Anti-interference self-adaptive robot for severe working conditions and control method

Abstract

The invention discloses an anti-interference self-adaptive robot for severe working conditions and a control method, and relates to the technical field of machine visual identification, and the anti-interference self-adaptive robot comprises a robot walking mechanism, an anti-interference visual detection unit, an autonomous handling mechanical arm unit and a communication unit. The anti-interference visual detection unit adopts the double-layer shell assembly under the severe environment of high dust, high humidity and high temperature, so that the protection performance of the visual detection unit is improved, and the visual detection unit can stably work under the severe working condition. The multiple visual camera modules are distributed on the periphery and the top of the double-layer shell assembly, visual pictures can be shot in all directions, and the coverage range and accuracy of visual detection are improved. Through research on composition, composition, components, characteristics and production process mechanisms of underground sundries, the technical problem of accurately distinguishing high-grade ores and iron sundries is solved based on machine vision identification and metal detection application technologies. And intelligent mine safety production of unmanned intelligent impurity removal is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of machine vision recognition, and in particular to an anti-interference adaptive robot and a control method for harsh working conditions. Background Art

[0002] In the field of smart mining and industrial intelligent manufacturing, efficient sorting of underground debris and reliable operation of robots in harsh environments are the core technical problems that restrict safe production and intelligent upgrading. The current technology has the following bottlenecks:

[0003] 1. Difficulty in accurately identifying ore and iron debris:

[0004] Traditional underground sorting technology relies on manual screening or a single sensor (such as electromagnetic induction), which makes it difficult to distinguish high-grade ore from iron debris (such as waste bolts and worn metal fragments) under complex working conditions. The two have similar metal properties, resulting in a high misjudgment rate, which causes waste of resources or equipment damage. Existing machine vision technology is easily disturbed in high-dust and uneven lighting environments, and cannot achieve collaborative analysis of multi-dimensional features (such as texture, composition, and electromagnetic properties). It is urgent to integrate multimodal data and intelligent algorithms to improve recognition accuracy.

[0005] 2. Robots are not adaptable enough to harsh environments:

[0006] The dusty, humid, and high-temperature underground environment causes multiple damages to industrial robots: dust intrusion causes mechanical structures to get stuck and circuits to short-circuit; humidity and high temperature accelerate metal corrosion and affect visual sensor imaging; traditional robot structural design does not systematically optimize "environment-material-function", relies on frequent manual maintenance, and is difficult to achieve autonomous operation. Existing solutions mostly use local protection (such as sealed housings), but sacrifice robot flexibility and cannot solve deep problems such as thermal stress deformation and material aging, and lack closed-loop design criteria from mechanism to engineering application.

[0007] 3. Lack of technical synergy:

[0008] Most of the existing sorting systems and operating robots are independent modules, and the full-link closed loop of perception-decision-execution has not been realized. For example, the debris recognition results fail to dynamically guide the robot's obstacle avoidance path planning, and the robot's mechanical arm lacks a flexible adaptation mechanism for grabbing special-shaped debris, resulting in low sorting efficiency. In addition, the data fusion depth of traditional metal detectors and visual systems is insufficient, making it difficult to build a high-precision three-dimensional feature library of debris, which restricts the reliability of intelligent debris removal. Summary of the invention

[0009] In order to solve the above technical problems, the present invention provides an anti-interference adaptive robot and a control method for harsh working conditions. The following technical solutions are adopted:

[0010] An anti-interference adaptive robot for harsh working conditions comprises a robot walking mechanism, an anti-interference visual detection unit, an autonomous handling robot arm unit and a communication unit, wherein the anti-interference visual detection unit comprises a double-layer shell component, a plurality of visual camera modules and a visual analysis module, wherein the bottom of the double-layer shell component is detachably mounted on the top of the robot walking mechanism, the four sides and the top of the double-layer shell component are provided with a plurality of openings for installing camera modules, the plurality of visual camera modules are respectively mounted at the plurality of openings of the double-layer shell component for shooting visual images around and on the top, the visual analysis module is respectively connected to the plurality of visual camera modules for communication, the visual analysis module uses a target-based visual detection algorithm to identify and locate anomalies in the visual images shot by the plurality of visual camera modules, and the autonomous handling robot arm The unit is installed on the top of the double-layer shell assembly and the position of the visual camera module is staggered. The autonomous disposal robot arm unit includes an autonomous robot arm, an active visual camera module, a metal detector and a chip-based robot arm controller. The active visual camera module is installed on one side of the active end of the autonomous robot arm to shoot the visual picture pointed by the active end of the autonomous robot arm. The metal detector is installed on the other side of the active end of the autonomous robot arm. The robot arm controller, the active visual camera module with a spotlight and the metal detector are respectively communicated with the visual analysis module. The robot arm controller receives interactive abnormal positioning data and controls the autonomous robot arm to point to the foreign object based on the abnormal positioning data. The communication unit is installed around the top of the double-layer shell assembly, and the visual analysis module is wirelessly connected to the remote server through the communication unit.

