WO2020140316A1 - 一种涂料生产安全智能巡检机器人、系统及方法 - Google Patents

一种涂料生产安全智能巡检机器人、系统及方法 Download PDF

Info

Publication number
WO2020140316A1
WO2020140316A1 PCT/CN2019/075903 CN2019075903W WO2020140316A1 WO 2020140316 A1 WO2020140316 A1 WO 2020140316A1 CN 2019075903 W CN2019075903 W CN 2019075903W WO 2020140316 A1 WO2020140316 A1 WO 2020140316A1
Authority
WO
WIPO (PCT)
Prior art keywords
module
inspection
information
data
robot
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/075903
Other languages
English (en)
French (fr)
Inventor
翟懿奎
陈丽燕
吴细
邓文博
柯琪锐
徐颖
甘俊英
应自炉
曾军英
秦传波
麦超云
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuyi University Fujian
Original Assignee
Wuyi University Fujian
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuyi University Fujian filed Critical Wuyi University Fujian
Publication of WO2020140316A1 publication Critical patent/WO2020140316A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
    • G01K1/00—Details of thermometers not specially adapted for particular types of thermometer
    • G01K1/02—Means for indicating or recording specially adapted for thermometers
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
    • G01N15/06—Investigating concentration of particle suspensions
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions

