Safe stopping device for AI vision humanoid detection control robot

Abstract

The utility model discloses an AI visual humanoid detection control robot safety stop device, which comprises an outer shell of the device, a control circuit board arranged in the outer shell, a switch connected with the control circuit board and used for collecting work video data of a robot, an output end of the switch is connected with an input end of an AI humanoid detection unit, and an output end of the AI humanoid detection unit is connected with an output end of the AI humanoid detection unit. An alarm output module is arranged at the AI human shape detection unit, the output end of the alarm output module is connected with the input end of a signal adapter, and the output end of the signal adapter is connected with multiple groups of robots through a safety stop interface. According to the utility model, the AI human shape detection unit processes image data in real time and rapidly triggers an alarm signal, so that a response can be rapidly made when a human shape target enters a dangerous area, the robot is timely controlled to stop safely, and the personnel safety is effectively guaranteed.

Description

Technical Field

[0001] This utility model relates to a safety stopping device for AI vision humanoid detection and control robots. Background Technology

[0002] In the field of industrial automation, the widespread application of robots has significantly improved production efficiency, but the potential safety risks during their operation remain a focus of industry attention. Traditional robot safety control solutions mainly rely on passive protection methods such as physical guardrails, photoelectric sensors, or emergency stop buttons. While these methods can limit personnel from entering hazardous areas to some extent, they have significant drawbacks: physical protective facilities lack flexibility and are difficult to adapt to dynamic and changing working environments; fixed sensors have limited detection ranges and their response speed is constrained by hardware characteristics, failing to meet the real-time requirements of complex scenarios. Furthermore, traditional solutions lack the ability to intelligently recognize human behavior, making them prone to misjudgments or missed detections in sudden human-robot interaction scenarios, leading to safety hazards.

[0003] In recent years, the development of artificial intelligence vision technology has provided new ideas for environmental perception. AI-based vision systems can detect human targets in real time through image recognition technology. However, existing AI vision systems mostly focus on data processing and algorithm optimization, without deep integration with the safety control interface of industrial robots. Specifically, traditional vision controllers typically only output alarm signals and cannot directly trigger the robot's safety stop mechanism, requiring intermediate links (such as manual intervention or third-party control systems) to achieve a response. This indirect control mode is prone to introducing communication delays, causing the robot to fail to stop in time in emergency situations. In addition, existing solutions lack redundancy design, and a single point of failure may cause system failure, further reducing safety and reliability.

[0004] Therefore, an AI-based visual humanoid detection and control robot safety stopping device is proposed. Utility Model Content

[0005] The purpose of this invention is to provide an AI vision-based humanoid detection and control robot safety stopping device to solve the problems mentioned in the background art, where traditional robot safety control relies heavily on physical guardrails or fixed sensors, resulting in poor flexibility, slow response speed, inability to adapt to dynamic environments, and control delays or failures in dangerous scenarios.

[0006] To solve the above-mentioned technical problems, this utility model provides the following technical solution: an AI visual human detection and control robot safety stop device, including a housing, a control circuit board located inside the housing, a switch for collecting robot working video data connected to the control circuit board, an input terminal of an AI human detection unit connected to the output terminal of the switch, an alarm output module provided at the AI ​​human detection unit, an input terminal of a signal adapter connected to the output terminal of the alarm output module, and an output terminal of the signal adapter connected to multiple groups of robots through multiple sets of safety stop interfaces, with each set of safety stop interfaces corresponding to a set of robots;

[0007] The switch outputs the collected robot working video data to the AI ​​human detection unit for detection and identification. After detecting a human target, it outputs an alarm signal through the alarm output module, and the signal converter sends a safety stop signal to the safety stop interface, and the robot stops working.

[0008] Preferably, the switch is provided with multiple sets of data input interfaces, and the data input interfaces are connected to the output end of the network camera, so that the network camera transmits real-time image data to the switch through the data input interfaces.

[0009] Preferably, the AI ​​human detection unit is also provided with HDMI, USB and SATA interfaces, so that the AI ​​human detection unit can be connected to the display screen, mouse and hard drive respectively, and used to connect to the display screen, connect to the mouse and store video data respectively.

