WO2022142973A1 - Système et procédé de protection de robot - Google Patents

Système et procédé de protection de robot Download PDF

Info

Publication number
WO2022142973A1
WO2022142973A1 PCT/CN2021/134656 CN2021134656W WO2022142973A1 WO 2022142973 A1 WO2022142973 A1 WO 2022142973A1 CN 2021134656 W CN2021134656 W CN 2021134656W WO 2022142973 A1 WO2022142973 A1 WO 2022142973A1
Authority
WO
WIPO (PCT)
Prior art keywords
early warning
image
image acquisition
target object
robot
Prior art date
Application number
PCT/CN2021/134656
Other languages
English (en)
Chinese (zh)
Inventor
周伟
杨林
聂凯
朱林楠
陈凌之
Original Assignee
广东美的白色家电技术创新中心有限公司
美的集团股份有限公司
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 广东美的白色家电技术创新中心有限公司, 美的集团股份有限公司 filed Critical 广东美的白色家电技术创新中心有限公司
Publication of WO2022142973A1 publication Critical patent/WO2022142973A1/fr

Links

Images

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00Programme-controlled manipulators
    • B25J9/16Programme controls
    • B25J9/1694Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
    • B25J9/1697Vision controlled systems
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J19/00Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
    • B25J19/02Sensing devices
    • B25J19/021Optical sensing devices
    • B25J19/023Optical sensing devices including video camera means
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J19/00Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
    • B25J19/06Safety devices
    • B25J19/061Safety devices with audible signals

