A snake-shaped inspection robot and inspection method for confined spaces

By using a snake-shaped inspection robot for narrow spaces, and employing a flexible drive body with multiple sets of drive joints connected in series and dual-mode positioning technology, the problem of passage and positioning in narrow and concealed spaces in power systems has been solved, realizing intelligent inspection and remote monitoring.

CN123103604APending Publication Date: 2026-11-17HUADIAN LAIZHOU POWER GENERATION
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611287867.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-24
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing wheeled or tracked inspection robots have difficulty moving flexibly in the narrow and concealed spaces of power systems, and cannot accurately locate or achieve remote real-time monitoring in areas without satellite signals.

Method used

The robot employs a snake-shaped design for inspecting confined spaces. It uses multiple sets of drive joints connected in series to form a flexible drive body. It combines satellite positioning and inertial navigation trajectory calculation for positioning and uses 5G and WiFi dual-mode communication to transmit data back.

Benefits of technology

It enables robots to walk autonomously and avoid obstacles in confined spaces, solves the problems of accurate positioning and fault point identification in areas without signal, and realizes intelligent inspection and remote real-time monitoring of power-concealed spaces.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123103604A_ABST
    Figure CN123103604A_ABST
Patent Text Reader

Abstract

This application provides a snake-shaped inspection robot and method for confined spaces, belonging to the field of intelligent power inspection equipment technology. The robot includes a head detection cabin, a mid-section flexible drive body, a tail energy communication cabin, and a main controller. The head detection cabin collects environmental perception data. The mid-section flexible drive body is composed of multiple sets of drive joints connected in series, each joint having a built-in servo motor that drives the robot to move autonomously through coordinated swinging. The tail energy communication cabin houses a power supply module, a positioning module, and a communication module. The main controller identifies the passage conditions based on the environmental perception data, matches control strategies for straight-through passage, curve passage, or narrow-slit crawling, and outputs PWM control signals to each joint to achieve autonomous walking and obstacle avoidance. The positioning module acquires real-time location information, and the communication module transmits the data outward. This application solves the problems of difficult passage in confined spaces and positioning in areas without signal, realizing fully automatic unmanned inspection of hidden power spaces.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application belongs to the field of power intelligent inspection equipment technology, specifically relating to a snake-shaped inspection robot and inspection method for narrow spaces. Background Technology

[0002] In power systems, a large number of cables, busbars, and pipelines are located in concealed spaces such as underground trenches, enclosed mezzanines, and narrow flues. These concealed spaces are generally narrow, poorly lit, prone to water and dust accumulation, and have complex electromagnetic environments, making manual on-site inspection impossible. As the power grid continues to expand, the aging of equipment and potential hazards in these concealed areas are becoming increasingly serious, making inspection a pressing need.

[0003] Currently, wheeled or tracked inspection robots are mainly used for replacement work, but existing equipment has significant shortcomings in applications in concealed power spaces. On the one hand, wheeled and tracked robots have large turning radii and high body rigidity, making them difficult to adapt to curves, narrow gaps, and variable-diameter passages. They have poor maneuverability in confined spaces and are prone to getting stuck or even overturning. On the other hand, underground enclosed spaces lack satellite signal coverage, rendering conventional positioning methods unusable. After anomalies are discovered during inspections, it is difficult to accurately pinpoint the fault location. At the same time, the high latency of on-site data transmission makes remote real-time monitoring impossible, making it difficult for maintenance personnel to keep track of equipment status in a timely manner.

[0004] Therefore, there is an urgent need for an inspection device that can flexibly navigate through narrow and concealed electrical spaces, possess autonomous walking and intelligent obstacle avoidance capabilities, achieve precise positioning in areas without signal, and support remote real-time data transmission. Summary of the Invention

[0005] In a first aspect, embodiments of this application provide a snake-shaped confined space inspection robot, including a head detection cabin, a mid-section flexible drive torso, a tail energy and communication cabin, and a main controller; The head detection cabin is located at the very front of the robot and is used to collect environmental perception data; The mid-section flexible drive torso is located at the rear end of the head probe cabin. The mid-section flexible drive torso includes N sets of drive joints, which are connected in series to form N sets of drive joints, where N≥2. Each set of drive joints is equipped with a servo motor, which is used to drive the corresponding drive joint to perform left and right yaw and / or up and down pitch movements. The flexible drive trunk in the middle section drives the robot to walk autonomously along the inspection channel through the coordinated swinging of each set of drive joints. The tail energy and communication compartment is located at the rear end of the mid-section flexible drive body, and the tail energy and communication compartment is equipped with a power supply module, a positioning module and a communication module; The power supply module supplies power to the head probe compartment, the mid-section flexible drive body and the tail energy and communication compartment through a power supply line that runs through the interior of the mid-section flexible drive body. The positioning module is used to obtain the robot's real-time location information; The communication module is used to transmit environmental sensing data and real-time location information to the outside world; The main controller is connected to the head probe module, the mid-section flexible drive body, and the tail power and communication module, and is configured as follows: It receives environmental perception data and real-time location information, and outputs control signals to the servo motors of each drive joint according to the environmental perception data, controlling each drive joint to coordinate actions according to a preset timing sequence, so as to realize the robot's autonomous walking and obstacle avoidance in a narrow space, and mark the fault location according to the real-time location information.

[0006] Secondly, embodiments of this application also provide an inspection method based on the snake-shaped confined space inspection robot described in the first aspect, comprising the following steps: S1. Collect environmental perception data through the head-mounted detection cabin; S2. The main controller receives environmental perception data and identifies the current passage conditions of the inspection channel based on the environmental perception data. It then selects a target walking control strategy that matches the current passage conditions from a variety of pre-stored bionic walking control strategies. S3. The main controller generates joint angle timing instructions for each drive joint according to the target walking control strategy, outputs PWM control signals to the servo motors of each drive joint, controls each drive joint to perform left and right yaw and / or up and down pitching actions according to the preset timing, and drives the robot to walk autonomously along the inspection channel. S4. Acquire the robot's real-time position information during the robot's autonomous walking process; S5. Transmit environmental perception data and real-time location information to the outside world through the communication module.

[0007] As can be seen from the above technical solutions, this application has the following advantages: The serpentine confined space inspection robot and inspection method provided in this application achieve serpentine biomimetic motion by using a flexible drive torso composed of multiple sets of drive joints connected in series. This allows the robot to flexibly navigate complex and confined spaces such as bends and narrow gaps that are difficult for traditional wheeled or tracked robots to pass through, thus improving its passability. At the same time, the robot integrates satellite positioning and inertial navigation trajectory calculation for dual positioning, solving the problems of accurate positioning and fault point identification in areas without satellite signals, such as underground pipe trenches. In addition, through the main controller's intelligent identification of traffic conditions and matching of corresponding walking strategies, the robot achieves autonomous walking, obstacle avoidance, and inspection. Combined with 5G and WiFi dual-mode communication, it transmits high-definition video, equipment temperature, environmental parameters, and real-time location data back to the monitoring backend, realizing intelligent inspection and remote real-time monitoring of power concealed spaces. Attached Figure Description

[0008] To more clearly illustrate the technical solution of this application, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the overall structure of the snake-shaped narrow space inspection robot of the present invention.