[0011] By adopting the above technical solution, the robot walking mechanism moves to the designated area, such as the underground tunnel. The anti-interference visual detection unit starts working, and multiple visual camera modules capture the visual images around and on the top. The visual analysis module processes the visual images captured by the camera module and uses the target-based visual detection algorithm to identify and locate the anomaly.

[0012] If the visual analysis module identifies an abnormality, the abnormal location data is sent to the robot arm controller of the autonomous disposal robot arm unit. The robot arm controller receives the abnormal location data and controls the autonomous robot arm to point to the foreign object. The active visual camera module captures the visual image pointed by the active end of the autonomous robot arm, and the metal detector detects the foreign object. The robot arm controller, active visual camera module and metal detector send data to the visual analysis module.

[0013] The visual analysis module interacts with the remote server wirelessly through the communication unit to interact with the abnormal target positive vision data and metal detection data of the visual camera module;

[0014] The remote server analyzes and processes the data, determines the nature of the foreign matter, and generates a corresponding disposal strategy. If it is determined that an abnormality has occurred, an alarm is issued, and the staff manually controls the robot's walking mechanism and the autonomous disposal robot arm unit through the remote control device to approach and further collect data, or they can manually go to the site for disposal;

[0015] In view of the harsh environment of high dust, high humidity and high temperature, the anti-interference visual detection unit adopts a double-layer shell component, which improves the protection performance of the visual detection unit and can work stably under harsh working conditions. Multiple visual camera modules are distributed around and on the top of the double-layer shell component, which can capture visual images in all directions, improving the coverage and accuracy of visual detection.

[0016] The visual analysis module adopts a target-based visual detection algorithm, which can effectively identify and locate anomalies, improving the accuracy and real-time performance of anomaly detection.

[0017] The autonomous disposal robot arm unit integrates an autonomous robot arm, an active visual camera module, a metal detector and a robot arm controller. It can automatically point to the location of foreign objects based on abnormal positioning data and perform corresponding disposal operations, thereby improving the robot's autonomous disposal capabilities.

[0018] Through the study of the composition, structure, ingredients, characteristics and generation process of underground debris, we have carried out research on application technology and key technologies based on machine vision recognition and metal detection, and have broken through the technical difficulties of accurately distinguishing "high-grade ore and iron debris". We have achieved safe production in smart mines with unmanned and intelligent debris removal.

[0019] Optionally, a double-layer shell assembly includes an outer shell, an inner shell and an anti-interference buffer layer, the outer shell is provided with multiple openings for installing a camera module around and on the top, the bottom of the outer shell is installed on the top of the robot walking mechanism, the outer wall of the inner shell is installed inside the outer shell through a rib plate, and multiple openings are also provided around the inner shell at positions corresponding to the multiple openings of the outer shell, the anti-interference buffer layer is installed on the inner wall of the inner shell, an electrical device installation space is formed inside the anti-interference buffer layer, and the visual analysis module is installed in the electrical device installation space.

[0020] By adopting the above technical solutions, multi-layer protection is provided, which enhances the mechanical strength and anti-interference ability of the entire visual inspection unit. As the outermost layer of protection, the shell can resist external impact and the influence of harsh environment.

[0021] The multiple openings in the housing facilitate the installation of multiple vision camera modules while maintaining all-round coverage of visual inspection, improving the efficiency and accuracy of visual inspection.

[0022] The inner shell is installed inside the outer shell through the rib plate, which increases the stability and supporting force of the structure. At the same time, the opening of the inner shell corresponds to the opening of the outer shell, ensuring the installation position and field of view of the camera module.

[0023] The anti-interference buffer layer is installed on the inner wall of the inner shell to form a space for installing electrical components. The anti-interference buffer layer can be conductive rubber, metal mesh or conductive paint. These materials can block external electromagnetic interference and protect internal electronic components from electromagnetic interference. It provides additional anti-interference protection for electronic components such as visual analysis modules. The buffer layer can absorb external vibration and impact, reduce the impact on internal electronic components, and improve the reliability and stability of the entire system.

[0024] The electrical component installation space provides an independent and safe working environment for the visual analysis module, which helps to improve its working performance and life. At the same time, this design also helps with heat dissipation and electromagnetic shielding, further improving the stability and anti-interference ability of the system.

[0025] The overall structure is compact and the modular design is easy to maintain and upgrade. The double-layer shell assembly can be easily installed on the top of the robot's walking mechanism, and it is also easy to disassemble and replace, which improves the robot's maintenance efficiency and flexibility.

[0026] This double-layer shell component design not only improves the anti-interference ability and mechanical strength of the visual inspection unit, but also provides additional protection for its internal electronic components, enhancing the stability and reliability of the entire system. At the same time, the modular design also facilitates maintenance and upgrades, improving the overall performance and adaptability of the robot.

[0027] Optionally, the visual camera module includes a visual camera and a ring light, wherein the base of the visual camera is installed at the opening of the inner shell, the visual camera is communicatively connected to the visual analysis module, and the ring light is installed on the outer wall of the opening of the inner shell and surrounds the visual camera.