Definitions

  • the present invention relates to the technical field of robots, in particular to a protection system and method for robots.
  • the technical problem mainly solved by the present invention is to provide a protection system and method for a robot, which can improve the accuracy of the alarm.
  • a technical solution adopted by the present invention is to provide a protection system for a robot, the protection system for the robot includes a fence system, an image acquisition system and a control system, and the fence system is used to define an early warning area and respond to The object enters the early warning area to generate an early warning signal, and the robot works in the early warning area; the image acquisition system is used to collect images of the objects entering the early warning area in response to the early warning signal; the control system is used to analyze the images collected by the image acquisition system. To determine whether the object is a preset target object, and in response to the object to generate a control signal for the target object, to control the robot to perform corresponding protective measures.
  • the fence system is used to generate detection light transmitted along the edge of the warning area, and to generate an early warning signal when the detection light is blocked by an object.
  • the fence system includes a transmitter and a receiver, wherein the transmitter is used for generating detection light, and the receiver is used for receiving detection light, and generates an early warning signal based on the change of signal intensity generated by the detection light being blocked by an object.
  • a pre-trained deep learning model is set in the control system, and the deep learning model extracts feature points of objects in the image, and analyzes whether the object is a target object based on the extracted feature points.
  • the image acquisition system is an infrared image acquisition system, and the control system detects the temperature of the object based on the image, and analyzes whether the object is a target object based on the detected temperature.
  • the target object is a human body or an animal body.
  • the number of image acquisition systems is at least two, the at least two image acquisition systems collect objects from different angles respectively, and the control system determines that the object is a target based on images collected by any one of the at least two image acquisition systems Objects generate control signals.
  • another technical solution adopted by the present invention is to provide a protection method for a robot, the method comprising: generating an early warning signal in response to an object entering an early warning area, wherein the robot works in the early warning area; in response to the early warning signal Image acquisition of objects entering the early warning area; analysis of the images to determine whether the object is a preset target object; in response to the object being the target object, a control signal is generated to control the robot to perform corresponding protective measures.
  • the step of analyzing the image to determine whether the object is a preset target object includes: using a deep learning model to extract feature points for the object in the image, and analyzing whether the object is a target object based on the extracted feature points; or The temperature of the object is detected based on the image, and whether the object is a target object is analyzed based on the detected temperature.
  • the step of analyzing the images includes: analyzing the at least two images respectively, and generating a control signal in response to the object as the target object to form a judgment result respectively includes: The control signal is generated in response to a determination result of any one of the at least two images that the object is the target object.
  • the protection system provided by the present invention includes a fence system, an image acquisition system and a control system, and through the fence system, it can detect and judge whether an object invades the early warning area, and in the After detecting that an object invades the early warning area, the image acquisition system is triggered, the image data of the intruding object is collected, and the collected image information is further analyzed by the control system. In this way, whether the intruding object is a target object can be accurately determined, the detection accuracy is improved, and the false alarm rate is also reduced.
  • the control system can control the robot to perform corresponding protective measures in linkage to realize automatic alarm and protection.
  • FIG. 1 is a schematic structural diagram of a robot protection system in an embodiment of the present application.
  • FIG. 2 is a schematic flowchart of a robot protection method in an embodiment of the present application.
  • the application provides a robot protection system, which combines an electronic fence system and a video recognition analysis technology to realize a protection alarm system based on electronic fence triggering and image recognition confirmation.
  • This protection system does not require physical fences, and the layout is convenient and simple; the detection of light triggering can ensure accurate prediction of intrusion objects, and through the deep learning-based video image analysis technology, the types of intrusion objects can be further distinguished, so that It can accurately alarm the human body intrusion and reduce the false alarm rate caused by the intrusion of other objects.
  • FIG. 1 is a schematic structural diagram of a robot protection system in an embodiment of the present application.
  • the protection system of the robot includes a fence system 10 , an image acquisition system 20 and a control system 30 .
  • the fence system 10 is used to define an early warning area, and generates an early warning signal in response to the object entering the early warning area; the image acquisition system 20 is used to perform image acquisition on the object entering the early warning area in response to the early warning signal; the control system 30 is used for the image acquisition system. 20 The collected images are analyzed to determine whether the object is a preset target object, and in response to the object being the target object, a control signal is generated to control the robot to execute corresponding protective measures.