[0010] Figure 2 This is a front view schematic diagram of the head detection chamber of the snake-shaped confined space inspection robot of the present invention.

[0011] Figure 3 This is a schematic diagram of the control of the snake-shaped narrow space inspection robot of the present invention.

[0012] Figure 4 This is a flowchart illustrating the inspection method of the snake-shaped confined space inspection robot of the present invention.

[0013] The components are as follows: 10-Head detection cabin; 11-Flexible anti-collision buffer cover; 12-High-definition night vision camera module; 13-Infrared non-contact temperature detection probe; 14-Laser ranging obstacle avoidance sensor; 15-Ring LED supplementary light; 16-Temperature and humidity sensor; 17-Harmful gas detection sensor; 18-Water level detection sensor; 19-Environmental parameter acquisition module; 20-Drive joint; 21-Attitude gyroscope; 22-Tilt sensor; 23-Walking mileage counting module; 30-Tail energy and communication cabin; 31-Power supply module; 32-Positioning module; 321-BeiDou and GPS dual-mode satellite positioning unit; 322-Inertial navigation trajectory calculation and positioning unit; 33-Communication module; 331-5G wireless transmission unit; 332-WiFi wireless transmission unit; 34-Communication antenna; 35-Working status indicator light; 40-Main controller. Detailed Implementation

[0014] Various embodiments of this disclosure will be described more fully in the following detailed description of the snake-shaped confined space inspection robot. This disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of this disclosure to the specific embodiments disclosed herein, but rather this disclosure should be understood to cover all adjustments, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of this disclosure.

[0015] This embodiment provides a snake-shaped inspection robot for narrow spaces, which enables flexible passage through narrow spaces and precise positioning in areas without signals, and completes fully automated unmanned inspection of power-concealed spaces.

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 and Figure 3 The diagram shown is a schematic of a snake-shaped confined space inspection robot in a specific embodiment. The robot includes a head detection cabin 10, a mid-section flexible drive torso, a tail energy and communication cabin 30, and a main controller 40. The head detection cabin 10 is located at the very front of the robot and is used to collect environmental perception data; It should be noted that the head detection cabin 10 is located at the very front of the entire machine to collect environmental perception data, enabling the robot to continuously perceive the environment and equipment status of the passage ahead during movement, providing a data basis for control decisions; The mid-section flexible drive torso is located at the rear end of the head probe cabin. The mid-section flexible drive torso includes N sets of drive joints 20, which are connected in series to form N sets of drive joints 20, where N≥2. Each set of drive joints 20 is equipped with a servo motor, which is used to drive the corresponding drive joint 20 to perform left and right yaw and / or up and down pitch movements. The flexible drive trunk in the middle section drives the robot to walk autonomously along the inspection channel through the coordinated swinging of each set of drive joints 20. It should be noted that the mid-section flexible drive torso is composed of N sets of drive joints connected in series (N≥2). Each set of drive joints has a built-in servo motor to perform left and right yaw and up and down pitch movements. The biomimetic walking drive force is generated by the coordinated swing of multiple drive joints. The robot can flexibly pass through curves, narrow gaps and variable diameter channels without turning radius. Its passage ability is better than that of traditional wheeled and tracked structures. The tail energy communication compartment 30 is located at the rear end of the mid-section flexible drive body. The tail energy communication compartment 30 is equipped with a power supply module 31, a positioning module 32 and a communication module 33. The power supply module 31 supplies power to the head probe compartment 10, the mid-section flexible drive body and the tail energy communication compartment 30 through a power supply line that runs through the interior of the mid-section flexible drive body; The positioning module 32 is used to acquire the robot's real-time position information; Communication module 33 is used to transmit environmental perception data and real-time location information to the outside. It should be noted that the tail energy communication compartment 30 has a built-in power supply module 31, positioning module 32 and communication module 33. The power supply module 31 supplies power to all compartments of the whole machine through the power supply line that runs through the inside of the body, realizing integrated energy management; the positioning module 32 obtains real-time location information, and the communication module 33 transmits environmental perception data and location information to the outside, so that the background can keep track of the robot's position and the working conditions of the inspection site in real time. The main controller 40 is connected to the head probe compartment 10, the mid-section flexible drive torso, and the tail power and communication compartment 30, and is configured as follows: It receives environmental perception data and real-time location information, and outputs control signals to the servo motors of each drive joint 20 according to the environmental perception data, controlling each drive joint 20 to coordinate actions according to a preset timing sequence, so as to realize the robot's autonomous walking and obstacle avoidance in a narrow space, and to mark the fault location according to the real-time location information. It should be noted that the main controller 40 receives environmental perception data and real-time position information, generates control signals based on the environmental perception data to drive each drive joint 20 to move in coordination to achieve autonomous walking and obstacle avoidance, and marks the fault location based on the real-time position information, so that the robot can complete the inspection task independently under the condition of no human intervention, and can accurately locate the fault section after discovering the abnormality, providing accurate location basis for maintenance.

[0018] This embodiment achieves flexible passage through narrow spaces through a serpentine flexible structure. Combined with dual-mode positioning and multi-parameter detection, it solves the problems of manual entry, easy jamming of conventional equipment, and inability to locate in areas without signal.