Definitions

  • the invention relates to the technical field of inspection and inspection, in particular to an intelligent inspection robot, system and method for paint production safety.
  • the paint production industry is a chemical industry.
  • the production of paints belongs to the production process of hazardous chemicals.
  • the raw materials are mostly all kinds of dangerous chemicals.
  • the current inspection of the workshop of the paint industry is obviously inadequate, safety accidents in the production process occur frequently, and automatic inspection cannot be carried out during the production process of the paint. Need to consume more human resources.
  • a sensor needs to be provided to detect the temperature, toxic gas concentration, and dust concentration in the workshop.
  • the existing detection method is to install the sensor in the workshop as much as possible, and the position distribution of the sensor needs to be considered, and the sensor is exposed for a long time. In the workshop environment, it is easy to damage, resulting in increased costs. At the same time, the existing detection methods cannot predict and analyze the collected temperature information, toxic gas concentration, and dust concentration.
  • the object of the present invention is to provide a safe and intelligent inspection robot, system and method for paint production.
  • the inspection robot can perform automatic inspection and obstacle avoidance, reduce the use of manpower, save costs, and at the same time inspect the robot
  • the sensor is installed on it, which can detect the temperature, toxic gas concentration and dust concentration in the workshop under the driving of the inspection sensor, and perform predictive analysis to ensure the safety of the workshop.
  • the present invention provides a safe and intelligent inspection robot for paint production, including a visual system for automatic navigation and obstacle avoidance, a sensor detection system for detecting the temperature of the workshop, the concentration of toxic gases and dust concentration, and A prediction system for predicting and analyzing temperature, toxic gas concentration, and dust concentration, the sensor detection system is connected to the prediction system;
  • the visual system includes a data collection module for collecting point cloud data of the inspection environment, and a construction inspection
  • a map construction module for an environmental three-dimensional map, a path planning module for planning inspection paths, a target recognition and positioning module for identifying and locating target objects, and a control module for data processing, the data collection module is constructed separately from the map
  • the module, the path planning module and the target positioning recognition module are connected, and the control module is data-connected to the map construction module, the path planning module and the target recognition positioning module, respectively.
  • the vision system further includes a transmission module for data transmission, and the control module receives the data transmitted by the map construction module, the path planning module, and the target recognition positioning module through the transmission module.
  • the sensor detection system includes a dust detection module, a toxic gas detection module, a temperature detection module, a signal processing module, and a data fusion module.
  • the signal processing module converts the information detected by the dust detection module, toxic gas detection module, and temperature detection module It is converted into electrical signals and transmitted to the data fusion module, which is connected to the prediction system.
  • the prediction system includes a prediction module based on data mining and an information display module for displaying temperature prediction information, dust prediction information, and toxic gas prediction information.
  • the prediction module is connected to the sensor detection system and the information display module, respectively.
  • the power detection module for detecting whether the power is too low, and the power detection module is connected to the path planning module.
  • an alarm system which is respectively connected to the power detection module and the prediction system.
  • the present invention also proposes a paint production safety and intelligent inspection system, which includes the paint production safety and intelligent inspection robot as described above, and also includes an intelligent cooling and exhaust system for removing workshop exhaust gas and reducing heat,
  • the intelligent cooling exhaust system is connected to the prediction system.
  • it also includes a charging position set on the inspection route of the inspection robot, and when the power of the inspection robot is low, it moves to the charging position for charging.
  • the present invention also proposes a safe and intelligent inspection method for paint production, including the following steps:
  • An object recognition algorithm based on convolutional neural network fusion of three-dimensional features is used to process the collected point cloud data, identify and locate the target object, and feed back the positioning information to the three-dimensional map;
  • the inspection robot collects temperature information, toxic gas concentration information and dust concentration information in the inspection environment during the inspection process, and uses data mining-based algorithms to predict temperature information, toxic gas concentration information and dust concentration information.
  • the one or more technical solutions provided in the embodiments of the present invention have at least the following beneficial effects: by collecting inspection environment information, constructing a three-dimensional map of the inspection environment, planning the best inspection path of the inspection, and waiting for The detected target is displayed on the three-dimensional map after positioning, which can enable the inspection robot to perform inspection according to the set inspection path, realize automatic inspection and obstacle avoidance, reduce the use of manpower, and at the same time collect the inspection during the inspection process
  • the temperature information, toxic gas concentration information and dust concentration information in the inspection environment, and predictive analysis of the collected information are used to predict the changes after these information to ensure the safety of the inspection environment.
  • FIG. 1 is a schematic diagram of an embodiment of a safe and intelligent inspection machine for paint production of the present invention
  • FIG. 2 is a schematic diagram of an embodiment of a paint production safety intelligent inspection system of the present invention
  • FIG. 3 is a flowchart of an embodiment of a safe and intelligent inspection method for paint production according to the present invention.
  • the paint production industry is a chemical industry.
  • the production of paints belongs to the production process of hazardous chemicals.
  • the raw materials are mostly all kinds of dangerous chemicals.
  • certain substances are likely to generate heat due to the physical, chemical and biochemical reactions that occur inside.
  • the accumulation of heat makes the temperature of combustibles higher and higher. When the temperature reaches a certain level, the combustible gas or toxic gas just reaches When it is thick, a combustion or explosion accident will occur, which reflects the unsafety of the paint production; the operator must be by the equipment when performing manual operations.
  • the toxic gas is easy to cause personal injury, and it is not easy to detect during the operation
  • the hidden dangers can not be avoided during the outbreak, threatening the life safety of the operator; a large amount of mineral powder is used in the production process, and long-term exposure will endanger the health.
  • the dust hazard is mainly in the batching position. If the dust concentration in the air is too high, or The air penetrates into the equipment pipeline through the crack, forming a flammable mixture and causing an explosion. Therefore, enterprises should pay great attention to production safety in order to ensure the normal operation and production efficiency of the entire production workshop;
  • the traditional safety detection method is in the workshop Install as many sensors as possible. In this way, the position distribution of the sensors needs to be considered, and the sensors are exposed to the workshop environment for a long time, which is easy to damage, resulting in increased costs.
  • the present invention provides a safe and intelligent inspection robot, system and method for paint production.
  • the inspection robot can automatically plan the inspection route, avoid obstacles in the inspection environment, reduce the use of manpower, and save costs
  • the sensor is set on the inspection robot.
  • the inspection robot can collect temperature information, toxic gas concentration information and dust concentration information in the inspection environment, and predict and analyze these information. If the parameter has a trend higher than the normal range within a certain period of time, you can notify the staff to deal with it accordingly.
  • an embodiment of the present invention provides a safe and intelligent inspection robot for paint production, including a visual system for automatic navigation and obstacle avoidance, and a sensor for detecting the temperature of the workshop, the concentration of toxic gas and the concentration of dust A system and a prediction system for predicting and analyzing temperature, toxic gas concentration and dust concentration, the sensor detection system is connected to the prediction system;
  • the vision system includes a data collection module 11 for collecting point cloud data of the inspection environment A map construction module 12 for constructing a three-dimensional map of the inspection environment, a path planning module 13 for planning an inspection path, a target recognition and positioning module 14 for identifying and positioning target objects, and a control module 15 for data processing,
  • the data collection module 11 is respectively connected to the map construction module 12, the path planning module 13 and the target positioning recognition module, and the control module 15 is respectively connected to the map construction module 12, the path planning module 13 and the target recognition positioning module 14.
  • the inspection robot can realize automatic inspection, obstacle avoidance and other functions through the vision system, which can reduce the use of manpower and reduce costs.
  • the temperature of each place in the inspection environment can be collected Information, toxic gas concentration information, and dust concentration information, etc., to carry out predictive analysis of these collected information to predict future trends. If the relevant parameter data has a trend higher than the normal range within a certain period of time, you can notify the staff to carry out Handle accordingly.
  • the data collection module 11 is composed of an RGBD camera, used to collect point cloud data of the inspection environment, and the RGBD camera can control the rotation angle to achieve 360-degree data collection of the inspection environment.
  • the map construction module 12 uses point cloud fusion Technology, construct a three-dimensional map of the inspection environment based on the point cloud data collected by the data collection module 11,
  • the path planning module 13 uses the artificial potential field method to calculate the distance to the obstacle based on the point cloud data collected by the data collection module 11, And plan the best path to reach the target position
  • the target recognition and positioning module 14 uses an object recognition algorithm based on convolutional neural network fusion of three-dimensional features to identify and locate the target object, and feedback the positioning information to the three-dimensional map to inspect
  • the robot can move to the corresponding target according to the corresponding inspection path, to achieve a fully automatic inspection process, without manual participation, reducing the use of manpower.
  • the sensor detection system includes multiple sensors for collecting different information, such as temperature information, toxic gas concentration information, dust concentration information, etc., the sensor is set on the surface of the inspection robot, in the process of inspection robot inspection Collect all kinds of information, and then transfer the collected data to the forecasting system.
  • the forecasting system performs predictive analysis to predict the future development trend of each parameter. When a certain parameter data has a trend higher than the normal range within a certain period of time At the time, you can notify the staff to take corresponding treatment to ensure the safety of the inspection environment, such as exhaust and cooling.
  • another embodiment of the present invention also provides a safe and intelligent inspection robot for paint production
  • the vision system further includes a transmission module 16 for data transmission
  • the control module 15 receives the map through the transmission module 16 The data transmitted by the building module 12, the path planning module 13, and the target recognition positioning module 14.
  • the transmission module 16 is connected to the control module 15 for transmitting the data transmitted by the map construction module 12, the path planning module 13 and the target recognition positioning module 14, and the control module 15 sends driving information according to the received data to control The inspection robot performs inspection according to the planned path.
  • the sensor detection system includes a dust detection module 21, a toxic gas detection module 22, a temperature detection module 23, a signal processing module 24, and The data fusion module 25, the signal processing module 24 converts the information detected by the dust detection module 21, the toxic gas detection module 22, and the temperature detection module 23 into electrical signals, and transmits them to the data fusion module 25, the data fusion module 25 Connect with prediction system.
  • the data detected by the dust detection module 21, the toxic gas detection module 22, and the temperature detection module 23 are first transmitted to the signal processing module 24 for preprocessing, and then the preprocessed information is processed by the data fusion module 25 For data fusion, the information after data fusion is transmitted to the prediction system for prediction and analysis.