[0028] Optionally, the visual camera module further includes a transparent ball cover, which is installed at the opening of the housing.

[0029] By adopting the above technical solution, the transparent ball cover is made of tempered glass or acrylic.

[0030] Optionally, the autonomous robotic arm is a miniature multi-axis robotic arm, and the active visual camera module includes a camera housing and an active camera, the camera housing is mounted on the side wall of the active end of the autonomous robotic arm, the active camera is mounted at the end of the camera housing, and the lens direction of the active camera is consistent with the direction of the active end of the autonomous robotic arm, and the active camera is communicatively connected to the visual analysis module.

[0031] Optionally, the visual analysis module includes a multi-channel video acquisition card, a memory, a visual analysis chip and a data analysis chip. The multi-channel video acquisition card is communicatively connected to multiple visual cameras and active cameras respectively, the memory is communicatively connected to the multi-channel video acquisition card, the visual analysis chip and the data analysis chip are communicatively connected to the memory respectively, and the data analysis chip exchanges the position and posture information data of the robot walking mechanism with the navigation module of the robot walking mechanism.

[0032] Optionally, the robotic arm controller includes an instruction cache and a control chip, the instruction cache stores instruction set data for controlling the robotic arm controller, the active visual camera module and the metal detector, the control chip is communicatively connected to the instruction cache, and the control chip controls the execution actions of the robotic arm controller, the active visual camera module and the metal detector respectively based on the instruction set data.

[0033] Optionally, the communication unit includes four wireless communication antennas and a baseband processing unit. The four wireless communication antennas are respectively installed around the top of the robot's walking mechanism. The four wireless communication antennas are respectively connected to the visual analysis module through the baseband processing unit, and the visual analysis module is wirelessly connected to the remote server through the communication unit.

[0034] By adopting the above technical solution, four wireless communication antennas are installed around the top of the robot's walking mechanism. This layout ensures that when the robot moves or turns, at least one antenna can maintain a good communication connection with the remote server. This all-round signal coverage improves the stability and reliability of communication and reduces communication interruptions caused by signal blind spots.

[0035] Multi-antenna systems (such as MIMO, multiple-input multiple-output technology) can significantly improve the capacity and efficiency of wireless communications. Four antennas can send and receive data simultaneously, and through spatial diversity and spatial multiplexing technology, the data transmission rate and communication quality can be increased, especially in complex environments, and can better resist signal fading and multipath interference.

[0036] Multi-antenna systems can effectively resist various electromagnetic interferences and improve the robustness of communications through diversity reception and transmission. This is especially important for robots working in harsh conditions, as they may encounter various electromagnetic interference sources.

[0037] The four wireless communication antennas, together with the baseband processing unit, can support more advanced wireless communication protocols such as 802.11ac or 802.11ax (Wi-Fi5 and Wi-Fi6), which can provide higher data transmission rates and lower latency to meet the robot's requirements for real-time and high bandwidth.

[0038] The baseband processing unit can flexibly configure the four wireless communication antennas and select the best antenna working mode, such as antenna selection diversity, antenna switching diversity or maximum ratio combining, according to the actual communication environment and needs, to optimize communication performance.

[0039] The visual analysis module is wirelessly connected to the remote server through the communication unit, so that the remote server can receive the video data and analysis results collected by the robot in real time, and send control instructions as needed to achieve remote monitoring and control of the robot.

[0040] The design of the antenna communication unit significantly improves the robot's communication performance under harsh working conditions, including signal coverage, communication quality, anti-interference capability and communication flexibility, providing a strong guarantee for the robot's efficient and reliable operation.

[0041] Optionally, the robot walking mechanism is an electric trolley, and an electrical wire opening is provided at the bottom of the double-layer shell assembly; the anti-interference visual detection unit, the autonomous handling robot arm unit and the communication unit respectively draw power from the battery pack of the robot walking mechanism through wires.

[0042] By adopting the above technical solution, the anti-interference visual detection unit, the autonomous handling robot arm unit and the communication unit are centrally powered by the battery pack of the robot walking mechanism, which simplifies the design of the power supply system, avoids the need to configure a separate power supply for each unit, and reduces the complexity and weight of the system. When maintenance or battery replacement is required, only the battery pack of the robot walking mechanism needs to be operated, and there is no need to handle the power supply of each unit separately, which greatly facilitates maintenance work and reduces downtime.

[0043] A control method for an anti-interference adaptive robot for harsh working conditions, used for controlling the anti-interference adaptive robot for harsh working conditions to perform abnormality recognition during underground operation, comprises the following steps:

[0044] Step 1: The robot walking mechanism moves in the underground tunnel based on the set navigation trajectory and speed;

[0045] Step 2, multiple visual camera modules respectively capture visual images around and on the top of the robot's walking mechanism and transmit them to the visual analysis module;

[0046] Step 3: The visual analysis chip performs target detection on the preprocessed visual image based on the target visual detection algorithm, and identifies abnormal targets based on the comparison of the stored standard underground tunnel visual images. The abnormal targets are foreign objects, tunnel cracks, structural damage, and roof subsidence.