[0019] Furthermore, as a refinement and extension of the specific implementation of the above embodiments, in order to fully illustrate the specific implementation process in this embodiment, another snake-shaped narrow space inspection robot is provided, which includes a head detection cabin 10, a mid-section flexible drive torso, a tail energy and communication cabin 30, and a main controller 40. The head detection cabin 10 is located at the very front of the robot and is used to collect environmental perception data; The mid-section flexible drive torso is located at the rear end of the head probe cabin. The mid-section flexible drive torso includes N sets of drive joints 20, which are connected in series to form N sets of drive joints 20, where N≥2. Each set of drive joints 20 is equipped with a servo motor, which is used to drive the corresponding drive joint 20 to perform left and right yaw and / or up and down pitch movements. The flexible drive trunk in the middle section drives the robot to walk autonomously along the inspection channel through the coordinated swinging of each set of drive joints 20. The tail energy communication compartment 30 is located at the rear end of the mid-section flexible drive body. The tail energy communication compartment 30 is equipped with a power supply module 31, a positioning module 32 and a communication module 33. The power supply module 31 supplies power to the head probe compartment 10, the mid-section flexible drive body and the tail energy communication compartment 30 through a power supply line that runs through the interior of the mid-section flexible drive body; The positioning module 32 is used to acquire the robot's real-time position information; Communication module 33 is used to transmit environmental perception data and real-time location information to the outside. The main controller 40 is connected to the head probe compartment 10, the mid-section flexible drive torso, and the tail power and communication compartment 30, and is configured as follows: It receives environmental perception data and real-time location information, and outputs control signals to the servo motors of each drive joint 20 according to the environmental perception data, controlling each drive joint 20 to coordinate actions according to a preset timing sequence, so as to realize the robot's autonomous walking and obstacle avoidance in a narrow space, and to mark the fault location according to the real-time location information. like Figure 2 As shown, the head detection cabin 10 has a cone-shaped or flat shell structure with a small front end and a large rear end. The head detection cabin 10 has a front end face, on which a high-definition night vision camera module 12, an infrared non-contact temperature detection probe 13, a laser rangefinder obstacle avoidance sensor 14, a ring LED fill light 15, and a flexible anti-collision buffer cover 11 are installed. The high-definition night vision camera module 12 is embedded in the center of the front end face. The lens of the high-definition night vision camera module 12 faces forward, and the end face of the lens is flush with or concave to the front end face. An infrared non-contact temperature detection probe 13 is embedded in the front end face and located on the left or right side of the high-definition night vision camera module 12. The detection surface of the infrared non-contact temperature detection probe 13 faces forward and is flush with or recessed into the front end face. The laser rangefinder obstacle avoidance sensor 14 is embedded in the front end face and is located above or below the high-definition night vision camera module 12. The emitting surface and the receiving surface of the laser rangefinder obstacle avoidance sensor 14 are both facing forward, and the emitting surface and the receiving surface are flush with or recessed into the front end face. The ring-shaped LED fill light 15 is embedded in the outer periphery of the front end face, surrounding the high-definition night vision camera module 12, the infrared non-contact temperature detection probe 13 and the laser rangefinding obstacle avoidance sensor 14, and the light-emitting surface of the ring-shaped LED fill light 15 faces forward. A flexible anti-collision buffer shield 11 is disposed on the outer peripheral edge of the front end face and protrudes forward. The flexible anti-collision buffer shield 11 surrounds the periphery of the ring LED supplement light 15, and the foremost end of the flexible anti-collision buffer shield 11 is located in front of the lens end face of the high-definition night vision camera module 12, the detection surface of the infrared non-contact temperature detection probe 13, and the emitting surface of the laser rangefinding obstacle avoidance sensor 14. The flexible anti-collision buffer shield 11 is used to absorb impact energy when a collision occurs. The high-definition night vision camera module 12, the infrared non-contact temperature detection probe 13, the laser rangefinder obstacle avoidance sensor 14, and the ring LED fill light 15 are all connected to the main controller 40. In this embodiment, the front end face of the head probe 10 has a circular or rectangular outline, and the specific layout relationship of each sensor is as follows: The high-definition night vision camera module 12 is embedded in the central area of ​​the front face, and its lens optical axis is parallel to the robot's walking direction; The infrared non-contact temperature detection probe 13 is located on the left or right side of the camera module 12, with a horizontal distance of 10-30mm between them to avoid mutual obstruction. The laser rangefinder obstacle avoidance sensor 14 is located above or below the camera module 12, with its emitting and receiving surfaces slightly protruding from the front end surface or flush with it. A ring-shaped LED fill light 15 surrounds the periphery of each of the above sensors, with the light-emitting surface facing forward. It automatically lights up when there is insufficient light, providing uniform fill light illumination for the camera module 12. A flexible anti-collision buffer shield 11 is located at the outermost perimeter of the front end face, forming a ring-shaped frame structure made of flexible buffer material (such as silicone rubber or polyurethane elastomer). The front end of the flexible anti-collision buffer shield 11 protrudes forward by 10-30mm, and its front edge extends beyond the lens end face of the high-definition night vision camera module 12, the detection surface of the infrared non-contact temperature detection probe 13, and the emitting surface of the laser ranging obstacle avoidance sensor 14, ensuring that the flexible anti-collision buffer shield 11 contacts the obstacle first when the robot's head collides with it. A 2-5mm gap is left between the inner peripheral wall of the impact shield 11 and the outer peripheral wall of the ring LED supplementary light 15 to prevent the heat of the supplementary light from being directly conducted to the flexible impact buffer shield 11, which would cause the material to age. When the robot head accidentally hits the channel wall or an obstacle, the flexible impact buffer shield 11 undergoes elastic deformation to absorb the impact energy. At the same time, its ring structure disperses the collision force to the shell of the head detection cabin 10, effectively protecting the high-definition night vision camera module 12, the infrared non-contact temperature detection probe 13, and the laser rangefinding obstacle avoidance sensor 14 from direct impact damage. The head probe compartment 10 is also equipped with side walls; The sidewall is embedded with at least one of a temperature and humidity sensor 16, a harmful gas detection sensor 17, and a water level detection sensor 18. In this embodiment, the water level detection sensor 18 is a photoelectric water level sensor. The photoelectric water level sensor works based on the principle of light reflection and refraction at the interface of two different media. It is flexible in installation and can be installed laterally on the side wall of the head detection chamber. It determines whether the water level has reached the preset warning height by detecting whether there is liquid at the probe and outputs a switch signal to the main controller. The sensor probe is flush with or slightly concave with the side wall to prevent it from being damaged by collisions with foreign objects during robot movement. The head probe cabin 10 also integrates an environmental parameter acquisition module 19. The environmental parameter acquisition module 19 is connected to the temperature and humidity sensor 16, the harmful gas detection sensor 17 and the water level detection sensor 18, and is used to receive and adjust the environmental parameter signals acquired by each sensor. The environmental parameter acquisition module 19 is connected to the main controller 40 and is used to output the conditioned environmental parameter signal to the main controller 40. In this embodiment, the environmental parameter acquisition module 19 is integrated inside the head detection chamber 10. The environmental parameter acquisition module 19 is an acquisition board containing a microcontroller and a multi-channel signal conditioning circuit. The temperature and humidity sensor 16, the hazardous gas detection sensor 17, and the water level detection sensor 18 are respectively embedded in the side wall of the head detection chamber 10. The signal output terminals of each sensor are introduced into the environmental parameter acquisition module 19 of the head detection chamber 10 through shielded wires. Among them, the temperature and humidity sensor 16 adopts a digital temperature and humidity integrated chip and outputs digital temperature and humidity values ​​to the environmental parameter acquisition module 19 through the I²C bus; the hazardous gas detection sensor 16... 