  • the dust detection module 21 mainly includes a new micro chip, a dust sensitive probe, a single chip microcomputer module, an electronic processing unit, a micro battery, a low-battery alarm module, a self-checking module of its own state, and its working principle is: 1) light scattering method; 2) LED emission The light meets the dust and generates reflected light; 3) The receiving sensor monitors the light intensity of the reflected light and outputs a signal; 4) The dust concentration is determined by the intensity of the light intensity. Based on the above principle, four sensitive probes are installed. When the dust particles pass near the probe , A weak electric quantity is sensed by the dust sensitive probe and transmitted to the electronic processing unit. The electronic processing unit converts the dust-related signals into electrical signals, which are processed again by the single-chip module, and finally can be displayed on the led digital tube in the form of values.
  • the new microchip of the dust detection module 21 can receive the signal from the dust sensitive probe about the change of the dust concentration in the production workshop and send it to the signal processing module 24. If the signal processing module 24 receives abnormal signals, the new microchip The signal data will be automatically saved, and the low-battery alarm module will activate the breathing lamp of the micro-battery in the dust detection module 21 when the battery is low, prompting to charge, and at the same time the inspection robot can return to the charging position by itself to charge;
  • the self-checking module of its own state can also realize self-checking, self-calibration and self-diagnostic functions. Ordinary sensors need to be regularly checked and calibrated to ensure that it has sufficient accuracy during normal use.
  • the toxic gas detection module 22 includes a high-performance sensitive element module, a high-brightness LED display module, a basic circuit structure module, an explosion-proof connection thread module, a single-chip microcomputer module, an analog voltage/current and serial port module, a micro battery, a new type semiconductor chip, and a low battery alarm Module, self-state self-inspection module, this module can detect a variety of toxic gases, the working process of the toxic gas detection module 22 is: the detection probe obtains the concentration data signal of toxic gases (including volatile organic compounds) in the workshop, and through the single-chip microcomputer The module converts it into a digital signal, and then the new semiconductor chip transmits the toxic gas concentration information to the signal processing module 24.
  • the data can also be displayed on the high-brightness LED display module.
  • the signal processing module 24 is abnormal, The chip will automatically save the signal data.
  • the micro battery of the toxic gas detection module 22 is low, the inspection robot can return to the charging position to charge itself.
  • the toxic gas detection module 22 of this embodiment is less affected by environmental interference and costs more Low, suitable for mass production, long life, small size, accuracy of 0.03%, after debugging, you can work efficiently.
  • the temperature detection module 23 includes an accurate temperature measurement module, a single-chip module, a drive circuit module and a display circuit module, a new chip module, a micro battery, a low-battery early warning module, and a self-test module.
  • the module accurately and timely senses the temperature change data signal of the workshop in the production of paint.
  • the single-chip module the sensed temperature change signal can be converted into a digital signal, and the temperature change can be visually displayed on the led module, which is convenient According to the inspection of the staff, the new chip module uploads the data signal about the temperature change to the signal processing module 24.
  • the low-battery warning module sends an early-warning signal when the micro battery is low, and the inspection robot can automatically return to the charging position for charging.
  • Each module in the temperature detection module 23 of the embodiment efficiently and orderly executes its respective functions, the working temperature is between -10 degrees Celsius and 80 degrees Celsius, and is suitable for use in the robot body.
  • the design is small, easy to maintain, and has stable performance.
  • the resolution accuracy is 0.01 degrees Celsius.
  • the signal processing module 24 can perform preprocessing such as amplification, filtering and A/D conversion on the received information, and then transmit it to the data fusion module 25.
  • the data fusion module 25 can perform data fusion on the received information, including three levels, 1 Data level fusion: for the data collected by the sensor detection module, depending on the sensor type, the same type of data fusion; 2 feature level fusion: extract the feature vector contained in the collected data to reflect the attributes of the monitored physical quantity, which is oriented to monitoring Object feature fusion; 3 Decision layer fusion: According to the data features obtained by feature level fusion, certain discrimination, classification and simple logical operations are performed, and higher-level decisions are made according to application requirements. It is an advanced fusion. Data fusion module 25 Related algorithms will be used to achieve the fusion of massive data information from various sensors, and the fusion will be sent to the prediction system.
  • the prediction system includes a prediction module 31 based on data mining and displays temperature prediction information, dust prediction information, and toxic gas prediction
  • the information display module 32 of the information, the prediction module 31 is respectively connected to the sensor detection system and the information display module 32.
  • the prediction module 31 receives the information sent by the data fusion module 25, uses a data mining-based algorithm to predict and analyze temperature information, dust concentration information, and toxic gas concentration information, and uses artificial intelligence, machine learning, and patterns Recognition, statistics, database, visualization technology and related algorithms and other technologies, highly automated analysis, synthesis, training, induction and reasoning of the received data to predict the future development trend of each type of detection information, and in the form of charts
  • the information display module 32 shows it clearly and clearly.
  • the staff can be notified to perform the corresponding processing, and the prediction module 31 based on data mining can play
  • the role of preventing accidents to a great extent reduces the probability of safety accidents, reduces costs, and has strong reliability. It maximizes the role of paint production safety monitoring and ensures the efficient production of paint.
  • another embodiment of the present invention also provides a paint production safety intelligent inspection robot, which further includes a power detection module 4 for detecting whether the power is too low, the power detection module 4 is connected to the path planning module 13 .
  • the power detection module 4 can be used to detect whether the battery power of the inspection robot is sufficient to support the power for one week of inspection, so the power detection module 4 is also connected to the path planning module 13.
  • another embodiment of the present invention also provides a safe and intelligent inspection robot for paint production, which further includes an alarm system 5, which is respectively connected to the electricity detection module 4 and the prediction system.
  • the alarm system 5 includes a low battery alarm module, a sensor alarm module, and a voice output module.
  • the low battery alarm module is connected to the power detection module 4 and the sensor alarm module is connected to the prediction module 31.
  • the alarm is generated by the low battery alarm module, and the voice output module generates a sound for alarm, and the sensor alarm module can receive the information sent by the prediction module 31.
  • the prediction module 31 issues an alarm signal
  • the sensor detection module can alarm and simultaneously Send control signals to the subsequent exhaust cooling system for cooling and exhaust treatment of the inspection environment.
  • the voice output module includes a noise sensor and a volume control module 15.
  • the volume value in the voice output unit is increased, or when it is lower than the second threshold, the volume value in the output unit is decreased.
  • a safe and intelligent inspection robot for paint production which includes a vision system, a sensor detection system, a prediction system, a power detection module 4 and an alarm system 5.
  • the vision system includes a data collection module 11 , Map construction module 12, path planning module 13, target recognition positioning module 14, control module 15 and transmission module 16,
  • the sensor detection system includes a dust detection module 21, a toxic gas detection module 22, a temperature detection module 23, a signal processing module 24 and
  • the prediction system includes a prediction module 31 and an information display module 32
  • the data acquisition module 11 is connected to the map construction module 12, the path planning module 13, and the target recognition and positioning module 14, respectively, and the control module 15 is respectively connected to the map through the transmission module 16
  • the building module 12, the path planning module 13, and the target recognition and positioning module 14 are connected
  • the signal processing module 24 is connected to the dust detection module 21, the toxic gas detection module 22, the temperature detection module 23, and the data fusion module 25, respectively
  • the power collection module is respectively connected to the alarm
  • the system 5 and the path planning module 13
  • the inspection robot can collect data on the inspection environment through the vision system, construct a three-dimensional map of the inspection environment and plan out the inspection path and identify and locate the target object, and then the inspection robot can according to different target objects
  • the inspection is carried out according to the set inspection path without manual participation, which reduces the use of manpower and costs.
  • various information on the inspection route can be collected through the sensor detection system, including temperature information, toxic gas concentration information and dust concentration information, etc.
  • the prediction module 31 will make a trend prediction, predicting a period of time in the future If a certain parameter is higher than the normal range within a certain period of time predicted, the alarm system 5 can be used to notify the staff to perform corresponding processing.
  • the inspection robot of this embodiment is also provided with The power detection module 4 can detect whether the inspection robot has enough power. When the power is not enough for the inspection robot to conduct a week of inspection, an alarm can be generated to prompt the need for charging. Multiple charging piles can be set on the inspection path to allow The inspection robot can automatically return to the charging pile for charging.
  • another embodiment of the present invention also provides a paint production safety and intelligent inspection system, which includes the paint production safety and intelligent inspection robot in any of the embodiments described above, and also includes Intelligent cooling exhaust system 6 for workshop exhaust gas and reducing heat, the intelligent cooling exhaust system 6 is connected to a prediction system.
  • the prediction system can send an alarm signal to the intelligent cooling exhaust system 6, the intelligent exhaust cooling system will send a signal to the intelligent exhaust fan and cooling device, and after receiving the signal, the intelligent exhaust fan and cooling device will switch to the low level In the wind exhaust mode or low-level cooling mode, when the prediction system still sends an alarm signal, the intelligent exhaust cooling system will send a signal to the exhaust device and the cooling device. After receiving the signal, the intelligent exhaust fan and cooling device switch to the emergency response mode. Under normal circumstances, through the intelligent exhaust cooling system to eliminate exhaust fumes and reduce heat, enhance air flow, the cooling device can inject fresh air into the room, and the pressure inside the room will increase, and the turbid gas inside will be filtered and discharged outside. So as to achieve the purpose of cooling and ventilation.
  • another embodiment of the present invention also provides a safe and intelligent inspection system for paint production, which further includes a charging position provided on an inspection route of the inspection robot, and the inspection robot moves to the charging position when the battery power is low Charging.
  • the charging position may be a charging pile
  • the inspection robot may automatically return to the charging pile for charging when the power is insufficient
  • the charging pile may be set on the inspection path to facilitate charging of the inspection robot.
  • another embodiment of the present invention also provides a safe and intelligent inspection method for paint production, including the following steps:
  • An object recognition algorithm based on convolutional neural network fusion of three-dimensional features is used to process the collected point cloud data, identify and locate the target object, and feed back the positioning information to the three-dimensional map;
  • a three-dimensional map, inspection path, and target object of the inspection environment can be obtained, and different inspection paths can be set for different target objects
  • an object recognition algorithm based on convolutional neural network fusion of three-dimensional features is used.
  • the specific steps are: take the three-dimensional features of the target object as input, train the model features of the target object to be recognized, and use the Softmax classifier As an output, identify the target object in the inspection process, and accurately locate the three-dimensional coordinates of the target object relative to the data collection module 11 according to the mapping relationship between the target object and the data collection module 11, and match the two
  • the characteristics of the key points of the author determine the position of the target object in the stereo map, and highlight the target object in the stereo map.
  • another embodiment of the present invention also provides a safe and intelligent inspection method for paint production.
  • the inspection robot collects temperature information, toxic gas concentration information, and dust concentration information in the inspection environment during the inspection process, and The algorithm based on data mining is used to predict temperature information, toxic gas concentration information and dust concentration information.
  • the inspection robot can collect temperature information, toxic gas concentration information and dust concentration information during the inspection process, and then the prediction system can perform predictive analysis based on the above information to predict the trend of various parameters in the future, If there is an increasing trend of parameters and it is higher than the normal range within a certain period of time, an alarm will be given and the staff can handle it accordingly.