[0047] Step 4: The data analysis chip calculates the position coordinates of the abnormal target in three-dimensional space based on the abnormal target detection results of the visual analysis chip, combined with the position and posture information data of the robot's walking mechanism and the three-dimensional space data of the underground tunnel;

[0048] Step 5: The robot controller interacts with the data analysis chip to obtain the position coordinates of the abnormal target, controls the active end of the robot controller to point to the position coordinates, and turns on the active visual camera module and the metal detector at the same time. The active visual camera module captures the visual image of the abnormal target, and the metal detector performs metal detection on the abnormal target.

[0049] Step 6: The active visual camera module and the metal detector communicate with the visual analysis module to exchange abnormal target visual image data and metal detection data. The visual analysis module communicates with the remote server wirelessly through the communication unit to exchange abnormal target visual image data and metal detection data.

[0050] In summary, the present invention includes at least one of the following beneficial technical effects:

[0051] The present invention can provide an anti-interference adaptive robot and control method for harsh working conditions. For harsh environments with high dust, high humidity and high temperature, the anti-interference visual detection unit adopts a double-layer shell component, which improves the protection performance of the visual detection unit and can work stably under harsh working conditions. Multiple visual camera modules are distributed around and on the top of the double-layer shell component, which can capture visual images in all directions, improving the coverage and accuracy of visual detection.

[0052] The visual analysis module adopts a target-based visual detection algorithm, which can effectively identify and locate anomalies, improving the accuracy and real-time performance of anomaly detection.

[0053] The autonomous disposal robot arm unit integrates an autonomous robot arm, an active visual camera module, a metal detector and a robot arm controller. It can automatically point to the location of foreign objects based on abnormal positioning data and perform corresponding disposal operations, thereby improving the robot's autonomous disposal capabilities.

[0054] Through the study of the composition, structure, ingredients, characteristics and generation process of underground debris, we have overcome the technical difficulties of accurately distinguishing "high-grade ore and iron debris" based on machine vision recognition and metal detection application technology, and achieved safe production in smart mines with unmanned and intelligent debris removal. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a side view structural schematic diagram of the anti-interference adaptive robot used in harsh working conditions of the present invention;

[0056] Figure 2 It is a schematic diagram of the internal structure of the double-layer shell assembly of the anti-interference adaptive robot used in harsh working conditions of the present invention;

[0057] Figure 3 It is a schematic diagram of the connection principle of the electrical components of the anti-interference adaptive robot invented for harsh working conditions;

[0058] Explanation of the accompanying drawings: 1. Robot walking mechanism; 2. Anti-interference visual detection unit; 21. Double-layer shell assembly; 211. Outer shell; 212. Inner shell; 213. Anti-interference buffer layer; 214. Electrical component installation space; 22. Visual camera module; 221. Visual camera; 222. Ring light; 223. Transparent ball cover; 23. Visual analysis module; 231. Multi-channel video acquisition card; 232. Memory; 233. Visual analysis chip; 234. Data analysis chip; 3. Autonomous disposal robot arm unit; 31. Autonomous robot arm; 32. Active visual camera module; 33. Metal detector; 34. Robot arm controller; 341. Instruction cache; 342. Control chip; 4. Communication unit; 41. Wireless communication antenna; 42. Baseband processing unit; 5. Remote server. DETAILED DESCRIPTION

[0059] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0060] The embodiments of the present invention disclose an anti-interference adaptive robot and a control method for harsh working conditions.

[0061] Reference Figure 1 - Figure 3Embodiment 1, an anti-interference adaptive robot for harsh working conditions, comprising a robot walking mechanism 1, an anti-interference visual detection unit 2, an autonomous disposal robot arm unit 3 and a communication unit 4, the anti-interference visual detection unit 2 comprising a double-layer shell component 21, a plurality of visual camera modules 22 and a visual analysis module 23, the bottom of the double-layer shell component 21 is detachably mounted on the top of the robot walking mechanism 1, the four sides and the top of the double-layer shell component 21 are provided with a plurality of openings for installing camera modules, the plurality of visual camera modules 22 are respectively mounted at the plurality of openings of the double-layer shell component 21, for shooting visual images around and on the top, the visual analysis module 23 is respectively connected to the plurality of visual camera modules 22 in communication, the visual analysis module 23 uses a target-based visual detection algorithm to identify and locate anomalies in the visual images shot by the plurality of visual camera modules 22, the autonomous disposal robot arm unit 3 is installed At the top of the double-layer shell component 21, and offset from the position of the visual camera module 22, the autonomous disposal robot arm unit 3 includes an autonomous robot arm 31, an active visual camera module 32, a metal detector 33 and a chip-based robot arm controller 34. The active visual camera module 32 is installed on one side of the active end of the autonomous robot arm 31 to shoot the visual picture pointed to by the active end of the autonomous robot arm 31. The metal detector 33 is installed on the other side of the active end of the autonomous robot arm 31. The robot arm controller 34, the active visual camera module 32 with a spotlight and the metal detector 33 are respectively communicated with the visual analysis module 23. The robot arm controller 34 receives interactive abnormality positioning data and controls the autonomous robot arm 31 to point to the foreign object based on the abnormal positioning data. The communication unit 4 is installed around the top of the double-layer shell component 21. The visual analysis module 23 is wirelessly connected to the remote server 5 through the communication unit 4.

