Intelligent detection robot and system for chemical engineering site

Abstract

The invention discloses an intelligent detection robot and system for a chemical field. The intelligent inspection robot comprises a main control system, a walking module, a communication module, a positioning module, an alarm module, a camera and an infrared imaging sensor, and the intelligent inspection robot is configured to operate according to the following modes: receiving a task instruction which indicates a plurality of task sites and one or more task contents of each task site; sequentially arriving at each task site, and going to the next task site after sequentially executing all task contents at each task site; and returning to the charging bin after the task contents of all the task sites are executed. And flexible and efficient on-site inspection is realized.

Description

Technical Field

[0001] The invention relates to an intelligent detection robot and system used in chemical industry sites. Background Art

[0002] At present, the on-site environmental status of the chemical industry is recorded by regular manual inspections, and the purpose of recording the on-site environmental status of the chemical industry is achieved by regular manual inspections. By manually recording the on-site environmental status parameters of the chemical industry, abnormalities of various devices and equipment can be discovered, thereby solving safety hazards or device failures in the on-site environment of the chemical industry. Summary of the invention

[0003] The present invention provides an intelligent detection robot and system for a chemical industry site, so as to detect the chemical industry site safely and efficiently.

[0004] The technical solution of the present invention is as follows: an intelligent detection robot for chemical sites, comprising a main control system, a walking module, a communication module, a positioning module, an alarm module, a camera and an infrared imaging sensor, wherein the intelligent detection robot is configured to operate in the following manner:

[0005] receiving a task instruction, the task instruction indicating a plurality of task locations and one or more task contents at each task location;

[0006] Arrive at each mission location in turn, and complete all missions at each mission location before proceeding to the next mission location;

[0007] After completing all the tasks at all the mission locations, return to the charging station;

[0008] Among them, the execution process of the task content includes: when the detection task content is pan-tilt photography, the target equipment at the task location is photographed, and the instrument value of the target equipment in the photo is identified. If the instrument value meets the set standard, the task content is completed; otherwise, an alarm is issued, and a maintenance work order is generated and sent to the background. The task content is completed after receiving the signal of successful identification alarm sent by the background; when the detection task content is thermal infrared imaging detection, thermal infrared photography is taken at the task location to identify the temperature value. If the temperature value meets the set standard, the task content is completed; otherwise, an alarm is issued, and a maintenance work order is generated and sent to the background. The task content is completed after receiving the signal of successful identification alarm sent by the background.

[0009] Optionally, when the task content is taking a photo with a gimbal, the task content indicates the x-axis and z-axis position points and the zoom multiples of the gimbal rotation.

[0010] Optionally, the instrument value of the target device in the photo is recognized by a recognition model built into the main control system or by an AI computing box to which the intelligent inspection robot is connected in communication.

[0011] Optionally, after receiving the business content, it also plans the forward route according to multiple task locations indicated by the business content.

[0012] Optionally, after receiving the service content, it also performs a self-check operation.

[0013] The technical solution of the present invention is as follows: a system, including the above-mentioned intelligent detection robot for chemical sites, and also including a background, which is configured to: in response to user operations, insert task content at the task location, and sort the task content of the task location, and then package the task content of each task location as a whole to form a task instruction, and send the task instruction to the intelligent detection robot.

[0014] Achieve flexible and efficient on-site inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a structural block diagram of the intelligent detection robot used in chemical industry site of the present invention.

[0016] Figure 2 yes Figure 1 The following is a workflow diagram of an intelligent detection robot used at chemical sites. DETAILED DESCRIPTION

[0017] The present invention is further described below in conjunction with specific embodiments, but the protection scope of the present invention is not limited thereto.

[0018] Figure 1 It is a structural block diagram of the intelligent detection robot used in chemical industry site of the present invention. Figure 2 yes Figure 1 The following is a workflow diagram of an intelligent detection robot used at chemical sites.

[0019] In the intelligent detection robot, 3D laser radar, ultrasonic obstacle avoidance sensor, flameproof housing (not shown), explosion-proof drive motor and reduction box (four-wheel drive, not shown in the figure), heavy-duty solid rubber wheels (as walking module), battery and charging module, communication module (such as 5G communication module), and magnetic navigation sensor are combined into the lower computer. The lower computer is equipped with GD32F470 chip (as processor).