Landscapes

  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Remote Sensing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Automation & Control Theory (AREA)
  • Medicinal Chemistry (AREA)
  • Food Science & Technology (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Dispersion Chemistry (AREA)
  • Manipulator (AREA)
  • General Factory Administration (AREA)

Abstract

提供一种涂料生产安全智能巡检机器人、系统及方法。巡检机器人包括用于实现自动导航和避障的视觉系统、用于检测车间温度、有毒气体浓度和粉尘浓度的传感器检测系统以及用于对温度、有毒气体浓度和粉尘浓度进行预测分析的预测系统，该巡检机器人可以对巡检环境进行信息采集，构建巡检环境的立体地图，将目标物体显示到立体地图上，并规划出巡检的最佳路径，巡检机器人则可以根据不同的目标物体选择不同的巡检路径进行巡检，无需人工参与，同时巡检机器人在巡检的过程中可以采集巡检环境内的温度信息、有毒气体浓度信息以及粉尘浓度信息，并由预测系统进行预测分析，对未来时间内各参数的变化趋势进行预测，保证巡检环境的安全。

Description

一种涂料生产安全智能巡检机器人、系统及方法
技术领域
本发明涉及巡检技术领域，特别是一种涂料生产安全智能巡检机器人、系统及方法。
背景技术
涂料生产行业是一种化工行业，涂料生产属于危化品生产过程，原料多为各种危险化学品，种类繁多且化学性质各异，其中多数具有可燃性和毒性，危险性质较严重，所以需要安排专门的巡检人员对车间内的各个部分进行全面、详细的巡检，然而目前涂料行业的车间安全巡检明显不足，生产过程的安全事故频繁发生，涂料生产过程中不能进行自动巡检，需要耗费较多的人力资源。
同时还需要设置有传感器用于检测车间内的温度、有毒气体浓度以及粉尘浓度等，现有检测方法是将传感器尽可能多的设置在车间内，需要考虑到传感器的位置分布，并且传感器长期暴露在车间环境中，容易损坏，导致成本提高，同时现有的检测方法无法对采集的温度信息、有毒气体浓度、粉尘浓度进行预测分析。
发明内容
为解决上述问题，本发明的目的在于提供一种涂料生产安全智能巡检机器人、系统及方法，巡检机器人可以进行自动巡检、避障，减少人力的使用，节省成本，同时在巡检机器人上设置有传感器，可以在巡检传感器的带动下检测车间内的温度、有毒气体浓度和粉尘浓度，并进行预测分析，保证车间安全。
本发明解决其问题所采用的技术方案是：
第一方面，本发明提出了一种涂料生产安全智能巡检机器人，包括用于实现自动导航和避障的视觉系统、用于检测车间温度、有毒气体浓度和粉尘浓度的传感器检测系统以及用于对温度、有毒气体浓度和粉尘浓度进行预测分析的预测系统，所述传感器检测系统与预测系统连接；所述视觉系统包括用于采集巡检环境点云数据的数据采集模块、用于构建巡检环境立体地图的地图构建模块、用于规划巡检路径的路径规划模块、用于识别和定位目标物体的目标识别定位模块以及用于进行数据处理的控制模块，所述数据采集模块分别与地图构建模块、路径规划模块以及目标定位识别模块连接，所述控制模块分别与地图构建模块、路径规划模块以及目标识别定位模块数据连接。
进一步，所述视觉系统还包括用于进行数据传输的传输模块，所述控制模块通过传输模块接收地图构建模块、路径规划模块以及目标识别定位模块传输的数据。
进一步，所述传感器检测系统包括粉尘检测模块、有毒气体检测模块、温度检测模块、信号处理模块以及数据融合模块，所述信号处理模块将粉尘检测模块、有毒气体检测模块以及温度检测模块检测的信息转换成电信号，并传输到数据融合模块中，所述数据融合模块与预测系统连接。
进一步，所述预测系统包括基于数据挖掘的预测模块以及用于显示温度预测信息、粉尘预测信息和有毒气体预测信息的信息显示模块，所述预测模块分别和传感器检测系统、信息显示模块连接。
进一步，还包括用于检测电量是否过低的电量检测模块，所述电量检测模块与路径规划模块连接。
进一步，还包括报警系统，所述报警系统分别与电量检测模块、预测系统连接。
第二方面，本发明还提出了一种涂料生产安全智能巡检系统，包括有如上所述的涂料生产安全智能巡检机器人，还包括用于排除车间废气和降低热量的智能冷却排风系统，所述智能冷却排风系统与预测系统连接。
进一步，还包括设置在巡检机器人巡检路线上的充电位置，所述巡检机器人电量低时移动到充电位置处进行充电。
第三方面，本发明还提出了一种涂料生产安全智能巡检方法，包括以下步骤：
采集巡检环境的点云数据；
根据点云数据的关键点特征，对点云数据进行逐个匹配，构建巡检环境的立体地图；
根据点云数据获取深度信息，判断障碍物的位置，规划巡检路径；
采用基于卷积神经网络融合三维特征的物体识别算法，对采集的点云数据进行处理，对目标物体进行识别定位，并将定位信息反馈到立体地图上；
控制巡检机器人在规划好的巡检路径上进行巡检。
进一步，巡检机器人在巡检过程中采集巡检环境中的温度信息、有毒气体浓度信息以及粉尘浓度信息，并采用基于数据挖掘的算法对温度信息、有毒气体浓度信息以及粉尘浓度信息进行预测。
本发明实施例中提供的一个或多个技术方案，至少具有如下有益效果：通过采集巡检环境信息，构建好巡检环境的立体地图，规划好巡检的最佳巡检路径，同时将待检测的目标定位后显示到立体地图上，可以使巡检机器人按照设置好的巡检路径进行巡检，实现自动巡检、避障，减少人力的使用，同时在巡检的过程中可以采集巡检环境内的温度信息、有毒气体浓度信息以及粉尘浓度信息，并对这些采集的信息进行预测分析，用于预测这些信息之后的变化情况，保证巡检环境的安全。
附图说明
下面结合附图和实例对本发明作进一步说明。
图1是本发明的一种涂料生产安全智能巡检机器的一个实施例的原理图；
图2是本发明的一种涂料生产安全智能巡检系统的一个实施例的原理图；
图3是本发明的一种涂料生产安全智能巡检方法的一个实施例的流程图。
具体实施方式
涂料生产行业是一种化工行业，涂料生产属于危化品生产过程，原料多为各种危险化学品，种类繁多且化学性质各异，其中多数具有可燃性和毒性，危险性质较严重；而且在化学品生产过程中，某些物质容易由于内部所发生的物理、化学及生物化学反应产生热量，热量的积聚使可燃物温度越来越高，当温度达到一定程度而可燃气体或有毒气体刚好达到一定浓时，就会发生燃烧或爆炸事故，体现了涂料生产的不安全性；操作人员进行手工操作时必定会守在设备旁，有毒气体容易对人身造成伤害，而操作过程中存在不容易察觉的隐患在爆发时也躲避不及，威胁操作人员的生命安全；生产过程中使用大量的矿物粉料，长期接触会危害健康，粉尘危害主要是在配料岗位，空气中的粉尘浓度如果过高，或空气通过裂缝深入到设备管线中，形成易燃混合物而引发爆炸事故。所以企业应极力重视生产安全，才能保证整个生产车间的正常运作和生产效率；
由于有毒气体的种类比较多，所需的传感器也比较多，但目前任意单个传感器所提供的信息是单方面的、不完整的，有可能是错误的、不精确的，传感器精度不相同，所提供信息类别也不相同，通过对多个传感器所提供的信息进行综合处理，将会比单个传感器做出更可靠的判断和决策，并且大大提高检测的精度，传统的安全检测方法是在车间内尽可能多的安装各类传感器，这样需要考虑传感器的位置分布，并且传感器长期暴露在车间环境中，容易损坏，导致成本提高。