  • the fence system by combining the fence system, the image acquisition system and the control system, on the premise of realizing regional early warning, the image recognition of objects entering the early warning area can also be performed, and the intruding objects can be further identified, which can reduce the intrusion caused by other objects.
  • the resulting false alarm rate, and the control system can jointly control the robot to execute corresponding protection measures, realize automatic alarm and protection, and realize a highly reliable and accurate robot protection system.
  • the robot protection system can be arranged in the working area of the automated robot, and the working area of the robot is defined in the early warning area, or the robot works in the early warning area to realize the protection of the robot.
  • the robot can be a robotic arm, a loading robot, a handling robot, etc. in the field of automated machinery. This application does not limit the protection objects, and can be extended to other scenarios that require protection and early warning.
  • the fence system 10 may be an electronic fence system, which uses electronic technology to form visible/invisible protective boundaries to define warning areas.
  • the electronic fence system does not need to install guardrails like physical fences, which is more convenient to use.
  • the fencing system 10 is used to generate detection light 101 that is transmitted along the edge of the warning area, and generates a warning signal when the detection light 101 is blocked by an object.
  • the detection light 101 acts as a protective boundary and defines an early warning area.
  • the detection light can be one, two or more, and can be distributed in different positions in the warning area.
  • a plurality of detection rays can be connected to form a closed early warning area.
  • Different levels of forecast areas can be constructed to carry out different degrees of early warning. For example, multiple circles of detection rays nested inside and outside can be formed; each circle of detection rays is separated by a predetermined distance. The closer to the robot work area, the higher the alarm level.
  • the detection light can be the light visible to the human eye, which can play a more direct warning and definition role.
  • the detection line can also be invisible to the human eye, which can reduce the light pollution of the environment and reduce the negative impact of the fence system on the robot, especially for precision work robots.
  • the detection light may affect the robot.
  • the detection light is generally continuous. When an object passes through, the object blocks the detection light, which will destroy its continuity. Therefore, early warning can be realized by monitoring the continuity of the detection light.
  • the fence system 10 includes a transmitter 110 and a receiver 120, wherein the transmitter 110 is used to generate the detection light, and the receiver 120 is used to receive the detection light, and based on the detection light is blocked by the object. Generate early warning signals.
  • the transmitter 110 and the receiver 120 may be disposed at opposite ends, respectively, to form a continuous detection light between the transmitter 110 and the receiver 120 . If the transmission path of the detection light is blocked, the receiver 120 will no longer be able to receive the detection light, or the signal strength of the received detection light will be weakened. Therefore, it is determined that the detection light is blocked, and it is possible that an object passes through the boundary of the warning area and enters the warning area.
  • an early warning signal when a change in the intensity of the detection light signal is detected, an early warning signal can be generated. Different degrees of early warning signals can be generated according to the degree to which the detected light is blocked, the time to be blocked, and the area to be blocked.
  • the receivers 120 and the transmitters 110 can be arranged in pairs, one receiver 120 corresponds to one transmitter 110; they can also be arranged in groups, one transmitter 110 corresponds to multiple receivers 120, or one receiver 120 corresponds to multiple transmitters 110, which is not limited here.
  • the fencing system may be an infrared radio fencing.
  • a certain number of vertical rod-shaped infrared transmitters and receivers can be set up on the perimeter of the early warning area to form a ring fence.
  • the distance between each pair of transmitters and receivers is about 3 to 50 meters according to the different transmission power.
  • Multiple parallel invisible infrared rays are emitted to form an optical circuit with the receiver. When the intruder jumps over, the infrared circuit will be cut off, thereby generating an alarm.
  • the fencing system may be a laser beam electronic fence.
  • a certain number of vertical rod-shaped laser ray transmitters and receivers can be set up on the perimeter of the early warning area to form a ring fence.
  • Each pair of transmitters and receivers can be separated by about 10 to 150 meters according to the different transmission power.
  • the transmitter emits multiple parallel invisible laser rays, forming an optical circuit with the receiver. When the intruder climbs over, the laser beam circuit will be cut off, thereby generating an alarm.
  • the present application does not limit the way in which the detection light of the fence system is generated.
  • the electronic beam fence system has the advantages of sensitive detection, simple wiring, safety and reliability. However, because the detection is too sensitive, it is easy to be disturbed and cause false alarms. In order to solve this technical problem, in the protection system provided by the present application, an image acquisition system is added to further identify and judge the intruding objects and improve the alarm accuracy.
  • the image acquisition system 20 includes an image acquisition device 210.
