certain extent and improve the deployment efficiency, the method cannot be dynamically adjusted according to the actual running state of the robot after deployment. Because the actual running environment of the robot is complex and changeable, the requirements on the algorithm performance are large in difference under different scenes, and the fixed algorithm parameters are difficult to adapt to diversified running requirements, the accuracy of the robot is suddenly reduced when the robot executes the task, and the task execution effect and quality are seriously affected. In the field of financial insurance, taking an insurance risk assessment robot as an example, a traditional method optimizes a risk assessment algorithm at one time before deployment, and fixes parameters. However, in an actual insurance scenario, the risk condition of the customer may dynamically change with time, market environment, and other factors. For example, the risk level of a client may change due to occupational change, health condition change and the like after the client makes an application, but the robot cannot adjust the evaluation strategy in time due to fixed algorithm parameters, and the risk evaluation is still performed according to the initial parameters, so that the deviation between the evaluation result and the actual situation is large, accurate risk reference cannot be provided for an insurance company, and reasonable pricing and risk management and control of insurance business are affected. In the field of digital medicine, as exemplified by medical analysis robots, similar problems exist. After the medical analysis algorithm is subjected to light weight treatment and parameters are fixed before deployment, the robot is difficult to dynamically adjust the algorithm according to new data and new characteristics in the actual analysis process in the face of complex and changeable diseases of different patients. For example, for some rare conditions or patients whose conditions develop rapidly, the fixed algorithm parameters may not accurately identify the characteristics of the condition, resulting in inaccurate medical analysis results, delayed patient treatment timing, and reduced medical quality of service. Therefore, a task processing method capable of improving accuracy and adaptability of a robot to perform tasks after the robot is deployed is needed. Disclosure of Invention The embodiment of the application aims to provide a task processing method, a task processing device, computer equipment and a storage medium based on a robot, so as to solve the technical problem that the accuracy of the existing robot is low when the task is executed. In a first aspect, a task processing method based on a robot is provided, including: receiving a task instruction corresponding to a task to be processed, and carrying out demand analysis on the task instruction to obtain corresponding task demand data; collecting hardware state data of a robot and collecting current environmental characteristic data; Analyzing and processing the task demand data, the hardware state data and the environment characteristic data based on a preset demand modeling engine to generate a corresponding demand instruction; Performing module splitting and priority ordering processing on a core algorithm of the robot based on the demand instruction to obtain a modularized algorithm structure and priority data corresponding to sub-modules, wherein the number of the sub-modules comprises a plurality of sub-modules; Optimizing the modular algorithm structure based on the priority data and a preset strategy library to obtain a corresponding target modular algorithm structure; generating a corresponding resource allocation scheme based on the target modularized algorithm structure and a preset hardware resource condition; And executing the task based on the target modularized algorithm structure and the resource allocation scheme to obtain a corresponding task processing result. In a second aspect, a robot-based task processing device is provided, including: The receiving module is used for receiving a task instruction corresponding to a task to be processed, and carrying out demand analysis on the t