[0010] Preferably, the output terminal of the alarm output module is connected to the robot via a K1 relay. The K1 relay is used to transfer alarm signals and is equipped with a safety stop interface.

[0011] Preferably, the K1 relay is also connected to a DIP switch for controlling the effectiveness of the alarm output of the AI ​​humanoid detection unit, which is used to disable the robot's safety stop function.

[0012] Preferably, the safety stop interface includes a first output terminal and a second output terminal, wherein the first output terminal includes a safety stop interface 1+ and a safety stop interface 1-; and the second output terminal includes a safety stop interface 2+ and a safety stop interface 2-.

[0013] When both the first and second output terminals are in a conducting state, the robot recognizes the alarm signal and triggers a safe stop; otherwise, the robot does not trigger a safe stop.

[0014] Preferably, the signal adapter is also connected to the input terminal of the switch via a microcontroller unit.

[0015] The microcontroller unit includes an STM32F407VET6 chip and a LAN8720 module. The STM32F407VET6 chip and the LAN8720 module are connected via an RMII interface and are used to convert alarm signals into Modbus TCP protocol data packets for network transmission.

[0016] Preferably, the switch is provided with a communication output port for networking and sending alarm data from the microcontroller unit. The communication output port is connected to the PLC controller and is used to remotely turn the robot's safety stop signal on / off.

[0017] Preferably, the output terminal of the K1 relay is also connected to the GPIO pin of the STM32F407VET6 microcontroller chip.

[0018] Preferably, the housing also contains a power supply unit, which is electrically connected to the control circuit board, switch, AI human detection unit, DIP switch and signal adapter.

[0019] Compared with the prior art, the beneficial effects of this utility model are:

[0020] This invention uses an AI human detection unit to process image data in real time and quickly trigger alarm signals, ensuring a rapid response when a human target enters a dangerous area, timely control of the robot to stop safely, and effectively protecting personnel safety.

[0021] The robot safety stop interface adopts a dual-channel redundant design to avoid system failure due to single point of failure, improve system reliability, ensure reliable robot stop in emergency situations, and reduce safety risks.

[0022] The effectiveness of alarm signals can be controlled by DIP switches, allowing for flexible selection of whether to disable the robot's safety stop function in special circumstances (such as equipment maintenance and debugging), thus avoiding accidental triggering that could affect normal operations.

[0023] The microcontroller unit converts alarm signals into Modbus TCP protocol data packets for network transmission, facilitating integration with devices that support this protocol (such as PLCs), enabling information sharing and collaborative work between systems, and improving the overall integrity of the automation system.

[0024] The PLC uses the Modbus protocol to verify signal integrity, effectively filtering out erroneous or incomplete signals. This ensures that only accurate alarm signals trigger the robot to stop safely, avoiding misoperation due to signal interference or transmission errors and improving system stability. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the overall structure of an embodiment of the present utility model;

[0026] Figure 2 This is a circuit diagram showing the connection between the alarm output module and the signal converter in an embodiment of the present invention.

[0027] Figure 3 This is a system flowchart of an embodiment of the present utility model. Detailed Implementation

[0028] To address the shortcomings of traditional robot safety control, which relies heavily on physical barriers or fixed sensors, resulting in poor flexibility, slow response, inability to adapt to dynamic environments, and control delays or failures in hazardous scenarios, this invention provides an AI-powered visual human detection and control robot safety stopping device. The technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0029] Please see Figure 1-3 This utility model provides an AI visual human detection and control robot safety stop device, including a housing. Inside the housing is a control circuit board connected to the control circuit board for collecting robot working video data. The output of the switch is connected to the input of an AI human detection unit. The AI ​​human detection unit has an alarm output module, the output of which is connected to the input of a signal converter. The output of the alarm output module is connected to the robot via a K1 relay, which processes alarm signals. The K1 relay has a safety stop interface. The safety stop interface includes a first output and a second output. The first output includes safety stop interface 1+ and safety stop interface 1-; the second output includes safety stop interface 2+ and safety stop interface 2-. When both the first and second outputs are in a conducting state, the robot recognizes the alarm signal and triggers a safety stop; otherwise, the robot does not trigger a safety stop.