[0062] The robot walking mechanism 1 moves to a designated area, such as an underground tunnel. The anti-interference visual detection unit 2 starts working, and multiple visual camera modules 22 capture visual images of the surroundings and the top. The visual analysis module 23 processes the visual images captured by the camera module 22 and uses a target-based visual detection algorithm to identify and locate anomalies.

[0063] If the visual analysis module 23 identifies an abnormality, the abnormality location data is sent to the robot arm controller 34 of the autonomous handling robot arm unit 3. The robot arm controller 34 receives the abnormality location data and controls the autonomous robot arm 31 to point to the foreign object. The active visual camera module 32 captures the visual image pointed by the active end of the autonomous robot arm 31, and the metal detector 33 detects the foreign object. The robot arm controller 34, the active visual camera module 32 and the metal detector 33 send data to the visual analysis module 23.

[0064] The visual analysis module 23 interacts with the remote server 5 via wireless communication with the communication unit 4 to interact with the abnormal target positive vision data and metal detection data of the visual camera module 32;

[0065] The remote server 5 analyzes and processes the data, determines the nature of the foreign matter, and generates a corresponding disposal strategy. If it is determined that an abnormality occurs, an alarm is issued, and the staff manually controls the robot walking mechanism 1 and the autonomous disposal robot arm unit 3 through the remote control device to approach and further collect data, or they can manually go to the site for disposal;

[0066] In view of the harsh environment of high dust, high humidity and high temperature, the anti-interference visual detection unit 2 adopts a double-layer shell component 21, which improves the protection performance of the visual detection unit and can work stably under harsh working conditions. Multiple visual camera modules 22 are distributed around and on the top of the double-layer shell component 21, which can capture visual images in all directions, thereby improving the coverage and accuracy of visual detection.

[0067] The visual analysis module 23 adopts a target-based visual detection algorithm, which can effectively identify and locate anomalies, thereby improving the accuracy and real-time performance of anomaly detection.

[0068] The autonomous disposal robot arm unit 3 integrates an autonomous robot arm 31, an active visual camera module 32, a metal detector 33 and a robot arm controller 34. It can automatically point to the location of foreign objects according to abnormal positioning data and perform corresponding disposal operations, thereby improving the robot's autonomous disposal capability.

[0069] Through the study of the composition, structure, ingredients, characteristics and generation process of underground debris, we have carried out research on application technology and key technologies based on machine vision recognition and metal detection, and have broken through the technical difficulties of accurately distinguishing "high-grade ore and iron debris". We have achieved safe production in smart mines with unmanned and intelligent debris removal.

[0070] Embodiment 2, the double-layer shell assembly 21 includes an outer shell 211, an inner shell 212 and an anti-interference buffer layer 213, and the outer shell 211 is provided with multiple openings for installing the camera module on the four sides and the top, the bottom of the outer shell 211 is installed on the top of the robot walking mechanism 1, the outer wall of the inner shell 212 is installed on the inside of the outer shell 211 through the rib plate, and multiple openings are also provided on the four sides of the inner shell 212 at positions corresponding to the multiple openings of the outer shell 211, the anti-interference buffer layer 213 is installed on the inner wall of the inner shell 212, and an electrical device installation space 214 is formed inside the anti-interference buffer layer 213, and the visual analysis module 23 is installed in the electrical device installation space 214.

[0071] Multiple layers of protection are provided to enhance the mechanical strength and anti-interference capability of the entire visual inspection unit 2. The housing 211, as the outermost layer of protection, can resist external impact and the influence of harsh environment.

[0072] The multiple openings of the housing 211 are designed to facilitate installation of multiple visual camera modules 22 while maintaining all-round coverage of visual inspection, thereby improving the efficiency and accuracy of visual inspection.

[0073] The inner shell 212 is installed inside the outer shell 211 through a rib plate, which increases the stability and supporting force of the structure. At the same time, the opening of the inner shell 212 corresponds to the opening of the outer shell 211, ensuring the installation position and field of view of the camera module 22.

[0074] The anti-interference buffer layer 213 is installed on the inner wall of the inner shell 212 to form an electrical device installation space 214. The anti-interference buffer layer 213 can be conductive rubber, metal mesh or conductive paint. These materials can block external electromagnetic interference and protect internal electronic components from electromagnetic interference. It provides additional anti-interference protection for electronic components such as the visual analysis module 23. The buffer layer can absorb external vibration and impact, reduce the impact on internal electronic components, and improve the reliability and stability of the entire system.

[0075] The electrical device installation space 214 provides an independent and safe working environment for the visual analysis module 23, which helps to improve its working performance and life. At the same time, this design also helps heat dissipation and electromagnetic shielding, further improving the stability and anti-interference ability of the system.