[0020] The infrared imaging sensor, the camera (specifically a zoom camera), the explosion-proof pan / tilt (not shown), and the alarm module (specifically an explosion-proof sound and light alarm module) are combined into a host computer, which is equipped with an RK3588 chip (as a processor).

[0021] The intelligent detection robot also includes a positioning module, a voice interaction module and a memory (such as Flash).

[0022] The intelligent inspection robot alarm prompt is realized through the explosion-proof sound and light alarm module, which includes an indicator light and a buzzer. When an alarm is generated, the indicator light flashes and the buzzer sounds an alarm. The process lasts for 1 minute to achieve the purpose of the alarm.

[0023] When there is no task, the intelligent inspection robot charges in the explosion-proof charging warehouse and waits for the platform to issue task instructions, during which it maintains wireless communication with the platform. When the platform issues a task instruction, the intelligent inspection robot first receives the task instruction issued by the platform through the 5G communication module (i.e., receives the business content). After receiving the task instruction, the main control system (e.g., including the processor) inside the intelligent inspection robot reads the task instruction and obtains task information such as the start time, task location, and corresponding content of the task.

[0024] When the task execution time arrives, the main control system of the intelligent inspection robot starts the task. Before setting off for the task location, the intelligent inspection robot performs a self-inspection, which includes communication detection of the drive device, power detection, ultrasonic obstacle avoidance sensor, communication detection of 3D laser radar, camera, infrared imaging sensor, communication module and positioning module. If the intelligent inspection robot fails the self-inspection, the explosion-proof sound and light alarm module will issue an alarm prompt, and the inspection robot main control system will generate a self-inspection work order after analysis, and send a signal to the cloud platform through wireless communication to page the maintenance personnel. The cloud platform then prompts the intelligent inspection robot that the inspection task is over and the inspection robot stops on the spot.

[0025] When the intelligent inspection robot passes the self-inspection successfully, it drives out of the charging compartment and drives four explosion-proof drive motors. The motors are connected to the reduction box, and the heavy-loaded solid rubber wheels are driven to the task location through the reduction box. If the left and right wheels need to be turned at different speeds, it can be done through the differential steering mode. The main control system of the intelligent inspection robot generates a global path planning based on relevant algorithms. During the forward process, the 3D laser radar, ultrasonic obstacle avoidance sensor and Beidou positioning system (i.e. positioning module) plan the forward route and avoid roadblocks. When encountering an unavoidable roadblock, the main control system determines whether the roadblock can be passed smoothly. If it cannot be passed smoothly, the explosion-proof sound and light alarm module will issue an alarm prompt, and the operator will be paged through wireless communication. The cloud platform prompts that the inspection task is over and the inspection robot stops on the spot. After the inspection robot successfully passes the roadblock, it continues to cooperate with the 3D laser radar, ultrasonic obstacle avoidance sensor and Beidou positioning system to reach the designated location of the task according to the global path modeling.

[0026] After the intelligent inspection robot arrives at the designated location of the task, the main control system determines the task content, which is divided into PTZ camera recognition and thermal infrared imaging recognition. After the task content is determined, the inspection robot starts to perform the task. When the task content is determined to be PTZ camera recognition, the PTZ is started, and the X-axis and Z-axis position points built into the main control system are read. After the reading is completed, the PTZ zoom camera rotates along the X-axis first and then along the Z-axis. After arriving at the designated location, the main control system reads the built-in zoom multiple, and the zoom camera zooms according to the command. After the zoom is completed, the camera takes a photo to determine whether the recognition model is located in the host computer. If the recognition model is located in the host computer, the built-in model of the main control system is called to perform intelligent recognition on the picture; if the recognition model is not located in the host computer, the picture is transmitted to the AI ​​computing box, and the built-in model of the AI ​​computing box is called to perform intelligent recognition on the picture. After the intelligent recognition is completed, the recognition result is obtained, and the recognition result is compared with the preset threshold of the main control system. If the recognition result exceeds the preset threshold of the main control system, it is determined that the device is abnormal, and the explosion-proof sound and light alarm module issues an alarm. The main control system generates a maintenance work order after analysis, and pages the staff through wireless communication. After the cloud platform successfully recognizes the alarm, the pan-tilt camera ends; if the recognition result is within the preset threshold of the main control system, it is determined that the device is normal, and the pan-tilt camera ends.