[0030] The K1 relay is also connected to a DIP switch for controlling the effectiveness of the alarm output of the AI ​​humanoid detection unit, and for disabling the robot's safety stop function.

[0031] The output terminal of the K1 relay is also connected to the GPIO pin of the STM32F407VET6 microcontroller chip.

[0032] The output of the signal converter is connected to multiple groups of robots through multiple sets of safety stop interfaces, and each set of safety stop interfaces is configured in correspondence with a group of robots.

[0033] The signal converter is also connected to the input of the switch via a microcontroller unit. The microcontroller unit includes an STM32F407VET6 chip and a LAN8720 module. The STM32F407VET6 chip and the LAN8720 module are connected via an RMII interface to convert alarm signals into Modbus TCP protocol data packets for network transmission. The switch is equipped with a communication output port for networking and sending alarm data from the microcontroller unit. This communication output port is connected to the PLC controller for remotely enabling / disabling the robot's safety stop signal.

[0034] The switch outputs the collected robot working video data to the AI ​​human detection unit for detection and identification. After detecting a human target, it outputs an alarm signal through the alarm output module, and the signal converter sends a safety stop signal to the safety stop interface, and the robot stops working.

[0035] The switch is equipped with multiple sets of data input interfaces, and the data input interfaces are connected to the output end of the network camera, so that the network camera can transmit real-time image data to the switch through the data input interfaces.

[0036] The AI ​​human detection unit is also equipped with HDMI, USB and SATA interfaces, so that the AI ​​human detection unit can be connected to the display screen, mouse and hard drive respectively, for connecting to the display screen, connecting to the mouse and storing video data.

[0037] The housing also houses a power supply unit, which is electrically connected to the control circuit board, switch, AI human detection unit, DIP switch, and signal adapter.

[0038] The working principle of the AI ​​vision-based humanoid detection and control robot safety stop device provided by this utility model is as follows:

[0039] In industrial automation environments, when a humanoid target (such as a worker) accidentally enters the dangerous area where the robot is working (human-defined), the device needs to immediately trigger an alarm and control the robot to stop safely.

[0040] Device operation process (taking robot No. 1 as an example):

[0041] The network camera connects to the switch via a data input interface to capture image data of the work area in real time. The switch unit then transmits the image data to the AI ​​human detection unit.

[0042] After receiving image data, the AI ​​human detection unit analyzes the image in real time to determine if a human-shaped target is present. When a human-shaped target is detected entering a dangerous area, the AI ​​human detection unit sends an alarm output signal through the alarm output module (AI human detection unit detection of human-shaped targets is existing technology, so it will not be described in detail).

[0043] The alarm output signals (such as 1Q and 1G) of the AI ​​humanoid detection unit are input through the input port of the signal adapter. The signal adapter processes the alarm signals. When the alarm signal is valid, the main coil of the K1 relay is in the conducting state, the normally open contact (NO) of the K1 relay is closed, and the normally closed contact (NC) is open. The two normally open contacts of the K1 relay form two circuits of the No. 1 robot safety interface (the first output terminal and the second output terminal), which are simultaneously connected to the microcontroller unit.

[0044] After the K1 relay contacts switch, the signal adapter transmits the alarm signal to the robot via the safety stop interface. The alarm signal is transmitted to the robot's safety input interface through a dual-loop design, ensuring signal reliability and preventing single-point failure.

[0045] The signal converter simultaneously transmits the alarm signal to the input interface of the STM32 F407VET controller via the K1 relay. The STM32F407VET6 controller confirms the integrity of the signal via the Modbus protocol and further executes the safety stop logic (sending the robot's safety stop signal to the PLC controller via the Modbus TCP protocol).

[0046] The DIP switch is connected to the main coil circuit of relay K1, which constitutes the function of shielding whether robot No. 1 has stopped safely.