[0076] The overall structure is compact and the modular design is convenient for maintenance and upgrading. The double-layer shell assembly 21 can be conveniently installed on the top of the robot walking mechanism 1, and is also convenient for disassembly and replacement, which improves the maintenance efficiency and flexibility of the robot.

[0077] The design of the double-layer housing assembly 21 not only improves the anti-interference ability and mechanical strength of the visual detection unit 2, but also provides additional protection for the electronic components inside it, thereby enhancing the stability and reliability of the entire system. At the same time, the modular design also facilitates maintenance and upgrading, thereby improving the overall performance and adaptability of the robot.

[0078] Embodiment 3, the visual camera module 22 includes a visual camera 221 and a ring light 222, the base of the visual camera 221 is installed at the opening of the inner shell 212, the visual camera 221 is communicated with the visual analysis module 23, and the ring light 222 is installed on the outer wall of the opening of the inner shell 212 and surrounds the visual camera 221.

[0079] In Embodiment 4, the visual camera module 22 further includes a transparent ball cover 223 , and the transparent ball cover 223 is installed at the opening of the housing 211 .

[0080] The transparent ball cover 223 is made of tempered glass or acrylic.

[0081] Embodiment 5, the autonomous robotic arm 31 is a miniature multi-axis robotic arm, and the active visual camera module 32 includes a camera housing and an active camera, the camera housing is mounted on the side wall of the active end of the autonomous robotic arm 31, the active camera is mounted at the end of the camera housing, and the lens direction of the active camera is consistent with the direction of the active end of the autonomous robotic arm 31, and the active camera is communicatively connected to the visual analysis module 23.

[0082] Embodiment 6, the visual analysis module 23 includes a multi-channel video acquisition card 231, a memory 232, a visual analysis chip 233 and a data analysis chip 234, the multi-channel video acquisition card 231 is respectively communicated with multiple visual cameras 221 and active cameras, the memory 232 is communicated with the multi-channel video acquisition card 231, the visual analysis chip 233 and the data analysis chip 234 are respectively communicated with the memory 232, and the data analysis chip 234 exchanges the position and posture information data of the robot walking mechanism 1 with the navigation module of the robot walking mechanism 1.

[0083] In Example 7, the robot arm controller 34 includes an instruction cache 341 and a control chip 342. The instruction cache 341 stores instruction set data for controlling the robot arm controller 34, the active visual camera module 32 and the metal detector 33. The control chip 342 is communicatively connected with the instruction cache 341. The control chip 342 controls the execution actions of the robot arm controller 34, the active visual camera module 32 and the metal detector 33 respectively based on the instruction set data.

[0084] In Example 8, the communication unit 4 includes four wireless communication antennas 41 and a baseband processing unit 42. The four wireless communication antennas 41 are respectively installed around the top of the robot walking mechanism 1. The four wireless communication antennas 41 are respectively connected to the visual analysis module 23 through the baseband processing unit 42, and the visual analysis module 23 is connected to the remote server 5 through the communication unit 4.

[0085] Four wireless communication antennas 41 are installed around the top of the robot walking mechanism 1. This layout ensures that when the robot moves or turns, at least one antenna can maintain a good communication connection with the remote server 5. This all-round signal coverage improves the stability and reliability of communication and reduces communication interruptions caused by signal blind spots.

[0086] Multi-antenna systems (such as MIMO, multiple-input multiple-output technology) can significantly improve the capacity and efficiency of wireless communications. Four antennas can send and receive data simultaneously, and through spatial diversity and spatial multiplexing technology, the data transmission rate and communication quality can be increased, especially in complex environments, and can better resist signal fading and multipath interference.

[0087] Multi-antenna systems can effectively resist various electromagnetic interferences and improve the robustness of communications through diversity reception and transmission. This is especially important for robots working in harsh conditions, as they may encounter various electromagnetic interference sources.

[0088] The four wireless communication antennas 41 cooperate with the baseband processing unit 42 to support more advanced wireless communication protocols, such as 802.11ac or 802.11ax (Wi-Fi5 and Wi-Fi6). These protocols can provide higher data transmission rates and lower latency to meet the robot's requirements for real-time and high bandwidth.

[0089] The baseband processing unit 42 can flexibly configure the four wireless communication antennas 41 and select the best antenna working mode, such as antenna selection diversity, antenna switching diversity or maximum ratio combining, according to the actual communication environment and needs, to optimize communication performance.

[0090] The visual analysis module 23 is wirelessly connected to the remote server 5 via the communication unit 4, so that the remote server can receive the video data and analysis results collected by the robot in real time, and send control instructions as needed to achieve remote monitoring and control of the robot.

[0091] The design of the antenna communication unit significantly improves the communication performance of the robot under harsh working conditions, including signal coverage, communication quality, anti-interference ability and communication flexibility, and provides a strong guarantee for the efficient and reliable operation of the robot. The remote server 5 is a computer with a wireless transceiver antenna.