[0027] When the task content is determined to be thermal infrared imaging recognition, the thermal infrared imager is started to take pictures, and the pictures are intelligently recognized to obtain the recognition results. The recognition results are compared with the preset threshold of the main control system. If the recognition result exceeds the preset threshold of the main control system, it is determined that the device is abnormal, and the explosion-proof sound and light alarm module issues an alarm prompt. The main control system generates a maintenance work order after analysis, and the staff is paged through wireless communication. After the cloud platform successfully recognizes the alarm, the thermal infrared imaging detection ends; if the recognition result is within the preset threshold of the main control system, it is determined that the device is normal, and the thermal infrared imager detection ends.

[0028] After the task point is completed, determine whether the task at the task location is completed. If the task at the task location is not completed, continue to perform the above business process for the task content at that location; if the task at the task location is completed, enter the next determination process. Determine whether there are tasks at other locations. If there are tasks at other locations, re-plan the route to the next task location. After arriving at the task location, execute the above business process until there are no tasks at other locations and this inspection task is completed; if there are no tasks at other locations, this inspection task is completed. After the inspection task is completed, the intelligent inspection robot returns to the charging warehouse according to the planned route, returns to the designated charging point for charging, and waits for the release of the next task instruction.

[0029] During the entire process of executing the task, the intelligent inspection robot must perform self-power checks at all times. If it is determined that the power is insufficient, it will automatically return to the charging warehouse for charging, and the inspection task will be completed. If there is no problem of insufficient power, the task will continue.

[0030] The intelligent inspection robot is controlled by the background. The background stores known routes, task locations and existing detection methods. According to work requirements, the background specifies the task point detection work and sorts the detection work. Sort by task order and task point order. After sorting, save the inspection task, upload the task to the cloud platform, and store it in the database. The executors of the specified work will issue them regularly according to the task requirements, and the intelligent inspection robot will execute the commands in the order of the task content and detect the equipment.

[0031] The intelligent inspection robot adopts front-end recognition. The main control system carried by the intelligent inspection robot stores the preset task content and the threshold value of the pre-stored device equipment parameter. According to the preset task content stored in the main control system, the inspection robot can quickly complete the intelligent recognition on site. After the intelligent recognition is completed, the intelligent recognition parameters are quickly compared with the threshold value according to the pre-stored device equipment parameter threshold value stored in the main control system. If the situation is abnormal, an on-site alarm is issued. The front-end recognition method used by the present invention improves the recognition and detection speed.

[0032] The intelligent inspection robot can perform multiple tasks at a time. The contents of multiple tasks do not interfere with each other. The task content and task path are decoupled and are only related to the task point. Tasks can be inserted at the same task point at will. This process is completed in the background. There are already task points in the background. After selecting the task point, the task content is randomly arranged in strips, and then the task instructions are sent to the intelligent inspection robot. After reading the task instructions, the intelligent inspection robot executes them in the order of the task content in the background. Decoupling the task content and the task path makes the changes between the task content and the task path independent of each other. Decoupling can also make the system more flexible. When the task content needs to be modified or replaced, it will not affect the task path. When the task path needs to be modified or replaced, it will not affect the task content, making the system more flexible and scalable.

[0033] After receiving the task instructions issued by the background, the intelligent inspection robot conducts self-inspection. After passing the self-inspection, it goes to the task location according to the global path modeling. After arriving at the task location, it selects different detection methods according to the task content for detection, intelligently identifies the detection results, compares the results with the pre-stored thresholds in the main control system, and makes a judgment on the spot. If the situation is abnormal, an alarm is issued. If the situation is normal, the task is completed. After the task is completed, it is determined whether the task of the task point is completed. If not, continue to execute the above detection process according to the task content. If completed, it will start to determine whether there are tasks at other locations. If there are tasks at other locations, go to the next task location according to the global path modeling to execute the above process. If there are no tasks at other locations, all tasks are completed and returned to the charging warehouse.

[0034] The intelligent inspection robot uses front-end recognition technology to achieve rapid intelligent recognition and equipment detection through preset tasks and equipment parameter thresholds stored in the main control system. When the intelligent inspection robot completes intelligent recognition, it will immediately compare it with the preset equipment parameter thresholds. If an abnormality is found, it will immediately issue an on-site alarm. This front-end recognition method improves the speed of recognition and detection.