目前涂料行业的车间安全巡检明显不足，生产过程的安全事故频繁发生，涂料生产过程中要做到无障碍、不间断人工巡检，具有一定难度，所以需要打造一种智能的巡检机器人来代替人工巡检，可以自主完成车间的安全巡检工作。
基于此，本发明提供了一种涂料生产安全智能巡检机器人、系统及方法，巡检机器人可以自动的规划好巡检路线，避开巡检环境中的障碍物，减少人力的使用，节省成本，同时将传感器设置在巡检机器人上，巡检机器人在巡检的过程中可以采集巡检环境中的温度信息、有毒气体浓度信息以及粉尘浓度信息，并对这些信息进行预测分析，若相关的参数在一定的时间内有高于正常范围的趋势，则可以通知工作人员进行相应的处理。
下面结合附图，对本发明实施例作进一步阐述。
参照图1，本发明的一个实施例提供了一种涂料生产安全智能巡检机器人，包括用于实现自动导航和避障的视觉系统、用于检测车间温度、有毒气体浓度和粉尘浓度的传感器检测系统以及用于对温度、有毒气体浓度和粉尘浓度进行预测分析的预测系统，所述传感器检测系统与预测系统连接；所述视觉系统包括用于采集巡检环境点云数据的数据采集模块11、用于构建巡检环境立体地图的地图构建模块12、用于规划巡检路径的路径规划模块13、用于识别和定位目标物体的目标识别定位模块14以及用于进行数据处理的控制模块15，所述数据采集模块11分别与地图构建模块12、路径规划模块13以及目标定位识别模块连接，所述控制模块15分别与地图构建模块12、路径规划模块13以及目标识别定位模块14数据连接。
在本实施例中，巡检机器人通过视觉系统可以实现自动巡检、避障等功能，可以减少人力的使用，降低成本，同时在巡检的过程中，可以采集巡检环境内各个地方的温度信息、有毒气体浓度信息以及粉尘浓度信息等，对这些采集的信息进行预测分析，预测未来的趋势，若相关的参数数据在一定的时间内有高于正常范围的趋势，则可以通知工作人员进行相应的处理。
其中数据采集模块11由RGBD摄像机构成，用于采集巡检环境的点云数据，并且RGBD摄像机可以控制旋转的角度，实现对巡检环境的360度数据采集，地图构建模块12通过采用点云融合技术，根据数据采集模块11采集的点云数据来构建巡检环境的立体地图，路径规划模块13采用基于人工势场法根据数据采集模块11采集的点云数据计算与障碍物之间的距离，并规划到达目标位置的最佳路径，目标识别定位模块14则是采用基于卷积神经网络融合三维特征的物体识别算法对目标物体进行识别定位，并将定位信息反馈到立体地图上，从而巡检机器人可以根据相应的巡检路径移动到相应的目标处，实现全自动的巡检过程，无需人工参与，减少人力的使用。
而传感器检测系统包括多个传感器，用于采集不同的信息，例如采集温度信息、有毒气体浓度信息、粉尘浓度信息等，将传感器设置在巡检机器人的表面，在巡检机器人巡检的过程中采集各种信息，然后将采集的数据传输给预测系统，由预测系统进行预测分析，预测每一种参数在未来的发展趋势，当某一个参数数据在一定的时间内有高于正常范围的趋势时，可以通知工作人员进行相应的处理，保证巡检环境的安全，例如可以进行排风以及冷却处理。
进一步地，本发明的另一个实施例还提供了一种涂料生产安全智能巡检机器人，所述视觉系统还包括用于进行数据传输的传输模块16，所述控制模块15通过传输模块16接收地图构建模块12、路径规划模块13以及目标识别定位模块14传输的数据。
在本实施例中，传输模块16与控制模块15连接，用于传输地图构建模块12、路径规划模块13以及目标识别定位模块14传输的数据，控制模块15根据接收的数据来发出驱动信息，控制巡检机器人按照规划好的路径进行巡检。
进一步地，本发明的另一个实施例还提供了一种涂料生产安全智能巡检机器人，所述传感器检测系统包括粉尘检测模块21、有毒气体检测模块22、温度检测模块23、信号处理模块24以及数据融合模块25，所述信号处理模块24将粉尘检测模块21、有毒气体检测模块22以及温度检测模块23检测的信息转换成电信号，并传输到数据融合模块25中，所述数据融合模块25与预测系统连接。
在本实施例中，粉尘检测模块21、有毒气体检测模块22、温度检测模块23检测的数据会先传输到信号处理模块24中进行预处理，然后再由数据融合模块25对预处理后的信息进行数据融合，经数据融合后的信息才传输给预测系统中进行预测分析。
粉尘检测模块21主要包括新型微型芯片，粉尘灵敏探头，单片机模块，电子处理单元，微型电池，低电量告警模块，自身状态自检模块，工作原理为：1)光散射方法；2）LED发射出光线遇粉尘产生反射光；3）接收传感器监测到反射光的光强，输出信号；4）通过光强的大小判断粉尘浓度，基于上述的原理，安装四个灵敏探头，当粉尘颗粒经过探头附近，一个微弱的电量被粉尘灵敏探头感应并传导给电子处理单元，电子处理单元将粉尘的有关信号变成了电信号，经由单片机模块再次处理，最后可以以数值的形式显示在led数码管上，
粉尘检测模块21的新型微型芯片可以接收粉尘灵敏探头发出的有关生产车间内粉尘浓度变化的信号并将其发送到信号处理模块24，如信号处理模块24出现接收异常的情况下，该新型微型芯片将自动保存信号数据，低电量告警模块在粉尘检测模块21中的微型电池的电量低的情况下将会启动该模块的呼吸灯，提示进行充电，同时巡检机器人可以自行返回充电位置进行充电；自身状态自检模块也可以实现自检、自校、自诊断功能，普通传感器需要定期检验和标定，以保证它在正常使用时足够的准确度，这些工作一般要求将传感器从使用现场拆卸到实验室或检验部门进行，而粉尘检测模块21可以进行自检，无需额外的定期维护和抽检，降低成本，并且粉尘检测模块21的检测误差不超过0.05%，采用的新型微型充电电池成本较低，无需维护。
有毒气体检测模块22包括高性能敏感元件模块、高亮度led显示模块、基本的电路结构模块、防爆连接螺纹模块、单片机模块、模拟电压/电流和串口模块、微型电池、新型半导体芯片、低电量告警模块、自身状态自检模块，该模块可以检测多种有毒气体，有毒气体检测模块22的工作过程为：检测探头获取车间内的有毒气体（包括挥发性有机化合物）的浓度数据信号，并通过单片机模块将其转换为数字信号，然后由新型半导体芯片将有毒气体浓度信息传输给信号处理模块24，同时也可以将数据在高亮度led显示模块上显示，如信号处理模块24出现异常的情况下，该芯片将自动保存信号数据，若有毒气体检测模块22的微型电池电量较低时，巡检机器人可以自行返回充电位置充电，本实施例的有毒气体检测模块22受环境干扰因素较小，成本较低，适合批量生产，寿命长，体型小，精度为0.03%，经调试后，便可高效工作。
而温度检测模块23包括精确测温模块、单片机模块、驱动电路模块以及显示电路模块、新型芯片模块、微型电池、低电量预警模块以及自检模块组成，工作过程：精确测温模块通过精确测温模块精确且及时感应生产涂料内的车间的温度变化数据信号，通过单片机模块就可以将感应到的有关温度变化的信号转换成数字信号，并将温度变化的情况直观地显示在led模块上，方便有关工作人员查看，新型芯片模块将有关温度变化的数据信号上传到信号处理模块24中，低电量预警模块在微型电池电量低的情况下发出预警信号，巡检机器人可以自动返回充电位置充电，本实施例的温度检测模块23内的各个模块高效地且有条不紊地执行各自的功能，工作温度在-10摄氏度-80摄氏度之间，适合在机器人本体中使用，设计精小，易维护，性能稳定，分辨精度为0.01摄氏度。