7. An electrochemical or semiconductor gas sensing element is used to output an analog voltage signal (e.g., 0-3.3V) that is proportional to the gas concentration. After being sampled by the operational amplifier conditioning and analog-to-digital converter (e.g., 12-bit precision ADC) built into the environmental parameter acquisition module 19, it is converted into a digital signal. The water level detection sensor 18 uses an electrode or capacitive level probe to output a switch signal or an analog signal. The environmental parameter acquisition module 19 acquires, filters, and converts the signals from each sensor through its internal microcontroller, and then outputs them to the main controller 40 through a serial communication interface (e.g., UART or CAN bus). Each set of drive joints 20 of the mid-section flexible drive trunk is covered with a flame-retardant, insulating, and waterproof flexible shell. Flame-retardant, insulating, and waterproof flexible shells are sequentially fitted onto the outer periphery of each set of drive joints along the length of the middle flexible drive body. A gap or flexible connecting section is provided between two adjacent sets of flame-retardant, insulating, and waterproof flexible shells to provide bending freedom. In this embodiment, the total number N of drive joints 20 is preferably 8 groups; each group of drive joints 20 is an independent deflection unit, and adjacent groups of drive joints 20 are connected in series through a hinge structure. Each group of joints has a built-in micro servo motor, which is a hollow cup DC servo motor with a precision reduction gearbox, and the output torque is not less than 2.5 kg·cm, which is sufficient to drive the robot to move in a channel with a slight slope; the output shaft of the servo motor is coaxial with the deflection axis of the joint, and the joint is driven to perform left and right deflection or up and down pitching relative to the adjacent joint by the rotation of the servo motor output shaft; the deflection axes of adjacent groups of drive joints 20 are arranged alternately orthogonally in space, that is, the deflection axis of the first group of joints is in the vertical direction to achieve left and right deflection, the deflection axis of the second group of joints is in the horizontal direction to achieve up and down pitching, the third group of joints is in the vertical direction again, and so on; this alternating orthogonal deflection axis configuration enables the robot torso to have omnidirectional bending capability in three-dimensional space, which can adapt to both horizontal curves and complex trench terrain with undulating slopes. The flame-retardant, insulating, and waterproof flexible shell is made of silicone rubber with a flame-retardant rating of V-0 and a Shore hardness of 50-70A. It is fitted onto the outside of each set of drive joints 20. A gap of 1-3mm is provided between two adjacent sets of flexible shells. This gap remains uniform when the joint is in a straight state. When the joint deflects, the gap on the narrow side decreases and the gap on the wide side increases, providing physical space for the joint deflection. Alternatively, two adjacent sets of flexible shells are connected by a flexible connecting section (such as a corrugated tubular structure or a helical spring structure). The axial expansion rate of this flexible connecting section is not less than 20%, and the radial bending angle is not less than ±30°. This provides bending freedom while ensuring the continuous sealing of the shell, effectively preventing external dust and moisture from entering the interior of the drive joint 20. The mid-section flexible drive body has an integrated wiring compartment inside; The integrated wiring compartment runs through each set of drive joints 20 along the length of the flexible drive body in the middle section. Power supply lines, signal lines and control lines are arranged in the integrated wiring compartment. The power supply lines, signal lines and control lines are provided with flexible and fold-resistant segments at corresponding positions of adjacent sets of drive joints 20. The flexible drive torso in the middle section is also equipped with an attitude gyroscope 21, an angle sensor 22, and a walking distance counting module 23. The attitude gyroscope 21, tilt sensor 22, and mileage counting module 23 are all connected to the main controller 40; The attitude gyroscope 21 is used to collect the robot's heading angle data and output it to the main controller 40; The tilt sensor 22 is used to collect the tilt angle data of the robot and output it to the main controller 40; The walking distance counting module 23 is used to collect the robot's walking distance increment data and output it to the main controller 40; The main controller 40 corrects the robot's walking posture in real time based on heading angle data, tilt angle data and walking distance increment data, and calculates the robot's real-time position in areas without satellite signals by combining trajectory estimation algorithms. In this embodiment, the main controller 40 has preset attitude control parameters, including: tilt angle reference value of 0°, tilt angle allowable deviation range [-5°, +5°], heading angle reference value, and heading angle deviation threshold of ±3°; the attitude gyroscope 21 and the tilt sensor 22 continuously collect the robot's heading angle data at a sampling frequency of not less than 50Hz. and tilt angle data The walking distance counting module 23 records the cumulative number of swings of the drive joint 20 or collects the incremental walking distance through a miniature encoder installed inside the drive joint. ; The main controller 40 will collect the tilt angle data in real time. Compare with the standard tilt range: when When the robot exceeds the standard range, it is determined that there is a risk of sideslip or tipping. At this time, the main controller 40 generates an attitude correction command and outputs a correction PWM control signal to the servo motors of each drive joint 20 to adjust the deflection angle of each joint to reduce the tilt amplitude of the robot's torso and restore the robot to a normal walking posture. At the same time, the main controller 40 will collect the heading angle data. The data is accumulated in real time. When the accumulated heading angle deviation exceeds the preset threshold, it is determined that the robot has deviated from its travel direction. At this time, the main controller 40 generates a corrective torque by adjusting the left and right deflection angles of each drive joint 20, so that the robot gradually returns to the preset travel direction. In the actual walking process, the tilt angle data and heading angle data are coupled with each other. The main controller 40 uses a PID control algorithm to jointly solve and coordinate the two to ensure that the robot maintains a stable walking posture on uneven ground. The tail-end energy and communication compartment 30 has an outer wall surface; A communication antenna 34 is provided on the top of the outer wall surface. The communication antenna 34 is connected to the communication module 33 and is used to enhance the transmission and reception of wireless signals. The outer wall of the tail energy communication compartment 30 is also equipped with a working status indicator light 35, which is connected to the main controller 40 and is used to indicate the robot's running status, fault status or standby status. The positioning module 32, communication module 33, and main controller 40 are all located inside the tail energy and communication compartment 30; The power supply module 31 includes a low-voltage lithium battery pack and a power supply regulator module. The output of the low-voltage lithium battery pack is connected to the head probe cabin 10, the mid-section flexible drive body and the tail energy communication cabin 30 after being regulated by the power supply module, and is used to provide low-voltage DC safe power supply. The positioning module 32 includes a Beidou and GPS dual-mode satellite positioning unit 321 and an inertial navigation trajectory calculation and positioning unit 322. The Beidou and GPS dual-mode satellite positioning unit 321 is used to obtain the robot's geographical coordinates in the satellite signal coverage area, and the inertial navigation trajectory calculation and positioning unit 322 is used to calculate the robot's real-time position in the area without satellite signal based on the walking distance, deflection angle and travel time. The communication module 33 includes a 5G wireless transmission unit 331 and a WiFi wireless transmission unit 332. Both the 5G wireless transmission unit 331 and the WiFi wireless transmission unit 332 are connected to the communication antenna 34 and are used to transmit environmental perception data and real-time location information to the outside. The tail-end energy and communication compartment 30 is also equipped with a remote wireless emergency stop receiver module 36; The remote wireless emergency stop receiver module 36 is connected to the main controller 40 and is used to receive remote emergency stop commands and output emergency stop trigger signals to the main controller 40.