[0035] The intelligent inspection robot can perform multiple tasks at the same time. These tasks are independent of each other. There is no coupling relationship between the task content and the execution path, which is only related to the task point. Users can insert tasks at the same task point at will, and this process is completed in the background. In the background, users can select the task point and arrange the task content in strips, and then send the task instructions to the intelligent inspection robot. After reading the task instructions, the intelligent inspection robot executes the tasks in the order of the task content in the background. By decoupling the task content and the execution path, the system becomes more flexible. When the task content needs to be modified or replaced, it will not affect the execution path; similarly, when the execution path needs to be modified or replaced, it will not affect the task content. This decoupling design makes the system more flexible and scalable.

[0036] After receiving the task instructions issued by the background, the intelligent inspection robot will perform self-inspection. After passing the self-inspection, it will go to the designated task location according to the global environment plan. After arriving at the task location, it will select the appropriate detection method according to the task content for detection and intelligent identification. The detection results will be compared with the thresholds pre-stored in the main control system. If an abnormality is found, an on-site alarm will be issued; if the situation is normal, the task will be completed. After the task is completed, it will be determined whether the tasks of the task point have been completed. If not, the above detection process will continue; if it has been completed, it will be determined whether there are tasks at other locations. If there are tasks at other locations, it will go to the next task location according to the global environment plan to execute the above process; if there are no tasks at other locations, all tasks will be completed and the inspection robot will return to the charging warehouse. This business process is concise and comprehensive, and can complete the inspection tasks well.

[0037] The communication solution used in the present invention is a 5G communication module. Other mobile communication technologies + short-range wireless transmission technologies + Bluetooth solutions can replace this communication module solution.

[0038] The positioning solution used in the present invention is Beidou positioning, and other wireless positioning methods such as base station positioning, GPS positioning, Bluetooth positioning, etc. can replace this positioning solution;

[0039] The multi-sensor comprehensive perception solution used in the present invention is an ultrasonic obstacle avoidance sensor, an infrared imaging sensor, a zoom camera, and a 3D lidar solution. Other types of ultrasonic obstacle avoidance sensors, thermal infrared imaging sensors, cameras, 3D lidars, etc. can replace this multi-sensor comprehensive perception solution.

[0040] The protection scope of the present invention is not limited to the above-mentioned embodiments. Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the scope and spirit of the present invention. If these changes and modifications fall within the scope of the claims of the present invention and their equivalents, the present invention is intended to include these changes and modifications.

Claims

1. An intelligent detection robot for chemical industry site, characterized in that: The intelligent detection robot comprises a main control system, a walking module, a communication module, a positioning module, an alarm module, a camera and an infrared imaging sensor. The intelligent detection robot is configured to operate in the following manner: receiving a task instruction, the task instruction indicating a plurality of task locations and one or more task contents at each task location; Arrive at each mission location in turn, and complete all missions at each mission location before proceeding to the next mission location; After completing all the tasks at all the mission locations, return to the charging station; Among them, the execution process of the task content includes: when the detection task content is pan-tilt photography, the target equipment at the task location is photographed, and the instrument value of the target equipment in the photo is identified. If the instrument value meets the set standard, the task content is completed; otherwise, an alarm is issued, and a maintenance work order is generated and sent to the background. The task content is completed after receiving the signal of successful identification alarm sent by the background; when the detection task content is thermal infrared imaging detection, thermal infrared photography is taken at the task location to identify the temperature value. If the temperature value meets the set standard, the task content is completed; otherwise, an alarm is issued, and a maintenance work order is generated and sent to the background. The task content is completed after receiving the signal of successful identification alarm sent by the background.

2. The intelligent detection robot according to claim 1, characterized in that: When the task content is to use the gimbal to take photos, the task content indicates the x-axis and z-axis position points of the gimbal's rotation and the zoom multiple.

3. The intelligent detection robot according to claim 1, characterized in that: The instrument value of the target device in the photo is recognized by the recognition model built into the main control system or by the AI ​​computing box to which the intelligent inspection robot is connected in communication.

4. The intelligent inspection robot according to claim 1, characterized in that: After receiving the business content, it also plans the forward route according to the multiple task locations indicated by the business content.

5. The intelligent inspection robot according to claim 1, characterized in that: After receiving the service content, it also performs a self-check operation.

6. A system, comprising an intelligent detection robot for a chemical site according to any one of claims 1 to 5, and also comprising a background, wherein the background is configured to: in response to user operations, insert task content at a task location, sort the task content at the task location, then package the task content of each task location as a whole to form a task instruction, and send the task instruction to the intelligent detection robot.

Images