[0092] In Example 9, the robot walking mechanism 1 is an electric trolley, an electrical wire opening is provided at the bottom of the double-layer shell assembly 21, and the anti-interference visual detection unit 2, the autonomous handling robot arm unit 3 and the communication unit 4 respectively draw power from the battery pack of the robot walking mechanism 1 through wires.

[0093] The battery pack of the robot walking mechanism 1 provides centralized power supply to the anti-interference visual detection unit 2, the autonomous handling robot arm unit 3, and the communication unit 4, which simplifies the design of the power supply system, avoids the need to configure a separate power supply for each unit, and reduces the complexity and weight of the system. When maintenance or battery replacement is required, only the battery pack of the robot walking mechanism 1 needs to be operated, and there is no need to handle the power supply of each unit separately, which greatly facilitates maintenance work and reduces downtime.

[0094] Embodiment 10, a control method for an anti-interference adaptive robot for harsh working conditions, is used to control the anti-interference adaptive robot for harsh working conditions to perform abnormality recognition when operating underground, comprising the following steps:

[0095] Step 1, the robot walking mechanism 1 moves in the underground tunnel based on the set navigation trajectory and speed;

[0096] Step 2, multiple visual camera modules 22 respectively capture visual images around and on the top of the robot walking mechanism 1 and transmit them to the visual analysis module 23;

[0097] Step 3: The visual analysis chip 233 performs target detection on the preprocessed visual image based on the target visual detection algorithm, and identifies abnormal targets based on the comparison of the stored standard underground tunnel visual images. The abnormal targets are foreign objects, tunnel cracks, structural damage, and roof subsidence.

[0098] Step 4, the data analysis chip 234 calculates the position coordinates of the abnormal target in the three-dimensional space based on the abnormal target detection result of the visual analysis chip 233, combined with the position and posture information data of the robot walking mechanism 1 and the three-dimensional space data of the underground tunnel;

[0099] Step 5: The robot controller 34 interacts with the data analysis chip 234 to obtain the position coordinates of the abnormal target, controls the active end of the robot controller 34 to point to the position coordinates, and turns on the active visual camera module 32 and the metal detector 33 at the same time. The active visual camera module 32 captures the visual image of the abnormal target, and the metal detector 33 performs metal detection on the abnormal target.

[0100] Step 6, the active visual camera module 32 and the metal detector 33 communicate with the visual analysis module 23 to exchange abnormal target visual image data and metal detection data, and the visual analysis module 23 communicates with the remote server 5 wirelessly through the communication unit 4 to exchange abnormal target visual image data and metal detection data.

[0101] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. Anti-interference adaptive robot for harsh working conditions, characterized by: The invention comprises a robot walking mechanism (1), an anti-interference visual detection unit (2), an autonomous handling robot arm unit (3) and a communication unit (4); the anti-interference visual detection unit (2) comprises a double-layer shell component (21), a plurality of visual camera modules (22) and a visual analysis module (23); the bottom of the double-layer shell component (21) is detachably mounted on the top of the robot walking mechanism (1); the four sides and the top of the double-layer shell component (21) are provided with a plurality of openings for mounting camera modules; the plurality of visual camera modules (22) are respectively mounted at the plurality of openings of the double-layer shell component (21) for capturing visual images of the four sides and the top; the visual analysis module (23) is respectively connected to the plurality of visual camera modules (22); the visual analysis module (23) uses a target-based visual detection algorithm to identify and locate anomalies in the visual images captured by the plurality of visual camera modules (22); the autonomous handling robot arm unit (3) is mounted on the top of the double-layer shell component (21); The position of the visual camera module (22) is staggered, and the autonomous handling robot arm unit (3) includes an autonomous robot arm (31), an active visual camera module (32), a metal detector (33) and a chip-based robot arm controller (34). The active visual camera module (32) is installed on one side of the active end of the autonomous robot arm (31) to capture the visual image pointed to by the active end of the autonomous robot arm (31). The metal detector (33) is installed on the other side of the active end of the autonomous robot arm (31). The robot arm controller (34), the active visual camera module with a spotlight (32) and the metal detector (33) are respectively connected to the visual analysis module (23) in communication. The robot arm controller (34) receives interactive abnormality positioning data and controls the autonomous robot arm (31) to point to the foreign object based on the abnormal positioning data. The communication unit (4) is installed around the top of the double-layer shell component (21). The visual analysis module (23) is connected to the remote server (5) in wireless communication via the communication unit (4).

2. The anti-interference adaptive robot for harsh working conditions according to claim 1, characterized in that: The double-layer shell component (21) comprises an outer shell (211), an inner shell (212) and an anti-interference buffer layer (213); the outer shell (211) is provided with a plurality of openings for installing a camera module on its periphery and top; the bottom of the outer shell (211) is installed on the top of a robot walking mechanism (1); the outer wall of the inner shell (212) is installed inside the outer shell (211) via a rib plate; a plurality of openings are also provided on the periphery of the inner shell (212) at positions corresponding to the plurality of openings of the outer shell (211); the anti-interference buffer layer (213) is installed on the inner wall of the inner shell (212); an electrical device installation space (214) is formed inside the anti-interference buffer layer (213); and the visual analysis module (23) is installed in the electrical device installation space (214).