[0020] like Figure 4 As shown, the following are embodiments of the inspection method of the snake-shaped confined space inspection robot provided in this disclosure. This method belongs to the same inventive concept as the snake-shaped confined space inspection robot in the above embodiments. For details not described in detail in the embodiments of the inspection method of the snake-shaped confined space inspection robot, please refer to the embodiments of the snake-shaped confined space inspection robot described above.

[0021] The method includes the following steps: S1. Collect environmental perception data through the head-mounted detection cabin; It should be noted that by using multiple types of sensors in the head detection cabin to simultaneously acquire information on the distribution of obstacles and the surface condition of equipment in the inspection channel, a data foundation is provided for working condition identification and control strategy selection. S2. The main controller receives environmental perception data and identifies the current passage conditions of the inspection channel based on the environmental perception data. It then selects a target walking control strategy that matches the current passage conditions from a variety of pre-stored bionic walking control strategies. It should be noted that this step enables the robot to switch control modes according to the characteristics of different channels, avoiding the shortcomings of a single control mode in the complex and ever-changing underground utility tunnel, and realizing adaptive walking in all scenarios. S3. The main controller generates joint angle timing instructions for each drive joint according to the target walking control strategy, outputs PWM control signals to the servo motors of each drive joint, controls each drive joint to perform left and right yaw and / or up and down pitching actions according to the preset timing, and drives the robot to walk autonomously along the inspection channel. It should be noted that this step transforms the control strategy into walking drive force. By precisely controlling the deflection and pitch of each joint, a snake-like bionic motion is achieved, enabling the robot to smoothly and efficiently traverse complex passages. S4. Acquire the robot's real-time position information during the robot's autonomous walking process; It should be noted that this step enables the robot to continuously update its own position coordinates while performing inspection tasks. In areas without signal, it maintains its positioning function by calculating its trajectory, ensuring that each frame of detection data can be accurately bound to the collection location, providing a position data foundation for the accurate marking of faulty sections. S5. Transmit environmental perception data and real-time location information to the outside world through the communication module; It should be noted that the communication module transmits environmental perception data and real-time location information to the outside world, enabling maintenance personnel to remotely monitor the on-site situation and achieve remote visual management.

[0022] In this embodiment, the robot is deployed from the manhole cover at the entrance of the cable trench. After the operator clicks the "Start Inspection" button on the monitoring backend, the robot automatically performs the inspection task according to the following complete process: First, the robot performs an initialization self-test process; the power supply module in the tail energy and communication cabin is powered on, and the main controller sequentially checks whether the sensors in the head detection cabin, the attitude gyroscope and tilt sensor in the mid-section flexible drive torso, and the positioning module and communication module in the tail energy and communication cabin are working properly; after each module passes the self-test, the ring LED fill light flashes twice to indicate the ready state, and the working status indicator light changes from red standby to green running. Subsequently, the robot begins autonomous inspection. The main controller reads the coordinates of the starting and target points of this inspection task from the pre-stored inspection path configuration, obtains the current real-time location information through the positioning module, and plans the inspection path accordingly. The robot walks autonomously along the planned path, executing steps S1 to S5 in real time during the process. When it reaches a cable joint, the infrared non-contact temperature detection probe detects that the surface temperature of the joint is 67.5℃, exceeding the preset alarm threshold of 60℃. The main controller immediately binds the temperature data with the current real-time location information, generates an alarm data packet containing the temperature anomaly, location coordinates, and acquisition time, and uploads it to the monitoring backend in real time through the communication module. At the same time, the robot pauses its advance and stays at the location for 3 seconds to retest and confirm. After confirming the temperature anomaly, the robot continues to move forward to complete the remaining inspection path. After the inspection task is completed, the robot automatically returns to the entrance manhole cover, and the main controller generates a complete inspection report containing all detection data and marked fault location information.

[0023] This embodiment solves the problem of traditional robots having difficulty navigating narrow bends through a serpentine flexible body design; combined with satellite and inertial navigation dual positioning, it achieves accurate positioning and fault calibration in areas without signal; the robot can autonomously identify working conditions, intelligently avoid obstacles, and transmit multi-source data in real time, realizing intelligent inspection and remote monitoring of power-concealed spaces.

[0024] Furthermore, as a refinement and extension of the specific implementation of the above embodiments, in order to fully illustrate the specific implementation process of this embodiment, another inspection method for a snake-shaped confined space inspection robot is provided, which includes the following steps: S1. Collect environmental perception data through the head-mounted detection cabin; The specific steps of step S1 are as follows: S11. The laser rangefinder and obstacle avoidance sensor embedded in the front end of the head detection cabin emits a laser beam forward and receives the reflected echo signal to obtain the distance data of the obstacle in front. S12. High-definition night vision camera module embedded in the front surface of the head detection cabin collects video image data in front of the inspection channel; S13. The head detection cabin transmits the distance data of obstacles in front and video image data as environmental perception data to the main controller; S2. The main controller receives environmental perception data and identifies the current passage conditions of the inspection channel based on the environmental perception data. It then selects a target walking control strategy that matches the current passage conditions from a variety of pre-stored bionic walking control strategies. The specific steps of step S2 are as follows: S21. The main controller extracts distance data of obstacles ahead and / or video image data from the environmental perception data; S22. The main controller determines the channel width, channel curvature and / or channel cross-sectional dimensions of the current inspection channel based on the distance data of the obstacle ahead and / or video image data, and identifies the current passage condition as straight pipe straight passage condition, curved passage condition or narrow gap creep passage condition; S23. Based on the identified current traffic conditions, the main controller selects a target control strategy that matches the current traffic conditions from the pre-stored straight-through control strategy, curve-through control strategy, and narrow-gap creep control strategy. S3. The main controller generates joint angle timing instructions for each drive joint according to the target walking control strategy, outputs PWM control signals to the servo motors of each drive joint, controls each drive joint to perform left and right yaw and / or up and down pitching actions according to the preset timing, and drives the robot to walk autonomously along the inspection channel. The specific steps of step S3 are as follows: S31. The main controller identifies the target control strategy; When the straight-pipe straight-through control strategy is adopted, proceed to step S32; When the curve passage control strategy is adopted, proceed to step S33; When the narrow slit creep control strategy is used, proceed to step S34; S32. The main controller controls each drive joint to swing in a sinusoidal phase, so that the robot body forms a sinusoidal phase wave from front to back; The joint angle of the i-th driving joint at time t Calculate using the following formula:

[0025] in, The amplitude of the swing. The angular frequency of the swing. The phase difference between adjacent drive joints. This refers to the serial number of the driven joint. , To drive the total number of joints; The main controller calculates the angles of each joint. PWM control signals are output to the servo motors of the corresponding drive joints to control each drive joint to swing in sequence, thus driving the robot forward. S33. The main controller calculates the target deflection angle of each drive joint based on the channel curvature in the environmental perception data, and controls the robot body to deflect sequentially from front to back to fit the inner wall of the curve. The target deflection angle of the i-th driving joint in the curve Calculate using the following formula:

[0026] in, For the maximum deflection angle, Here, is the bending adjustment factor, and i is the sequence number of the driven joint. , To drive the total number of joints; The main controller calculates the target deflection angle of each drive joint. PWM control signals are output to the servo motors of the corresponding drive joints to control each drive joint to deflect in sequence, so that the robot body can smoothly pass through the curve by conforming to the inner wall of the curve. S34. The main controller controls each drive joint to perform alternating extension and retraction movements, dividing each drive joint into a front end group and a rear end group. The front end group is controlled to extend forward and position itself first, and the rear end group then retracts and follows forward, alternating in a cycle. Extension of the front end drive joint and the amount of contraction of the rear end group drive joint Calculate according to the following formulas respectively:

[0027]