  • the image acquisition device 210 can be set at the boundary of the early warning area, and the monitoring field of view of the image acquisition device 210 should cover the entire early warning area.
  • the number of image capturing devices 210 may be one, two or more, and the fields of view of the multiple image capturing devices 210 may be arranged to overlap and overlap, so as to ensure that the entire warning area can be covered.
  • the number of image capturing devices 210 is at least two, and the at least two image capturing devices 210 capture objects from different angles respectively, or in other words, each area should be covered by the field of view of at least two image capturing devices 210 .
  • an alarm signal/and/or a control signal may be generated.
  • the image capture device 210 may continuously capture video information, or may capture image information at intervals.
  • the image acquisition device 210 can perform simple processing on the image just acquired, or directly send it to the control system without processing.
  • the image acquisition device 210 may be a conventional camera for acquiring two-dimensional image information, or a depth image camera system for acquiring three-dimensional image information.
  • the image acquisition device 210 may also be an infrared image acquisition system.
  • the infrared image acquisition system can collect and obtain the temperature information of the object, detect the temperature of the object in the image, and analyze whether the object is a target object based on the detected temperature. For example, if the preset target object is a person or an animal, if the temperature is not within the body temperature range of the person or animal, it can basically rule out that the intruding object is the target object, and image recognition is no longer required. In this way, a preliminary screening is performed from the temperature, which can reduce the amount of calculation and improve the recognition speed. At the same time, the infrared camera system is also good for night vision and has wider application scenarios.
  • the image acquisition system 20 can make a prediction for the intrusion object in other ways.
  • the volume size, shape, etc. of the detected intrusion objects can be obtained; in this way, the intrusion objects can be identified and detected simply and quickly, and it is not necessary to perform image recognition on each of them.
  • the image acquisition system 20 is controlled by the fence system 10, or the image acquisition system 20 performs image acquisition on objects entering the early warning area in response to the early warning signal. That is, the image acquisition system 20 only needs to start working when an object invades, and the object intrusion is detected by the fence system 10 . In other words, only when the fence system 10 detects an intrusion of an object, will it send an instruction (early warning signal) to the image acquisition system 20 to trigger the start of the image acquisition system 20 to acquire images of the intrusion object.
  • the image acquisition device 210 can acquire image information of the early warning area in real time, and monitor the early warning area. It can only acquire image information data in real time, but does not need to identify and detect each frame of image.
  • the detection may be performed only when an early warning signal of the fence system 10 is received.
  • the image acquisition system 20 can identify and detect the collected suspicious objects according to its own judgment; for example, the image acquisition device 210 collects objects that invade the warning area, and the temperature is about 36 degrees, but the fence system 10 does not provide When an early warning signal is issued, the image acquisition system 20 judges that the object at this temperature has a high probability of being a human body. Therefore, it can perform detection by itself without the warning signal to further prevent missed detection and improve detection accuracy.
  • a wireless/wired communication connection is suggested between the image acquisition system 20 and the fence system 10 to realize the transmission of information, instructions, notifications and signals.
  • a local area network may be established in the early warning area, and the fence system 10 and the image acquisition system 20 are communicated and connected through the local area network.
  • Information transmission may be performed by means of WiFi, Bluetooth, near field communication, or the like. Information transmission may also be achieved through a wired communication connection. This application does not limit the communication method between the image acquisition system 20 and the fence system 10 .
  • the robot protection system of the present application further includes a control system 30, the control system 30 is configured to analyze the images collected by the image acquisition system 20 to determine whether the intrusion object is a preset target object, and to generate a target object in response to the intrusion object Control signals to control the robot to perform corresponding protective measures.
  • the control system 30 is mainly used to identify and analyze the images collected by the image collection system 20 .
  • the control system 30 is provided with a pre-trained deep learning model, the deep learning model extracts feature points for objects in the image, and analyzes whether the object is a target object based on the extracted feature points.
  • the deep learning model can be a deep learning key point detection model to detect human body shape; parallel statistical learning of 3D depth image information to detect human body parts; human body curve measurement based on deep learning, etc.
  • the model used can be a network model for humanoid detection only. Because the protection system does not need to specifically distinguish the identity of the person/individual, it only needs to detect the human body to determine whether it is a person; it does not require further evaluation and judgment such as face recognition and age recognition.
  • the control system 30 can also control the entire protection system, for example, it can provide instructions to the robot to control the behavior of the robot according to the detection result. If an intrusion is detected, the robot can be controlled to issue an alarm, stop motion, etc. to remind the intruder and avoid harm to the intruder.