信号处理模块24可以对接收的信息进行放大、滤波以及A/D转换等预处理，然后在传输到数据融合模块25，数据融合模块25可以对接收的信息进行数据融合，包括三个层次，①数据级融合：针对传感器检测模块采集的数据，依赖于传感器类型，进行同类数据的融合；②特征级融合：提取所采集数据包含的特征向量，用来体现所监测物理量的属性，这是面向监测对象特征的融合；③决策层融合：根据特征级融合所得到的数据特征，进行一定的判别、分类以及简单的逻辑运算，根据应用需求进行较高级的决策，是高级的融合，数据融合模块25会运用相关算法实现对各个传感器传来的海量数据信息的融合，融合完成后将发送到预测系统。
进一步地，本发明的另一个实施例还提供了一种涂料生产安全智能巡检机器人，所述预测系统包括基于数据挖掘的预测模块31以及用于显示温度预测信息、粉尘预测信息和有毒气体预测信息的信息显示模块32，所述预测模块31分别和传感器检测系统、信息显示模块32连接。
在本实施例中，预测模块31接收数据融合模块25发送的信息，采用基于数据挖掘的算法对温度信息、粉尘浓度信息、有毒气体浓度信息进行预测分析，并利用利用人工智能、机器学习、模式识别、统计学、数据库、可视化技术以及相关算法等技术，对接收数据进行高度自动化地分析、综合、训练、归纳和推理，预测每一种检测信息未来的发展趋式，并以图表的形式在信息显示模块32上简单明了地展现出来，若某一个参数数据在一定的时间内有高于正常范围的趋势，则可以通知工作人员进行相应的处理，通过基于数据挖掘的预测模块31可以起到防患于未然的作用，在很大的程度上减少安全事故发生的概率，降低成本，可靠性强，最大限度地发挥了涂料生产安全监测的作用，保证涂料的高效生产。
进一步地，本发明的另一个实施例还提供了一种涂料生产安全智能巡检机器人，还包括用于检测电量是否过低的电量检测模块4，所述电量检测模块4与路径规划模块13连接。
在本实施例中，电量检测模块4可以用于检测巡检机器人的电池电量是否足够支持巡查一周的电量，所以电量检测模块4还会与路径规划模块13相连接。
进一步地，本发明的另一个实施例还提供了一种涂料生产安全智能巡检机器人，还包括报警系统5，所述报警系统5分别与电量检测模块4、预测系统连接。
在本实施例中，报警系统5包括低电量报警模块、传感器报警模块以及语音输出模块，低电量报警模块与电量检测模块4连接，传感器报警模块与预测模块31连接，当巡检机器人电量过低时，通过低电量报警模块进行报警，同时语音输出模块发生声音进行报警，而传感器报警模块可以接收预测模块31发送的信息，当预测模块31发出报警信号时，传感器检测模块可以进行报警，并同时发送控制信号到后续的排风冷却系统中，用于对巡检环境进行降温排气处理。
语音输出模块包括噪音传感器和音量控制模块15，当车间周围噪音音量大于第一阈值时，提高语音输出单元中的音量值，或者低于第二阈值时，降低输出单元中的音量值。
进一步地，本发明的另一个实施例还提供了一种涂料生产安全智能巡检机器人，包括视觉系统、传感器检测系统、预测系统、电量检测模块4以及报警系统5，视觉系统包括数据采集模块11、地图构建模块12、路径规划模块13、目标识别定位模块14、控制模块15以及传输模块16，传感器检测系统包括粉尘检测模块21、有毒气体检测模块22、温度检测模块23、信号处理模块24以及数据融合模块25，预测系统包括预测模块31以及信息显示模块32，数据采集模块11分别与地图构建模块12、路径规划模块13、目标识别定位模块14连接，控制模块15通过传输模块16分别与地图构建模块12、路径规划模块13、目标识别定位模块14连接，信号处理模块24分别与粉尘检测模块21、有毒气体检测模块22、温度检测模块23以及数据融合模块25连接，电量采集模块分别与报警系统5、路径规划模块13连接，而预测模块31分别与信息显示模块32、报警系统5连接。
在本实施例中，巡检机器人可以通过视觉系统对巡检环境进行数据采集，构建巡检环境的立体地图并规划出巡检路径以及识别定位目标物体，然后巡检机器人可以根据不同的目标物体按照设定好的巡检路径进行巡检，无需人工参与，减少人力的使用，降低成本。
同时通过传感器检测系统可以对巡检路线上的各种信息进行采集，包括温度信息、有毒气体浓度信息以及粉尘浓度信息等，信息采集后会由预测模块31进行趋势预测，预测在后续的一段时间内各个参数的变化，若某一个参数在预测的一定的时间内高于正常范围，则可以通过报警系统5进行报警，通知工作人员进行相应的处理，同时本实施例的巡检机器人还设置有电量检测模块4，可以检测巡检机器人的电量是否充足，当电量不足以使巡检机器人进行巡查一周时，可以发生警报，提示需要进行充电，可以在巡检路径上设置多个充电桩，让巡检机器人可以自动回到充电桩处进行充电。
此外，参照图2，本发明的另一个实施例还提供了一种涂料生产安全智能巡检系统，包括有如上所述任一实施例中的涂料生产安全智能巡检机器人，还包括用于排除车间废气和降低热量的智能冷却排风系统6，所述智能冷却排风系统6与预测系统连接。
在本实施例中，预测系统可以发出报警信号到智能冷却排风系统6中，智能排风冷却系统会向智能排风扇和冷却装置发出信号，接收到信号后智能排风扇和冷却装置将会切换成低级风力排风模式或者低级冷却模式，当该预测系统依然发出报警信号时，智能排风冷却系统将向排风装置和冷却装置发出信号，接收到信号后智能排风扇和冷却装置切换到紧急应对模式，在一般情况下，通过智能排风冷却系统排除车间废气和降低热量,加强空气流动，冷却装置可以将新鲜的空气注入室内，同时室内的压力会增大，里面的浑浊气体会经过过滤排出室外，从而达到降温通风的目的。
进一步地，本发明的另一个实施例还提供了一种涂料生产安全智能巡检系统，还包括设置在巡检机器人巡检路线上的充电位置，所述巡检机器人电量低时移动到充电位置处进行充电。
在本实施例中，充电位置可以是充电桩，巡检机器人在电量不足时，可以自动回到充电桩处进行充电，而充电桩可以设置在巡检路径上，方便巡检机器人的充电。
此外，参照图3，本发明的另一个实施例还提出了一种涂料生产安全智能巡检方法，包括以下步骤：
采集巡检环境的点云数据；
根据点云数据的关键点特征，对点云数据进行逐个匹配，构建巡检环境的立体地图；
根据点云数据获取深度信息，判断障碍物的位置，规划巡检路径；
采用基于卷积神经网络融合三维特征的物体识别算法，对采集的点云数据进行处理，对目标物体进行识别定位，并将定位信息反馈到立体地图上；
控制巡检机器人在规划好的巡检路径上进行巡检。
在本实施例中，通过对巡检环境的点云数据进行采集并进行处理，从而可以得到巡检环境的立体地图、巡检路径以及目标物体，针对不同的目标物体可以设置不同的巡检路径，而对于目标物体的识别定位采用的是基于卷积神经网络融合三维特征的物体识别算法，具体步骤为：以目标物体的三维特征为输入，训练待识别目标物体的模型特征，以Softmax分类器作为输出，识别巡检过程中的目标物体，并根据目标物体与数据采集模块11之间的映射关系，精确定位目标物体相对于数据采集模块11的三维坐标，并通过匹配目标物体与立体地图两者关键点的特征，确定目标物体在立体地图中的位置，在立体地图中高亮显示目标物体。
进一步地，本发明的另一个实施例还提供了一种涂料生产安全智能巡检方法，巡检机器人在巡检过程中采集巡检环境中的温度信息、有毒气体浓度信息以及粉尘浓度信息，并采用基于数据挖掘的算法对温度信息、有毒气体浓度信息以及粉尘浓度信息进行预测。
在本实施例中，巡检机器人在巡检过程中可以采集温度信息、有毒气体浓度信息以及粉尘浓度信息，然后预测系统可以根据上述信息进行预测分析，预测各个参数在未来一段时间内的趋势，若有参数呈增长趋势并在一定时间内高于正常范围时，则进行报警，工作人员可以进行相应的处理。
以上是对本发明的较佳实施进行了具体说明，但本发明并不局限于上述实施方式，熟悉本领域的技术人员在不违背本发明精神的前提下还可作出种种的等同变形或替换，这些等同的变形或替换均包含在本申请权利要求所限定的范围内。