[0028] in, For maximum extension stroke, For the scaling angular frequency, For time; The main controller calculates the extension amount. and shrinkage PWM control signals are output to the servo motors of the corresponding drive joints to control the robot to move forward in a peristaltic manner; S4. Acquire the robot's real-time position information during the robot's autonomous walking process; The specific steps of step S4 are as follows: S41. The positioning module determines whether there are satellite signals in the area where the robot is currently located; When the robot is located in the satellite signal coverage area, proceed to step S42; When the robot is in an area without satellite signal, proceed to step S43; S42. Receive satellite signals through the Beidou and GPS dual-mode satellite positioning unit, calculate the robot's geographic coordinates as real-time location information, and output the real-time location information to the main controller; S43. Activate the inertial navigation trajectory estimation and positioning unit, and calculate the robot's real-time position using the following trajectory estimation algorithm: Let the robot's initial position coordinates be... The position coordinates at time t Calculate according to the following recursive formula:

[0029]

[0030] in, The sampling time interval, This represents the increment of walking mileage collected by the walking mileage counting module during the kth sampling period. The heading angle of the robot collected by the attitude gyroscope during the kth sampling period; The main controller calculates the real-time position coordinates. Update the robot's real-time location information; In this embodiment, when the trajectory estimation and positioning unit executes the trajectory estimation algorithm, it uses a sliding window filter to apply the input travel mileage increment. and heading angle Preprocessing is performed; specifically, a sliding window of length 5 is set inside the trajectory estimation and positioning unit to process five consecutive sampling periods. and Perform mean filtering separately:

[0031] in, This represents the filtered increment of travel distance. The filtered heading angle; after using sliding window mean filtering, the mileage counting error generated by the walking mileage counting module when walking on uneven ground, as well as the zero drift and noise in the heading angle obtained by integrating the angular velocity data output by the attitude gyroscope, can be effectively suppressed, thereby improving the positioning accuracy of the trajectory calculation; the filtered data is then substituted into the position recursion formula for dead reckoning. When the robot returns from an area with no satellite signal to an area with satellite signal coverage, the main controller immediately obtains the current precise geographical coordinates through the Beidou and GPS dual-mode satellite positioning unit. Using these geographical coordinates as a reference, it performs a single calibration correction on the accumulated position error of the trajectory estimation positioning unit, resets the estimation starting position of the trajectory estimation positioning unit to the current satellite positioning coordinates, and sets the accumulated position error of the trajectory estimation to zero. This eliminates the accumulation of errors caused by sensor drift during the trajectory estimation process, providing a precise starting position reference for the next entry into an area with no signal. S5. Transmit environmental perception data and real-time location information to the outside world through the communication module; The specific steps of step S5 are as follows: S51. The main controller packages the environmental perception data, real-time position information, and robot operating status data to generate uplink data transmission frames; S52. The main controller selects the unit with the higher signal strength from the 5G wireless transmission unit and the WiFi wireless transmission unit as the current transmission channel based on the current communication signal strength, or sends the uplink data transmission frame outward through both the 5G wireless transmission unit and the WiFi wireless transmission unit at the same time. S53. The monitoring backend receives and parses the uplink data transmission frames, and synchronously displays the video images of the inspection channel, equipment temperature data, distance to obstacles in front, real-time position and running status of the robot on the display interface, and automatically stores and generates inspection reports. For example, the monitoring backend is a computer terminal with inspection management software installed, which also supports remote access from tablets and smartphones; the uplink data transmission frame contains the following structured data fields: data frame header (e.g., 4 bytes, fixed identifier 0xAA55AA55), timestamp (e.g., 8 bytes, accurate to milliseconds), robot ID (e.g., 2 bytes), real-time position coordinates (e.g., 16 bytes, longitude / latitude or XY coordinates, double-precision floating-point number), video keyframe (e.g., variable length, JPEG compression format), infrared temperature value (e.g., 2 bytes × N measurement points, unit 0.1℃), obstacle distance (e.g., 2 bytes, unit cm), ambient temperature and humidity (e.g., 4 bytes), harmful gas concentration (e.g., 2 bytes), robot operating status (e.g., 1 byte, 0x00 indicates normal operation / 0x01 indicates fault / 0x02 indicates alarm / 0x03 indicates standby), and check frame tail (e.g., 2 bytes, CRC16 check). After receiving the uplink data transmission frame, the monitoring backend first performs a CRC check to confirm data integrity. After the check passes, the data is unpacked according to the above data format and displayed synchronously in different areas on the display interface: the central area plays real-time video footage of the inspection channel, with the robot's position coordinates at the current moment superimposed on the video footage; the left information panel updates the infrared temperature curve, obstacle distance, and environmental parameter values ​​in real time; the right side displays the robot's real-time walking trajectory in the pipe gallery in the form of a two-dimensional map or a planar layout diagram; all data is automatically archived and stored in the backend database, and after the inspection task is completed, the system automatically generates an inspection report containing anomaly point markers, temperature rise curves, and position trajectories according to a preset template.

[0032] In this embodiment, as the robot autonomously walks along the inspection channel, the laser rangefinder obstacle avoidance sensor continuously collects distance data of obstacles ahead at a sampling frequency of 50Hz; the main controller has three preset safe distance thresholds: warning distance 50cm, safe distance 30cm, and emergency braking distance 10cm.

[0033] When an obstacle is detected ahead that is less than the warning distance of 50cm but greater than the safe distance of 30cm, the main controller 40 generates a deceleration control command, which reduces the sinusoidal swing amplitude A of each drive joint from the default value of 30° to 20° and the swing angular frequency ω from 2Hz to 1.5Hz, so that the robot decelerates and approaches the obstacle while maintaining a straight-line direction of travel.

[0034] When an obstacle is detected to be less than 30cm from the safe distance but greater than 10cm from the emergency braking distance, the main controller generates a detour control command. The main controller analyzes the distance data trends over multiple consecutive sampling periods, combined with the scanning data from the laser rangefinder obstacle avoidance sensor 14 in the horizontal and vertical directions, to determine the spatial distribution characteristics of the obstacle. If the obstacle is determined to be a small, localized obstacle (such as fallen stones or construction debris), and there is a passable gap of at least 15cm on both sides of the passage, the main controller re-plans the timing sequence of the joint angles of each drive joint: first, it controls the first three sets of drive joints to deflect 15° to the passable side; then, the four sets of joints in the middle follow in sequence; and finally, the tail joint deflects to complete the overall offset, causing the robot to form an arc-shaped detour trajectory, maintaining continuous movement while avoiding obstacles.

[0035] When an obstacle ahead is detected to be less than 10cm from the emergency braking distance, and the distance measured by the laser rangefinder obstacle avoidance sensor is less than 20cm for 3 consecutive seconds, the main controller determines that the path ahead is either at the end or completely blocked. At this point, the main controller generates an emergency braking command, outputting a zero-speed lock signal to the servo motors of each drive joint, causing the robot to immediately stop moving forward. Subsequently, the main controller executes the turnaround control logic: first, it controls all drive joints of the flexible drive trunk in the middle section to deflect sequentially, causing the robot's body to form a U-shape. Specifically, it deflects joint by joint from head to tail, with the front four joints deflecting to the right to their limit angle (e.g., ±60°), while the rear four joints maintain their original direction and follow forward. In a confined space, through body deformation and controllable contact with the path wall, it completes the in-situ turning maneuver. After the turnaround is complete, the main controller controls the robot to return along the original path until it returns to the starting point or receives the next instruction from maintenance personnel. If an obstacle is encountered during the turnaround, the remote wireless emergency stop receiver module is ready to receive remote emergency stop commands, ensuring that operators can proactively intervene in control in emergency situations.