  • the control system may be a local computer device, an independent server device, a server cluster, etc., or a cloud service system, or the like.
  • the present application provides a simple, safe and reliable robot protection system.
  • the protection system is based on the fence system, and an image acquisition system and a control system are added.
  • the image acquisition system acquires the image information of the intrusion object, and uses the control system to further discriminate the intrusion object; and when it is determined that there is a preset target object intrusion, the robot is controlled to execute corresponding protective measures.
  • the high reliability and high safety requirements of the electronic fence early warning system in the safe area of production can be met, and the behavior of the robot can be controlled by linkage after the alarm.
  • FIG. 2 is a schematic flowchart of a robot protection method in an embodiment of the present application.
  • the robot protection method includes:
  • S410 Generate an early warning signal in response to the object entering the early warning area.
  • the early warning signal is generated by the fence system and used for sending to the image acquisition system.
  • the fence system can be a through-beam electronic fence, and the early warning signal can be generated by a receiver.
  • the receiver can be set to generate an early warning signal when the detected light signal strength is lower than the threshold.
  • the early warning signal can carry the number information of the receiver, the position information of the receiver, and the information of the formation time of the early warning signal. By carrying this information, the image acquisition system can roughly know where in the early warning area there is an object intrusion, and can correspondingly activate the adjacent image acquisition equipment. Multiple early warning signals can be generated at the same time. Multiple early warning signals are parallel and do not conflict.
  • the image acquisition system will start image acquisition when any early warning signal is received.
  • connection between the fence system and the control system can also be suggested, or an early warning signal can be sent directly to the control system, so that the control system can judge and evaluate according to the early warning signal. , the warning message is issued directly based on the early warning signal.
  • the image acquisition system After receiving the early warning signal, the image acquisition system starts the image acquisition equipment to collect the image of the early warning area. It can be to acquire continuous video signals, or it can be to acquire picture information. The image acquisition system can also acquire some temperature information, sound information, depth information, etc.
  • the image acquisition system may include multiple image acquisition devices, especially when the early warning area is large, the early warning signal can be analyzed, the position information in the early warning signal can be obtained, the approximate location of the intrusion of the object can be judged, and the corresponding image acquisition equipment can be activated. In this way, the accuracy rate, acquisition speed, etc. of the image can be improved. At the same time, the amount of data is reduced, and it is not necessary to shoot the entire warning area every time.
  • a clearer enlarged image of a local area can be obtained, which is beneficial to subsequent detection. Otherwise, there may be many background factors in the collected information, and if the intrusion target is small, it is easy to miss detection or false detection.
  • multiple image acquisition devices can be used to acquire image information from multiple angles.
  • the captured images can be in color or grayscale.
  • S430 Analyze the image to determine whether the object is a preset target object.
  • the control system can be equipped with multiple trained models, which can directly input the received image information into the model to obtain the recognition structure.
  • a deep learning model can be used to extract feature points for objects in the image, and based on the extracted feature points, it is analyzed whether the object is a target object.
  • the model can be trained by using historical detection information as a sample, which is more suitable for practical applications. Because the intruders are likely to be related staff, they may appear around the warning area many times. However, since there are too few samples of intrusion objects, the trained model may not be good.
  • the trained model can be used and the model can be updated regularly.
  • There can be a variety of models on the control system which are used to detect different target objects, or to achieve detection and recognition with different accuracy.
  • the control system may receive multiple image information, some of which are similar in content.
  • the control system should analyze each image separately to obtain multiple judgment results.
  • the judgment result of any one of the at least two images is that the object is the target object, and the control signal is generated. It can compare and analyze the judgment results; record historical data to facilitate periodic statistics and guide production work specifications.
  • the control system may detect the temperature of the object based on the image, and analyze whether the object is the target object based on the detected temperature. Specifically, when the control system performs image recognition, it can first make some basic judgments to screen out some samples that are obviously different from the target object, such as the temperature is not within a reasonable range, the volume is not within a reasonable range, it is a static intrusion, and the posture of the intruding object Wait.
  • the protection system may also include an alarm device.
  • the control system can send instructions to the alarm device according to the detection result, and control the alarm device to perform corresponding alarm actions.