Claims (10)

  1. 一种涂料生产安全智能巡检机器人，其特征在于：包括用于实现自动导航和避障的视觉系统、用于检测车间温度、有毒气体浓度和粉尘浓度的传感器检测系统以及用于对温度、有毒气体浓度和粉尘浓度进行预测分析的预测系统，所述传感器检测系统与预测系统连接；所述视觉系统包括用于采集巡检环境点云数据的数据采集模块、用于构建巡检环境立体地图的地图构建模块、用于规划巡检路径的路径规划模块、用于识别和定位目标物体的目标识别定位模块以及用于进行数据处理的控制模块，所述数据采集模块分别与地图构建模块、路径规划模块以及目标定位识别模块连接，所述控制模块分别与地图构建模块、路径规划模块以及目标识别定位模块数据连接。
  2. 根据权利要求1所述的一种涂料生产安全智能巡检机器人，其特征在于：所述视觉系统还包括用于进行数据传输的传输模块，所述控制模块通过传输模块接收地图构建模块、路径规划模块以及目标识别定位模块传输的数据。
  3. 根据权利要求1所述的一种涂料生产安全智能巡检机器人，其特征在于：所述传感器检测系统包括粉尘检测模块、有毒气体检测模块、温度检测模块、信号处理模块以及数据融合模块，所述信号处理模块将粉尘检测模块、有毒气体检测模块以及温度检测模块检测的信息转换成电信号，并传输到数据融合模块中，所述数据融合模块与预测系统连接。
  4. 根据权利要求1所述的一种涂料生产安全智能巡检机器人，其特征在于：所述预测系统包括基于数据挖掘的预测模块以及用于显示温度预测信息、粉尘预测信息和有毒气体预测信息的信息显示模块，所述预测模块分别和传感器检测系统、信息显示模块连接。
  5. 根据权利要求1所述的一种涂料生产安全智能巡检机器人，其特征在于：还包括用于检测电量是否过低的电量检测模块，所述电量检测模块与路径规划模块连接。
  6. 根据权利要求5所述的一种涂料生产安全智能巡检机器人，其特征在于：还包括报警系统，所述报警系统分别与电量检测模块、预测系统连接。
  7. 一种涂料生产安全智能巡检系统，其特征在于：包括权利要求1-6任一所述涂料生产安全智能巡检机器人，还包括用于排除车间废气和降低热量的智能冷却排风系统，所述智能冷却排风系统与预测系统连接。
  8. 根据权利要求7所述的一种涂料生产安全智能巡检系统，其特征在于：还包括设置在巡检机器人巡检路线上的充电位置，所述巡检机器人电量低时移动到充电位置处进行充电。
  9. 一种应用权利要求7所述涂料生产安全智能巡检系统的涂料生产安全智能巡检方法，其特征在于：包括以下步骤：
    采集巡检环境的点云数据；
    根据点云数据的关键点特征，对点云数据进行逐个匹配，构建巡检环境的立体地图；
    根据点云数据获取深度信息，判断障碍物的位置，规划巡检路径；
    采用基于卷积神经网络融合三维特征的物体识别算法，对采集的点云数据进行处理，对目标物体进行识别定位，并将定位信息反馈到立体地图上；
    控制巡检机器人在规划好的巡检路径上进行巡检。
  10. 根据权利要求9所述的一种涂料生产安全智能巡检方法，其特征在于：巡检机器人在巡检过程中采集巡检环境中的温度信息、有毒气体浓度信息以及粉尘浓度信息，并采用基于数据挖掘的算法对温度信息、有毒气体浓度信息以及粉尘浓度信息进行预测。
PCT/CN2019/075903 2019-01-02 2019-02-22 一种涂料生产安全智能巡检机器人、系统及方法 Ceased WO2020140316A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910000701.8 2019-01-02
CN201910000701.8A CN109752300A (zh) 2019-01-02 2019-01-02 一种涂料生产安全智能巡检机器人、系统及方法

Publications (1)

Publication Number Publication Date
WO2020140316A1 true WO2020140316A1 (zh) 2020-07-09

Family

ID=66405086

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/075903 Ceased WO2020140316A1 (zh) 2019-01-02 2019-02-22 一种涂料生产安全智能巡检机器人、系统及方法

Country Status (2)

Country Link
CN (1) CN109752300A (zh)
WO (1) WO2020140316A1 (zh)

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112362821A (zh) * 2020-11-24 2021-02-12 南京工业大学 一种布置在胶囊制作车间的空气质量巡检机器人及方法
CN112484734A (zh) * 2020-12-08 2021-03-12 中国矿业大学 基于特征提取自适应神经网络和co2的室内人员定位方法
CN113720394A (zh) * 2021-09-08 2021-11-30 苏州融萃特种机器人有限公司 一种智能探测机器人及其搜查方法
CN113739850A (zh) * 2021-09-03 2021-12-03 浙江国净净化科技有限公司 一种基于数据自动处理的巡检采样机器人
CN115062940A (zh) * 2022-06-07 2022-09-16 北京市博汇科技股份有限公司 一种基于数字孪生的发射台智能可视化巡检方法及系统
CN115562212A (zh) * 2022-10-24 2023-01-03 新特能源股份有限公司 一种基于机器学习的化工安全生产预警系统
CN115958990A (zh) * 2021-10-11 2023-04-14 国家电网有限公司 一种v2g充电桩信息诊断与安全防护方法
CN116448276A (zh) * 2022-01-07 2023-07-18 宝山钢铁股份有限公司 一种电炉钢水测温取样机器人测点位置控制方法
CN116993166A (zh) * 2023-09-25 2023-11-03 北京帮安迪信息科技股份有限公司 一种园区安全风险监测方法
CN117405178A (zh) * 2023-12-15 2024-01-16 成都电科星拓科技有限公司 一种自动化检测室内环境参数的移动监测平台及方法
EP4327878A2 (en) 2017-05-01 2024-02-28 Juno Therapeutics, Inc. Combination of a cell therapy and an immunomodulatory compound
CN118211741A (zh) * 2024-05-21 2024-06-18 山东道万电气有限公司 一种基于多路巡检数据的巡检机器人智能调度管理方法
CN118776608A (zh) * 2024-06-24 2024-10-15 湖北中烟工业有限责任公司 一种用于叶丝干燥工序的粉尘防爆监控方法及装置
CN118819209A (zh) * 2024-09-20 2024-10-22 湖州盛世华骏新型材料股份有限公司 一种醋丙乳液制备温度调节方法及系统
CN120255529A (zh) * 2025-06-04 2025-07-04 北京陆航翼龙智能科技有限公司 基于动态避障的巡检机器人路径优化方法和系统

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110209552B (zh) * 2019-05-29 2023-09-22 北京百度网讯科技有限公司 一种设备管理方法和装置
CN110246235B (zh) * 2019-06-18 2021-10-08 广州供电局有限公司 一种基于Hololens混合现实技术的配电房现场巡检方法及系统
CN110442127B (zh) * 2019-07-15 2022-06-17 广州杰赛科技股份有限公司 智能巡检方法、设备和装置
CN110370284A (zh) * 2019-08-07 2019-10-25 北京凌天世纪控股股份有限公司 一种防爆自主巡检机器人的自动化控制系统
CN110456001A (zh) * 2019-08-14 2019-11-15 中国工商银行股份有限公司 检测方法、检测装置和电子设备
CN110909653B (zh) * 2019-11-18 2022-03-15 南京七宝机器人技术有限公司 一种室内机器人自动标定配电室屏柜的方法
CN112034108A (zh) * 2020-09-16 2020-12-04 上海市环境科学研究院 区域污染情况的分析装置、方法及计算机可读存储介质
CN112332541B (zh) * 2020-10-29 2024-07-12 国网山西省电力公司检修分公司 一种用于变电站的监测系统和方法
CN112904844A (zh) * 2021-01-14 2021-06-04 广东核电合营有限公司 核电站智能巡检方法及系统
CN113617575B (zh) * 2021-07-30 2022-07-29 北京戴纳实验科技有限公司 一种自动防爆喷涂房控制系统
CN115705050A (zh) * 2021-08-06 2023-02-17 中能智旷(北京)科技有限公司 智能巡检机器人系统
CN113376331A (zh) * 2021-08-16 2021-09-10 广东科凯达智能机器人有限公司 一种气体检测机器人系统和预警方法
CN114330761A (zh) * 2021-10-26 2022-04-12 中国核电工程有限公司 一种核电厂设备定位、导航智慧终端系统及其应用方法
CN115541031A (zh) * 2022-09-28 2022-12-30 新特能源股份有限公司 一种多晶硅配电室的温度检测方法及装置
CN115830578B (zh) * 2023-02-23 2023-05-23 北京金石视觉数字科技有限公司 物品巡检方法、装置和电子设备
CN116959118A (zh) * 2023-08-24 2023-10-27 广州为乐信息科技有限公司 一种实验安全操作监测分析及规范系统及方法
CN120255491A (zh) * 2025-06-05 2025-07-04 内蒙古自治区烟草公司巴彦淖尔市公司 一种基于ros的数据中心设备巡检机器人