[0036] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0037] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A snake-shaped inspection robot for confined spaces, characterized in that, It includes a head probe compartment (10), a mid-section flexible drive torso, a tail energy and communication compartment (30), and a main controller (40). The head detection cabin (10) is located at the very front of the robot and is used to collect environmental perception data; The mid-section flexible drive torso is located at the rear end of the head probe cabin. The mid-section flexible drive torso includes N sets of drive joints (20), which are connected in series to form N sets of drive joints (20), where N≥2. Each drive joint (20) is equipped with a servo motor to drive the corresponding drive joint (20) to perform left and right yaw and / or up and down pitch movements; The flexible drive trunk in the middle section drives the robot to walk autonomously along the inspection channel through the coordinated swing of each set of drive joints (20); The tail energy communication compartment (30) is located at the rear end of the mid-section flexible drive body. The tail energy communication compartment (30) is equipped with a power supply module (31), a positioning module (32) and a communication module (33). The power supply module (31) supplies power to the head probe compartment (10), the mid-section flexible drive body and the tail energy communication compartment (30) through the power supply line that runs through the interior of the mid-section flexible drive body; The positioning module (32) is used to obtain the robot's real-time position information; The communication module (33) is used to transmit environmental perception data and real-time location information to the outside. The main controller (40) is connected to the head probe compartment (10), the mid-section flexible drive torso, and the tail energy and communication compartment (30), and is configured as follows: The robot receives environmental perception data and real-time location information, and outputs control signals to the servo motors of each drive joint (20) according to the environmental perception data, so as to control each drive joint (20) to coordinate actions according to a preset timing sequence, so as to realize the robot's autonomous walking and obstacle avoidance in a narrow space, and mark the fault location according to the real-time location information.

2. The snake-shaped confined space inspection robot according to claim 1, characterized in that, The head detection cabin (10) has a cone-shaped or flat shell structure with a small front end and a large rear end. The head detection cabin (10) has a front end face, on which a high-definition night vision camera module (12), an infrared non-contact temperature detection probe (13), a laser rangefinder obstacle avoidance sensor (14), a ring LED supplementary light (15), and a flexible anti-collision buffer shield (11) are installed. The high-definition night vision camera module (12) is embedded in the center of the front end face. The lens of the high-definition night vision camera module (12) faces forward, and the end face of the lens is flush with or concave to the front end face. An infrared non-contact temperature detection probe (13) is embedded in the front end face and located on the left or right side of the high-definition night vision camera module (12). The detection surface of the infrared non-contact temperature detection probe (13) faces forward and is flush with or recessed into the front end face. The laser rangefinder obstacle avoidance sensor (14) is embedded in the front end face and is located above or below the high-definition night vision camera module (12). The emitting surface and receiving surface of the laser rangefinder obstacle avoidance sensor (14) are both facing forward, and the emitting surface and receiving surface are flush with or recessed into the front end face. The ring-shaped LED fill light (15) is embedded in the outer periphery of the front end face, surrounding the high-definition night vision camera module (12), the infrared non-contact temperature detection probe (13) and the laser rangefinding obstacle avoidance sensor (14), and the light-emitting surface of the ring-shaped LED fill light (15) faces forward. The flexible anti-collision buffer shield (11) is set on the outer periphery of the front end face and protrudes forward. The flexible anti-collision buffer shield (11) surrounds the periphery of the ring LED supplement light (15), and the front end of the flexible anti-collision buffer shield (11) is located in front of the lens end face of the high-definition night vision camera module (12), the detection surface of the infrared non-contact temperature detection probe (13), and the emitting surface of the laser rangefinding obstacle avoidance sensor (14). The flexible anti-collision buffer shield (11) is used to absorb impact energy when a collision occurs. The high-definition night vision camera module (12), infrared non-contact temperature detection probe (13), laser rangefinder obstacle avoidance sensor (14), and ring LED fill light (15) are all connected to the main controller (40).

3. The snake-shaped confined space inspection robot according to claim 2, characterized in that, The head probe compartment (10) is also equipped with side walls; The sidewall is embedded with at least one of a temperature and humidity sensor (16), a harmful gas detection sensor (17), and a water level detection sensor (18); An environmental parameter acquisition module (19) is also integrated inside the head probe cabin (10); The environmental parameter acquisition module (19) is connected to the temperature and humidity sensor (16), the harmful gas detection sensor (17), and the water level detection sensor (18), and is used to receive and adjust the environmental parameter signals acquired by each sensor. The environmental parameter acquisition module (19) is connected to the main controller (40) and is used to output the conditioned environmental parameter signal to the main controller (40).

4. The snake-shaped confined space inspection robot according to claim 1, characterized in that, Each set of drive joints (20) of the mid-section flexible drive trunk is covered with a flame-retardant, insulating, and waterproof flexible shell. Flame-retardant, insulating, and waterproof flexible shells are sequentially fitted onto the outer periphery of each set of drive joints along the length of the middle flexible drive body. A gap or flexible connecting section is provided between two adjacent sets of flame-retardant, insulating, and waterproof flexible shells to provide bending freedom. The mid-section flexible drive body has an integrated wiring compartment inside; The integrated wiring compartment runs through each set of drive joints (20) along the length of the flexible drive body in the middle section. Power supply lines, signal lines and control lines are arranged in the integrated wiring compartment. The power supply lines, signal lines and control lines are provided with flexible and fold-resistant segments at the corresponding positions of the two adjacent sets of drive joints (20). The flexible drive torso in the middle section is also equipped with an attitude gyroscope (21), an tilt sensor (22), and a walking distance counting module (23). The attitude gyroscope (21), tilt sensor (22), walking distance counting module (23) are all connected to the main controller (40); The attitude gyroscope (21) is used to collect the robot's heading angle data and output it to the main controller (40); The tilt sensor (22) is used to collect the tilt data of the robot and output it to the main controller (40). The walking distance counting module (23) is used to collect the robot's walking distance increment data and output it to the main controller (40). The main controller (40) corrects the robot's walking posture in real time based on heading angle data, tilt angle data and walking distance increment data, and calculates the robot's real-time position in areas without satellite signals by combining the trajectory estimation algorithm.