  • an alarm signal can be formed by means of sound, light and the like. Such as simultaneous voice warning and red light warning. Different levels of alarm signals can be formed. Such as the speed of the alarm sound, the volume of the sound, the color and brightness of the alarm light, etc.
  • the alarm device can only serve as a reminder. At this time, even if the intruder finds that he has entered the dangerous area by mistake under the prompt, if he cannot leave in time, he will be easily injured by the robot at work.
  • the control system can also be linked with the robot control system to control the robot to stop working, etc., to prevent harm to intruders.
  • the control system can only provide instructions to control the stop work, and notify the relevant staff to go to the site to check, and when it is confirmed that the intruder has left and can work safely, give control instructions to control the robot to resume work.
  • the control system can also track and detect the early warning area to determine whether the intruder has left and whether there is a new intruder. To confirm that there are no intruders in the early warning area, send a start-up work instruction to the robot control system to control the robot to restart work. In this case, automation can be achieved without human involvement.
  • control method can solve the problem of accuracy and reliability of the electronic fence alarm under high security environment requirements, and reduce the loss caused by the production process stop due to false alarm.
  • a protection system can be constructed for the working area of an automated robotic arm.
  • a rectangular early warning area can be constructed around the working area of the robotic arm.
  • the warning area should completely surround the working area of the robotic arm, and a certain safety distance should be reserved. For example, a safety distance of 0.5 meters can be reserved. Inadvertently intruding, it will not hurt.
  • each laser transmitter has a corresponding laser receiver.
  • a rectangular border made of laser light can be formed.
  • one, two or more fence boundaries can be set according to the height of the robotic arm, that is, on a pole, in the vertical direction, multiple sets of laser transmitters and receivers can be set at intervals.
  • two laser transmitters and two laser receivers may also be provided, and the laser receivers and the laser transmitters may be arranged at intervals. In this way, one laser receiver can receive laser light from two laser transmitters, and one laser transmitter can send laser light to two laser receivers.
  • a server cluster can be built to manage multiple protection systems in a centralized and unified manner, or electronic equipment can be built to independently control each protection system.
  • the control system is equipped with a detection algorithm trained by a deep learning model to improve the detection accuracy.
  • the target object can be set to be a human/animal, that is, to prevent the human or animal from entering the early warning area and prevent the mechanical arm from causing damage to it.
  • Connect the video camera to the control system to receive image information from the video camera.
  • the control system also cooperates with the control system of the robotic arm to control the robot to perform corresponding protective measures in linkage when detecting the intrusion of a target object.
  • the laser transmitter and laser receiver can be turned on to detect the warning area.
  • all 4 laser receivers receive the laser signal normally, it means that there is no object intrusion in the boundary area of the fence. If any of the 4 laser receivers are interrupted or not received the laser signal. At this time, a trigger signal is generated, indicating that an object has invaded the warning boundary.
  • the trigger signal is sent to the video camera, and the two video cameras are activated to continuously collect the video signal and transmit it to the control system.
  • Two video cameras can detect the same area/object at the same time to improve redundancy and reduce the missed detection rate.
  • the control system can detect the received image information, and use the detection algorithm of human key points based on deep learning to detect whether the intruding object is a human being.
  • the detection algorithm of human key points of deep learning By using the detection algorithm of human key points of deep learning, the accuracy of human detection can be greatly improved.
  • the alarm signal can be formed by means of sound and light. Such as simultaneous voice warning and red light warning.
  • the alarm information is sent to the robotic arm control system to control the robotic arm to automatically stop working to prevent damage to the human body.
  • an electronic protection system based on laser triggering and image recognition and confirmation is realized by combining the laser beam shooting system and the video recognition analysis technology.
  • the use of this protection system does not require physical fences, and the layout is convenient and simple; the use of laser triggering can ensure accurate prediction of intruding objects, and through the deep learning-based video image analysis technology, the types of intruding objects can be further distinguished, which can Accurately alarms human intrusion and reduces the false alarm rate caused by intrusion of other objects.
  • the fence system and the image acquisition system both carry processors, which are used for early warning signal processing, equipment management, and the like.
  • the processor can also be called CPU (Central Processing Unit, central processing unit).
  • a processor may be an integrated circuit chip with signal processing capabilities.
  • the processor may also be a general purpose processor, digital signal processor (DSP), application specific integrated circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA field programmable gate array
  • a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the disclosed system, apparatus and method may be implemented in other manners.
  • the apparatus embodiments described above are only illustrative.
  • the division of units is only a logical function division.
  • there may be other division methods for example, multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented.
  • the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

