Control method for drone inspection of chemical production plant, electronic device and storage medium

Patent metadata

Field	Value
Publication number	US-12487149-B2
Application number	US-202418955501-A
Country	US
Kind code	B2
Filing date	Nov 21, 2024
Priority date	Dec 7, 2023
Publication date	Dec 2, 2025
Grant date	Dec 2, 2025

Abstract

Provided is a control method, system and apparatus for drone inspection of a chemical production plant, relating to the field of chemical fiber intelligent technology. The control method includes: determining a first drone for performing a preset inspection task in a target scene, wherein the first drone is responsible for the preset inspection task of the chemical production plant; obtaining data of a target object collected by the first drone in the target scene; determining a state of the target object in the target scene based on the data of the target object; and outputting prompt information matching a preset state in the target scene when the state represents that the target object is in the preset state of the target scene.

First claim

What is claimed is:

1 . A control method for drone inspection of a chemical production plant, comprising:
determining a first drone for performing a preset inspection task in a target scene, wherein the first drone is responsible for the preset inspection task of the chemical production plant; obtaining data of a target object collected by the first drone in the target scene; determining a state of the target object in the target scene based on the data of the target object; outputting prompt information matching a preset state in the target scene in a case of the state represents that the target object is in the preset state of the target scene; receiving an item demand sent by the first drone; and dispatching a second drone for the first drone based on the item demand, so that the second drone transports a target item to an inspector corresponding to the first drone based on the item demand; wherein the preset inspection task comprises: in a case of the target scene belongs to an active inspection scene and the target object is a device, the preset inspection task is a routine inspection task or a sampling inspection task, wherein the routine inspection task is pre-set, and the sampling inspection task is randomly generated; in a case of the target scene belongs to an active inspection scene and the target object is a person, the preset inspection task is a routine inspection task or a sampling inspection task; in a case of the target scene belongs to a passive inspection scene and the target object is a device, the preset inspection task is a first dedicated inspection task, and the first dedicated inspection task is set based on an attribute of the device; in a case of the target scene belongs to a passive inspection scene and the target object is a person, the preset inspection task is a second dedicated inspection task, and the second dedicated inspection task is generated based on a personnel requirement.

2 . The method of claim 1 , wherein the determining a first drone for performing a preset inspection task in a target scene, comprises:
in response to receiving a follow instruction forwarded by a drone, determining the drone that forwards the follow instruction as the first drone, wherein the follow instruction is an instruction sent by an inspector to the drone that sends the follow instruction and is used to instruct the drone to follow the inspector to work, and the follow instruction comprises information characterizing the preset inspection task.

3 . The method of claim 1 , wherein the determining a first drone for performing a preset inspection task in a target scene, comprises:
in response to receiving an operation requirement sent by a terminal, determining the first drone for performing the preset inspection task in the target scene for the terminal from candidate drones based on the operation requirement and state information of the candidate drones, wherein the operation requirement comprises information characterizing the preset inspection task.

4 . The method of claim 1 , wherein the determining a first drone for performing a preset inspection task in a target scene, comprises:
determining the first drone for performing the preset inspection task in the target scene from candidate drones according to state information of the candidate drones in combination with an inspection cycle.

5 . The method of claim 1 , wherein the preset state comprises a device abnormal state, and the target object comprises a target device;
wherein the outputting prompt information matching a preset state in the target scene in the case of the state represents that the target object is in the preset state of the target scene, comprises: outputting prompt information matching the device abnormal state in the target scene in a case of the state represents that the target device is in the device abnormal state of the target scene.

6 . The method of claim 1 , wherein the preset state comprises a person abnormal state, and the target object comprises a target person;
wherein the outputting prompt information matching a preset state in the target scene in the case of the state represents that the target object is in the preset state of the target scene, comprises: outputting prompt information matching the person abnormal state in the target scene in a case of the state represents that the target person is in the person abnormal state of the target scene.

7 . The method of claim 1 , wherein the determining a state of the target object in the target scene based on the data of the target object, comprises:
obtaining a known corresponding relationship, wherein the known corresponding relationship at least comprises a relationship between each state of the target object and corresponding interval data; and analyzing the state corresponding to the data of the target object in the target scene based on the known corresponding relationship.

8 . The method of claim 7 , further comprising:
establishing a corresponding relationship between the data of the target object and the state of the target object in the target scene; and updating the known corresponding relationship based on the corresponding relationship.

9 . The method of claim 1 , wherein the determining a state of the target object in the target scene based on the data of the target object, comprises:
inputting the data of the target object in the target scene into a state analysis model, wherein the state analysis model is obtained by training based on a known corresponding relationship and is used to predict the state of the target object in the target scene; and obtaining the state of the target object in the target scene output by the state analysis model.

10 . The method of claim 9 , further comprising:
establishing a corresponding relationship between the data of the target object and the state of the target object in the target scene; and updating the known corresponding relationship based on the corresponding relationship.

11 . The method of claim 1 , wherein the first drone is provided with a first voice recognition model, the target object comprises an inspector, and the method further comprises:
sending the item demand to a system background if the item demand of the inspector is recognized by the first voice recognition model in a case of the first drone receives voice of the inspector.

12 . The method of claim 1 , wherein the first drone is provided with a second voice recognition model, the target object comprises a target device, and the method further comprises:
inputting collected sound of the target device into the second voice recognition model in a process of performing the preset inspection task by the first drone, and sending the item demand to a system background if the item demand matching a fault of the target device is analyzed by the second voice recognition model.

13 . The method of claim 1 , further comprising:
determining an inspection path for the first drone to perform the preset inspection task; and sending the inspection path to the first drone, so that the first drone performs the preset inspection task based on the inspection path.

14 . The method of claim 1 , wherein the target object comprises an inspector, and the method further comprises:
receiving a control instruction of the inspector sent by the first drone and an execution record of the first drone for the control instruction, wherein the control instruction is received in a process of performing the preset inspection task by the first drone, and the first drone recognizes that the inspector has control authority to send the contro

Assignees

• Zhejiang Hengyi Petrochemical Co Ltd

Inventors

• Peng Xiantao
• Wang Peng
• Qiu Yibo
• Li Dake
• Xu Feng
• Jin Junliang

Classifications

• G05D2105/89

  for inspecting structures, e.g. wind mills, bridges, buildings or vehicles · CPC title

• G05D2105/28

  of freight · CPC title

• G05D1/686

  Maintaining a relative position with respect to moving targets, e.g. following animals or humans (for pointing payloads towards targets G05D1/689; involving controlling the position or course of two or more vehicles G05D1/69) · CPC title

• G05D1/689 Primary

  Pointing payloads towards fixed or moving targets (positioning towed, pushed or suspended implements G05D1/672) · CPC title

• G05D1/2285

  using voice or gesture commands · CPC title