5. The snake-shaped confined space inspection robot according to claim 1, characterized in that, The tail-end energy and communication compartment (30) has an outer wall; A communication antenna (34) is provided on the top of the outer wall surface. The communication antenna (34) is connected to the communication module (33) to enhance the transmission and reception of wireless signals. The outer wall of the tail energy communication compartment (30) is also equipped with a working status indicator (35), which is connected to the main controller (40) and is used to indicate the robot's running status, fault status or standby status; The positioning module (32), communication module (33) and main controller (40) are all located inside the tail energy communication compartment (30); The power supply module (31) includes a low-voltage lithium battery pack and a power supply regulator module. The output of the low-voltage lithium battery pack is connected to the head probe cabin (10), the mid-section flexible drive body and the tail energy communication cabin (30) after being regulated by the power supply module, and is used to provide low-voltage DC safe power supply. The positioning module (32) includes a Beidou and GPS dual-mode satellite positioning unit (321) and an inertial navigation trajectory estimation positioning unit (322). The Beidou and GPS dual-mode satellite positioning unit (321) is used to obtain the robot's geographical coordinates in the satellite signal coverage area. The inertial navigation trajectory estimation positioning unit (322) is used to calculate the robot's real-time position in the area without satellite signal based on the walking distance, deflection angle and travel time. The communication module (33) includes a 5G wireless transmission unit (331) and a WiFi wireless transmission unit (332). Both the 5G wireless transmission unit (331) and the WiFi wireless transmission unit (332) are connected to the communication antenna (34) and are used to transmit environmental perception data and real-time location information to the outside. The tail-end energy communication compartment (30) is also equipped with a remote wireless emergency stop receiver module (36). The remote wireless emergency stop receiver module (36) is connected to the main controller (40) and is used to receive remote emergency stop commands and output emergency stop trigger signals to the main controller (40).

6. An inspection method based on the snake-shaped confined space inspection robot according to any one of claims 1-5, characterized in that, Includes the following steps: S1. Collect environmental perception data through the head-mounted detection cabin; S2. The main controller receives environmental perception data and identifies the current passage conditions of the inspection channel based on the environmental perception data. It then selects a target walking control strategy that matches the current passage conditions from a variety of pre-stored bionic walking control strategies. S3. The main controller generates joint angle timing instructions for each drive joint according to the target walking control strategy, outputs PWM control signals to the servo motors of each drive joint, controls each drive joint to perform left and right yaw and / or up and down pitching actions according to the preset timing, and drives the robot to walk autonomously along the inspection channel. S4. Acquire the robot's real-time position information during the robot's autonomous walking process; S5. Transmit environmental perception data and real-time location information to the outside world through the communication module.

7. The inspection method according to claim 6, characterized in that, The specific steps of step S1 are as follows: S11. The laser rangefinder and obstacle avoidance sensor embedded in the front end of the head detection cabin emits a laser beam forward and receives the reflected echo signal to obtain the distance data of the obstacle in front. S12. High-definition night vision camera module embedded in the front surface of the head detection cabin collects video image data in front of the inspection channel; S13. The head detection cabin transmits the distance data of obstacles in front and video image data as environmental perception data to the main controller; The specific steps of step S2 are as follows: S21. The main controller extracts distance data of obstacles ahead and / or video image data from the environmental perception data; S22. The main controller determines the channel width, channel curvature and / or channel cross-sectional dimensions of the current inspection channel based on the distance data of the obstacle ahead and / or video image data, and identifies the current passage condition as straight pipe straight passage condition, curved passage condition or narrow gap creep passage condition; S23. Based on the identified current traffic conditions, the main controller selects a target control strategy that matches the current traffic conditions from the pre-stored straight-through control strategy, curve-through control strategy, and narrow-gap creep control strategy.

8. The inspection method according to claim 7, characterized in that, The specific steps of step S3 are as follows: S31. The main controller identifies the target control strategy; When the straight-pipe straight-through control strategy is adopted, proceed to step S32; When the curve passage control strategy is adopted, proceed to step S33; When the narrow slit creep control strategy is used, proceed to step S34; S32. The main controller controls each drive joint to swing in a sinusoidal phase, so that the robot body forms a sinusoidal phase wave from front to back; The joint angle of the i-th driving joint at time t Calculate using the following formula: in, The amplitude of the swing. The angular frequency of the swing. The phase difference between adjacent drive joints. This refers to the serial number of the driven joint. , To drive the total number of joints; The main controller calculates the angles of each joint. PWM control signals are output to the servo motors of the corresponding drive joints to control each drive joint to swing in sequence, thus driving the robot forward. S33. The main controller calculates the target deflection angle of each drive joint based on the channel curvature in the environmental perception data, and controls the robot body to deflect sequentially from front to back to fit the inner wall of the curve. The target deflection angle of the i-th driving joint in the curve Calculate using the following formula: in, For the maximum deflection angle, Here, is the bending adjustment factor, and i is the sequence number of the driven joint. , To drive the total number of joints; The main controller calculates the target deflection angle of each drive joint. PWM control signals are output to the servo motors of the corresponding drive joints to control each drive joint to deflect in sequence, so that the robot body can smoothly pass through the curve by conforming to the inner wall of the curve. S34. The main controller controls each drive joint to perform alternating extension and retraction movements, dividing each drive joint into a front end group and a rear end group. The front end group is controlled to extend forward and position itself first, and the rear end group then retracts and follows forward, alternating in a cycle. Extension of the front end drive joint and the amount of contraction of the rear end group drive joint Calculate according to the following formulas respectively: in, For maximum extension stroke, For the scaling angular frequency, For time; The main controller calculates the extension amount. and shrinkage PWM control signals are output to the servo motors of the corresponding drive joints to control the robot to move forward in a creeping manner.

9. The inspection method according to claim 8, characterized in that, The specific steps of step S4 are as follows: S41. The positioning module determines whether there are satellite signals in the area where the robot is currently located; When the robot is located in an area covered by satellite signals, proceed to step S42; When the robot is in an area without satellite signal, proceed to step S43; S42. Receive satellite signals through the Beidou and GPS dual-mode satellite positioning unit, calculate the robot's geographic coordinates as real-time location information, and output the real-time location information to the main controller; S43. Activate the inertial navigation trajectory estimation and positioning unit, and calculate the robot's real-time position using the following trajectory estimation algorithm: Let the robot's initial position coordinates be... The position coordinates at time t Calculate according to the following recursive formula: in, The sampling time interval, This represents the increment of walking mileage collected by the walking mileage counting module during the kth sampling period. The heading angle of the robot collected by the attitude gyroscope during the kth sampling period; The main controller calculates the real-time position coordinates. Update the robot's real-time location information.

10. The inspection method according to claim 9, characterized in that, The specific steps of step S5 are as follows: S51. The main controller packages the environmental perception data, real-time position information, and robot operating status data to generate uplink data transmission frames; S52. The main controller selects the unit with the higher signal strength from the 5G wireless transmission unit and the WiFi wireless transmission unit as the current transmission channel based on the current communication signal strength, or sends the uplink data transmission frame outward through both the 5G wireless transmission unit and the WiFi wireless transmission unit at the same time. S53. The monitoring backend receives and parses the uplink data transmission frames, and synchronously displays the video images of the inspection channel, equipment temperature data, distance to obstacles ahead, real-time position and running status of the robot on the display interface, and automatically stores and generates inspection reports.