Landscapes

  • Engineering & Computer Science (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Multimedia (AREA)
  • Alarm Systems (AREA)
  • Burglar Alarm Systems (AREA)

Abstract

La présente invention concerne un système et un procédé de protection de robot. Le système de protection de robot comprend un système de clôture, un système d'acquisition d'images et un système de commande ; le système de clôture est configuré pour définir une zone d'avertissement précoce et générer un signal d'avertissement précoce en réponse à l'entrée d'un objet dans la zone d'avertissement précoce, un robot travaillant dans la zone d'avertissement précoce ; le système d'acquisition d'images est configuré pour réaliser une acquisition d'image sur l'objet entrant dans la zone d'avertissement précoce en réponse au signal d'avertissement précoce ; et le système de commande est configuré pour analyser une image acquise par le système d'acquisition d'image pour déterminer si oui ou non l'objet est un objet cible prédéfini, et générer un signal de commande en réponse à l'objet qui est l'objet cible, pour commander le robot pour l'exécution de mesures de protection correspondantes. De cette manière, la présente invention peut améliorer la précision d'alarme.
PCT/CN2021/134656 2020-12-31 2021-11-30 Système et procédé de protection de robot WO2022142973A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011627666.1A CN112757300A (zh) 2020-12-31 2020-12-31 机器人的防护系统及方法
CN202011627666.1 2020-12-31

Publications (1)

Publication Number Publication Date
WO2022142973A1 true WO2022142973A1 (fr) 2022-07-07

Family

ID=75699310

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/134656 WO2022142973A1 (fr) 2020-12-31 2021-11-30 Système et procédé de protection de robot

Country Status (2)

Country Link
CN (1) CN112757300A (fr)
WO (1) WO2022142973A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116778653A (zh) * 2023-06-25 2023-09-19 国网湖北省电力有限公司宜昌供电公司 用于应急电源抢修车供电时的标识及检测警示方法

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112757300A (zh) * 2020-12-31 2021-05-07 广东美的白色家电技术创新中心有限公司 机器人的防护系统及方法
CN113821040A (zh) * 2021-09-28 2021-12-21 中通服创立信息科技有限责任公司 一种深度视觉相机与激光雷达融合导航的机器人
CN113733163B (zh) * 2021-11-05 2022-02-08 杭州骏沃机电科技有限公司 一种机械手用适配型安全装置
CN114407025B (zh) * 2022-03-29 2022-06-28 北京云迹科技股份有限公司 一种机器人急停模式自动控制方法、装置及机器人

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100208067A1 (en) * 2009-02-13 2010-08-19 Dietmar Doettling Method and apparatus for determining an overtravel time of a machine
CN109311164A (zh) * 2016-06-20 2019-02-05 库卡德国有限公司 对机器人组的监视
CN110001597A (zh) * 2018-11-08 2019-07-12 蔚来汽车有限公司 安全防护系统、方法及包括该安全防护系统的换电站
CN111178257A (zh) * 2019-12-28 2020-05-19 深圳奥比中光科技有限公司 一种基于深度相机的区域安全防护系统及方法
CN111489521A (zh) * 2020-05-22 2020-08-04 成都电科慧安科技有限公司 一种基于温度检测的网络电子围栏系统及其实施方法
CN112757300A (zh) * 2020-12-31 2021-05-07 广东美的白色家电技术创新中心有限公司 机器人的防护系统及方法