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103823465A (zh) * 2012-11-17 2014-05-28 安徽蓝德集团股份有限公司 一种无线自动巡检报警车
US20160195613A1 (en) * 2015-01-05 2016-07-07 Robert M. Knox Dual Mode Undercarriage Vehicle Inspection System
JP2017133838A (ja) * 2016-01-25 2017-08-03 三菱電機株式会社 設備管理データ抽出装置および設備管理データ抽出方法
CN206598277U (zh) * 2016-08-31 2017-10-31 杭州申昊科技股份有限公司 一种巡检机器人
JP2017223467A (ja) * 2016-06-13 2017-12-21 三菱自動車工業株式会社 車両の運転支援装置
CN108171796A (zh) * 2017-12-25 2018-06-15 燕山大学 一种基于三维点云的巡检机器人视觉系统及控制方法
CN108189043A (zh) * 2018-01-10 2018-06-22 北京飞鸿云际科技有限公司 一种应用于高铁机房的巡检方法及巡检机器人系统

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2879190Y (zh) * 2005-12-22 2007-03-14 辽宁工程技术大学 多传感器瓦斯突出预测仪
CN103995509B (zh) * 2014-05-13 2017-06-20 北京华科数能科技发展有限公司 用于禽舍环境监测的机器人及其监测方法和系统
CN104345020A (zh) * 2014-10-24 2015-02-11 海安能度软件科技有限公司 粉尘检测报警系统
CN205484275U (zh) * 2016-03-24 2016-08-17 无锡泓瑞航天科技有限公司 大空间有害气体自动巡检装置
CN106598052A (zh) * 2016-12-14 2017-04-26 南京阿凡达机器人科技有限公司 一种基于环境地图的机器人安防巡检方法及其机器人
CN106780822A (zh) * 2017-01-09 2017-05-31 安徽杰瑞信息科技有限公司 一种用于工厂的智能巡检系统
CN106873677A (zh) * 2017-03-01 2017-06-20 连京华 一种禽舍环境巡检及调控系统
CN107167139A (zh) * 2017-05-24 2017-09-15 广东工业大学 一种变电站巡检机器人视觉定位导航方法及系统
CN207139822U (zh) * 2017-09-12 2018-03-27 北京中油瑞飞信息技术有限责任公司 数据中心巡检机器人
CN108908372A (zh) * 2018-08-13 2018-11-30 四川桑瑞思环境技术工程有限公司 一种巡检系统

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103823465A (zh) * 2012-11-17 2014-05-28 安徽蓝德集团股份有限公司 一种无线自动巡检报警车
US20160195613A1 (en) * 2015-01-05 2016-07-07 Robert M. Knox Dual Mode Undercarriage Vehicle Inspection System
JP2017133838A (ja) * 2016-01-25 2017-08-03 三菱電機株式会社 設備管理データ抽出装置および設備管理データ抽出方法
JP2017223467A (ja) * 2016-06-13 2017-12-21 三菱自動車工業株式会社 車両の運転支援装置
CN206598277U (zh) * 2016-08-31 2017-10-31 杭州申昊科技股份有限公司 一种巡检机器人
CN108171796A (zh) * 2017-12-25 2018-06-15 燕山大学 一种基于三维点云的巡检机器人视觉系统及控制方法
CN108189043A (zh) * 2018-01-10 2018-06-22 北京飞鸿云际科技有限公司 一种应用于高铁机房的巡检方法及巡检机器人系统

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4327878A2 (en) 2017-05-01 2024-02-28 Juno Therapeutics, Inc. Combination of a cell therapy and an immunomodulatory compound
CN112362821A (zh) * 2020-11-24 2021-02-12 南京工业大学 一种布置在胶囊制作车间的空气质量巡检机器人及方法
CN112484734A (zh) * 2020-12-08 2021-03-12 中国矿业大学 基于特征提取自适应神经网络和co2的室内人员定位方法
CN113739850A (zh) * 2021-09-03 2021-12-03 浙江国净净化科技有限公司 一种基于数据自动处理的巡检采样机器人
CN113739850B (zh) * 2021-09-03 2024-01-19 浙江国净净化科技有限公司 一种基于数据自动处理的巡检采样机器人
CN113720394B (zh) * 2021-09-08 2023-10-31 苏州融萃特种机器人有限公司 一种智能探测机器人及其搜查方法
CN113720394A (zh) * 2021-09-08 2021-11-30 苏州融萃特种机器人有限公司 一种智能探测机器人及其搜查方法
CN115958990A (zh) * 2021-10-11 2023-04-14 国家电网有限公司 一种v2g充电桩信息诊断与安全防护方法
CN116448276A (zh) * 2022-01-07 2023-07-18 宝山钢铁股份有限公司 一种电炉钢水测温取样机器人测点位置控制方法
CN115062940A (zh) * 2022-06-07 2022-09-16 北京市博汇科技股份有限公司 一种基于数字孪生的发射台智能可视化巡检方法及系统
CN115562212A (zh) * 2022-10-24 2023-01-03 新特能源股份有限公司 一种基于机器学习的化工安全生产预警系统
CN116993166A (zh) * 2023-09-25 2023-11-03 北京帮安迪信息科技股份有限公司 一种园区安全风险监测方法
CN117405178A (zh) * 2023-12-15 2024-01-16 成都电科星拓科技有限公司 一种自动化检测室内环境参数的移动监测平台及方法
CN117405178B (zh) * 2023-12-15 2024-03-15 成都电科星拓科技有限公司 一种自动化检测室内环境参数的移动监测方法
CN118211741A (zh) * 2024-05-21 2024-06-18 山东道万电气有限公司 一种基于多路巡检数据的巡检机器人智能调度管理方法
CN118776608A (zh) * 2024-06-24 2024-10-15 湖北中烟工业有限责任公司 一种用于叶丝干燥工序的粉尘防爆监控方法及装置
CN118819209A (zh) * 2024-09-20 2024-10-22 湖州盛世华骏新型材料股份有限公司 一种醋丙乳液制备温度调节方法及系统
CN118819209B (zh) * 2024-09-20 2025-01-21 湖州盛世华骏新型材料股份有限公司 一种醋丙乳液制备温度调节方法及系统
CN120255529A (zh) * 2025-06-04 2025-07-04 北京陆航翼龙智能科技有限公司 基于动态避障的巡检机器人路径优化方法和系统

Also Published As

Publication number Publication date
CN109752300A (zh) 2019-05-14

Similar Documents

Publication Publication Date Title
WO2020140316A1 (zh) 一种涂料生产安全智能巡检机器人、系统及方法
CN109882244B (zh) 一种井下巡检机器人智能建图系统
CN113478500B (zh) 一种基于智能巡检机器人的船舶机舱多源数据收集系统
CN107437080A (zh) 一种综合管廊动态智能巡检系统
CN107831387A (zh) 一种电力站设备运行状况智能监测方法
CN113086549A (zh) 一种火电厂输煤皮带多智能体协同监测系统
CN105574683A (zh) 一种全方位变电站巡检系统及方法
US20240019151A1 (en) Unmanned inspection system and method of HVAC room based on artificial intelligence
US20260095033A1 (en) Electric patrol inspection control method and system based on unmanned aerial vehicle
CN114137302B (zh) 一种电能计量器具检定全过程监控系统
CN115056236A (zh) 一种电厂智能巡检机器人
CN105629840A (zh) 一种集成多传感器的管道爬行智能检测设备
CN102494723A (zh) 一种无人值守环境试验系统和方法
CN115741746A (zh) 一种智能巡检系统及其巡检方法
CN111983344A (zh) 一种智能电力巡检机器人系统
CN115165362A (zh) 一种基于巡检及振动信号的轴承故障智能检测系统和方法
CN117291551B (zh) 一种基于数字可视化的环境监测预警系统
CN113433900B (zh) 一种油田站场无人值守智能集控方法和系统
CN117310093A (zh) 基于物联网的大气环境监测系统
CN211876785U (zh) 一种铝电解槽漏炉智能巡检装置
CN108093210A (zh) 一种变压器油位告警系统及其告警方法
CN205721190U (zh) 一种集成多传感器的管道爬行智能检测设备
CN114791268A (zh) 基于光纤光栅的机载雷达罩的局部应变监测报警系统
CN119382326A (zh) 一种供配电设施智能巡检方法及设备
CN119418475A (zh) 电厂区域人员安全管理的多源智能分析系统及方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19907684

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19907684

Country of ref document: EP

Kind code of ref document: A1