[0047] The remote start signal of the PLC controller is connected to the main circuit of the K9 relay. When this signal is 24V, the main coil of the K9 relay is in the conducting state, the auxiliary contacts of the K9 relay switch, and the digital alarm signal is activated. When this signal is not activated, the digital alarm signal is in the off state.

[0048] Upon receiving the alarm signal, the robot immediately executes a safety stop to ensure the safety of the humanoid target. Simultaneously, the PLC controller records the alarm event and triggers other safety measures (such as audible and visual alarms).

[0049] Once the humanoid target leaves the danger zone, the AI ​​humanoid detection unit stops outputting alarm signals after a 5-second delay (this is existing technology and will not be elaborated further). After the signal converter detects that the alarm signal has disappeared, the relay contacts return to their initial state (normally open contacts open, normally closed contacts close), and the robot resumes normal operation.

[0050] Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An AI vision human shape detection control robot safety stop device, comprising a device outer shell, characterized in that: The inside of the shell is provided with a control circuit board, a switch connected to the control circuit board for collecting robot working video data, an input end of an AI human detection unit connected to the output end of the switch, an alarm output module arranged at the AI human detection unit, an input end of a signal adapter connected to the output end of the alarm output module, and a plurality of groups of safety stop interfaces connected to the output end of the signal adapter and a plurality of groups of robots, and one group of safety stop interfaces and one group of robots are arranged in correspondence. The switch outputs the collected robot working video data to the AI human detection unit for detection and identification, outputs an alarm signal through the alarm output module after detecting a human target, and sends a safety stop signal to the safety stop interface through the signal adapter, so that the robot stops working.

2. The AI vision human detection control robot safety stop device according to claim 1, wherein: A plurality of groups of data input interfaces are arranged at the switch, and the data input interfaces are connected to the output end of the network camera, so that the network camera transmits real-time image data to the switch through the data input interface.

3. The AI vision human detection control robot safety stop device according to claim 2, wherein: The AI human detection unit is also provided with HDMI, USB and SATA interfaces, so that the AI human detection unit is connected to a display screen, a mouse and a hard disk respectively, and is used for connecting the display screen, connecting the mouse and storing video data respectively.

4. The AI vision human detection control robot safety stop device according to claim 1, wherein: The output end of the alarm output module is connected to the robot through a K1 relay, the K1 relay is used for switching processing of the alarm signal, and the K1 relay is provided with a safety stop interface.

5. The AI vision human detection control robot safety stop device according to claim 4, wherein: The K1 relay is also connected with a dial switch for controlling the alarm output effectiveness of the AI human detection unit, and is used for shielding the safety stop function of the robot.

6. The AI vision human detection control robot safety stop device according to claim 4, wherein: The safety stop interface includes a first output end and a second output end, the first output end includes a safety stop interface 1+ and a safety stop interface 1-, and the second output end includes a safety stop interface 2+ and a safety stop interface 2-. When the first output end and the second output end are both in a conduction state, the robot recognizes the alarm signal and triggers the safety stop of the robot; otherwise, the safety stop of the robot is not triggered.

7. The AI vision human detection control robot safety stop device according to claim 1, wherein: The signal adapter is also connected to the input end of the switch through a micro control unit, The micro control unit includes an STM32F407VET6 chip and a LAN8720 module, the STM32F407VET6 chip and the LAN8720 module are connected through an RMII interface, and are used for converting the alarm signal into a ModbusTCP protocol data packet for network transmission.

8. The AI vision human detection control robot safety stop device according to claim 7, wherein: The switch is provided with a communication output port for networking and sending alarm data at the micro control unit, the communication output port is connected to a PLC controller, and is used for remotely opening / closing the safety stop signal of the robot.

9. The AI vision human detection control robot safety stop device according to claim 4, characterized in that: The output end of the K1 relay is also connected to the GPIO pin of the STM32F407VET6 chip of the micro control unit.

10. The AI vision human detection control robot safety stop device according to claim 9, wherein: The inside of the shell is also provided with a power supply unit, and the power supply unit is electrically connected to the control circuit board, the switch, the AI human detection unit, the dial switch and the signal adapter.

Images