3. The anti-interference adaptive robot for harsh working conditions according to claim 2, characterized in that: The visual camera module (22) comprises a visual camera (221) and an annular light (222), wherein the base of the visual camera (221) is mounted at the opening of the inner shell (212), the visual camera (221) is communicatively connected with the visual analysis module (23), and the annular light (222) is mounted on the outer wall of the opening of the inner shell (212) and surrounds the visual camera (221).

4. The anti-interference adaptive robot for harsh working conditions according to claim 3, characterized in that: The visual camera module (22) further comprises a transparent ball cover (223), wherein the transparent ball cover (223) is installed at the opening of the housing (211).

5. The anti-interference adaptive robot for harsh working conditions according to claim 4, characterized in that: The autonomous robotic arm (31) is a miniature multi-axis robotic arm, and the active visual camera module (32) comprises a camera housing and an active camera, wherein the camera housing is mounted on the side wall of the active end of the autonomous robotic arm (31), and the active camera is mounted on the end of the camera housing, and the lens orientation of the active camera is consistent with the orientation of the active end of the autonomous robotic arm (31), and the active camera is communicatively connected to the visual analysis module (23).

6. The anti-interference adaptive robot for harsh working conditions according to claim 5, characterized in that: The visual analysis module (23) comprises a multi-channel video acquisition card (231), a memory (232), a visual analysis chip (233) and a data analysis chip (234); the multi-channel video acquisition card (231) is respectively connected to multiple visual cameras (221) and active cameras in communication; the memory (232) is connected to the multi-channel video acquisition card (231) in communication; the visual analysis chip (233) and the data analysis chip (234) are respectively connected to the memory (232) in communication; and the data analysis chip (234) exchanges position and posture information data of the robot walking mechanism (1) with a navigation module of the robot walking mechanism (1).

7. The anti-interference adaptive robot for harsh working conditions according to claim 6, characterized in that: The robot arm controller (34) comprises an instruction buffer (341) and a control chip (342), wherein the instruction buffer (341) stores instruction set data for controlling the robot arm controller (34), the active visual camera module (32) and the metal detector (33), and the control chip (342) is communicatively connected with the instruction buffer (341), and the control chip (342) controls the execution actions of the robot arm controller (34), the active visual camera module (32) and the metal detector (33) respectively based on the instruction set data.

8. The anti-interference adaptive robot for harsh working conditions according to claim 7, characterized in that: The communication unit (4) comprises four wireless communication antennas (41) and a baseband processing unit (42). The four wireless communication antennas (41) are respectively installed around the top of the robot walking mechanism (1). The four wireless communication antennas (41) are respectively connected to the visual analysis module (23) through the baseband processing unit (42). The visual analysis module (23) is connected to the remote server (5) through the communication unit (4) through wireless communication.

9. The anti-interference adaptive robot for harsh working conditions according to claim 8, characterized in that: The robot walking mechanism (1) is an electric trolley. An electrical wire opening is provided at the bottom of the double-layer shell component (21). The anti-interference visual detection unit (2), the autonomous handling robot arm unit (3) and the communication unit (4) respectively draw power from the battery pack of the robot walking mechanism (1) through wires.

10. A control method for an anti-interference adaptive robot for harsh working conditions, characterized in that: The method for controlling the anti-interference adaptive robot for harsh working conditions as claimed in claim 9 to perform abnormality recognition during underground operation comprises the following steps: Step 1, the robot walking mechanism (1) moves in the underground tunnel based on the set navigation trajectory and speed; Step 2, multiple visual camera modules (22) respectively capture visual images around and on the top of the robot walking mechanism (1) and transmit them to the visual analysis module (23); Step 3, the visual analysis chip (233) performs target detection on the preprocessed visual image based on the target visual detection algorithm, and identifies abnormal targets based on the comparison of the stored standard underground tunnel visual images, wherein the abnormal targets are foreign objects, tunnel cracks, structural damage, and roof subsidence; Step 4, the data analysis chip (234) calculates the position coordinates of the abnormal target in the three-dimensional space based on the abnormal target detection result of the visual analysis chip (233) and the position and posture information data of the robot walking mechanism (1) and the three-dimensional space data of the underground tunnel; Step 5, the mechanical arm controller (34) exchanges the position coordinates of the abnormal target with the data analysis chip (234), controls the active end of the mechanical arm controller (34) to point to the position coordinates, and simultaneously turns on the active visual camera module (32) and the metal detector (33), the active visual camera module (32) captures the visual image of the abnormal target, and the metal detector (33) performs metal detection on the abnormal target; Step 6, the active visual camera module (32) and the metal detector (33) communicate with the visual analysis module (23) to exchange abnormal target visual image data and metal detection data, and the visual analysis module (23) communicates with the remote server (5) wirelessly through the communication unit (4) to exchange abnormal target visual image data and metal detection data.

Images