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105729468B (zh) * 2016-01-27 2018-01-09 浙江大学 一种基于多深度摄像机增强的机器人工作台
CN106845450A (zh) * 2017-02-22 2017-06-13 武汉科技大学 基于近红外成像与深度学习的黑暗环境人脸识别方法
CN107820051A (zh) * 2017-11-07 2018-03-20 北京京电电网维护集团有限公司 监控系统及其监控方法和装置
US11106927B2 (en) * 2017-12-27 2021-08-31 Direct Current Capital LLC Method for monitoring an interior state of an autonomous vehicle
CN108647562A (zh) * 2018-03-27 2018-10-12 北京中硕众联智能电子科技有限公司 识别图像中物体和温度的装置和相应的方法
CN110666814A (zh) * 2019-10-30 2020-01-10 广州供电局有限公司 配电网巡检机器人
CN111144232A (zh) * 2019-12-09 2020-05-12 国网智能科技股份有限公司 基于智能视频监控的变电站电子围栏监测方法、存储介质及设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100208067A1 (en) * 2009-02-13 2010-08-19 Dietmar Doettling Method and apparatus for determining an overtravel time of a machine
CN109311164A (zh) * 2016-06-20 2019-02-05 库卡德国有限公司 对机器人组的监视
CN110001597A (zh) * 2018-11-08 2019-07-12 蔚来汽车有限公司 安全防护系统、方法及包括该安全防护系统的换电站
CN111178257A (zh) * 2019-12-28 2020-05-19 深圳奥比中光科技有限公司 一种基于深度相机的区域安全防护系统及方法
CN111489521A (zh) * 2020-05-22 2020-08-04 成都电科慧安科技有限公司 一种基于温度检测的网络电子围栏系统及其实施方法
CN112757300A (zh) * 2020-12-31 2021-05-07 广东美的白色家电技术创新中心有限公司 机器人的防护系统及方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116778653A (zh) * 2023-06-25 2023-09-19 国网湖北省电力有限公司宜昌供电公司 用于应急电源抢修车供电时的标识及检测警示方法
CN116778653B (zh) * 2023-06-25 2024-04-19 国网湖北省电力有限公司宜昌供电公司 用于应急电源抢修车供电时的标识及检测警示方法

Also Published As

Publication number Publication date
CN112757300A (zh) 2021-05-07

Similar Documents

Publication Publication Date Title
WO2022142973A1 (fr) Système et procédé de protection de robot
CN111770266B (zh) 一种智能视觉感知系统
US8908034B2 (en) Surveillance systems and methods to monitor, recognize, track objects and unusual activities in real time within user defined boundaries in an area
KR101644443B1 (ko) 시공적 상황데이터를 이용한 경고 방법 및 시스템
CN107645652A (zh) 一种基于视频监控的非法越界报警系统
JP2021500668A (ja) 監視装置、産業用設備、監視方法及びコンピュータプログラム
CN209993060U (zh) 一种基于物联网技术的电子围栏联动系统
KR101575011B1 (ko) 영상을 이용한 적외선감지기와 그 동작방법 및 이를 이용한 보안시설물 통합관리시스템
US11386669B2 (en) Building evacuation method and building evacuation system
US9613510B2 (en) Apparatus and method for rapid human detection with pet immunity
KR20080044812A (ko) 컴퓨터 영상 분석기술을 이용한 범죄 및 사고예방 자동경비 시스템
KR102499410B1 (ko) 무인 경비를 위한 스마트 알림 시스템 및 그 방법
JP6195447B2 (ja) 撮影システム
CN213876966U (zh) 一种铁路周界安全防范系统
KR101415848B1 (ko) 차량 및 보행자 검지를 이용한 스쿨존 방범장치
JP2013008298A (ja) 警備システム
JP2006323709A (ja) 防犯監視装置
KR102513372B1 (ko) 음향과 영상을 연계한 하이브리드형 객체 연속 추적시스템 및 제어방법
JP5876253B2 (ja) 不審物体監視システム
KR102538548B1 (ko) 딥러닝 기반의 객체 인식 및 모니터링 시스템
JP2012103901A (ja) 侵入物体検出装置
JP2018192844A (ja) 監視装置、監視システム、監視プログラム、および、記憶媒体
JP7384601B2 (ja) 監視システム
KR20170133592A (ko) 승강장 추락방지 시스템
CN113810656A (zh) 一种可自动调节的视频监控行为识别装置

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: 21913714

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 16/11/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21913714

Country of ref document: EP

Kind code of ref document: A1