Multi-rotor wall-climbing robot based on path autonomous planning

Abstract

The invention discloses a multi-rotor wall-climbing robot based on path autonomous planning, belongs to the technical field of wall-climbing robots, and aims to improve the structure of a traditional multi-rotor wall-climbing robot and construct a mixed configuration of a double-degree-of-freedom rotor assembly and an adjustable wall-climbing wheel set. Wall-climbing postures corresponding to different wall surfaces are obtained through angle adjustment of multiple wall-climbing wheels, a conventional single rotating rotor wing structure is arranged to be a double-degree-of-freedom rotor wing assembly, in the wall-climbing posture switching process, continuous and smooth conversion of thrust vectors is achieved to stably adjust pitch angle control, and the control precision of the wall-climbing postures is improved. Seamless conversion of thrust from resisting gravity to providing adherence pressure is achieved, the problem of dynamic instability of a traditional rotor wing structure overall single rotation scheme in the transition stage is effectively solved, remarkable advantages are brought to path planning based on the flight and wall climbing dual-mode switching characteristic, and through the obstacle perception fusion and graded obstacle avoidance strategy, the path planning precision is improved. Real-time path dynamic optimization is realized, and the operation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wall-climbing robots, and more particularly to a multi-rotor wall-climbing robot based on autonomous path planning. Background Art

[0002] Due to the low efficiency, low safety factor and high cost of manual high-altitude operations, wall-climbing robots are increasingly used in inspection, flaw detection, reconnaissance and other aspects. However, traditional wall-climbing robots have the limitation of single-form movement, slow climbing speed, difficulty in hovering and positioning, and cannot cross obstacles for barrier-free movement.

[0003] Due to the technical limitations of traditional wall-climbing robots, multi-rotor wall-climbing robots are gradually entering the market and playing an important role in many fields such as cleaning and monitoring. For example, the content disclosed in patent number CN114084346B combines the rotor flight capabilities of drones with the wall-climbing functions of ordinary wall-climbing robots to achieve efficient drone inspections, wall-mounted inspections, and non-destructive testing.

[0004] However, the fixed wheeled mobile structure relies solely on propeller downforce to provide wall-adhering force, making it unable to adapt to different wall surfaces. Furthermore, the entire flight assembly is rotated freely to adjust the pitch angle. The originally horizontal rotors at the four corners change their thrust direction during overall rotation, potentially causing lateral drift. Furthermore, path planning for multi-rotor wall-climbing robots is more complex than for conventional drones or wall-climbing robots.

[0005] However, how to efficiently cover the detection area, avoid obstacles, and quickly switch the control strategy when switching motion modes to achieve autonomous path planning is a difficult problem. For this reason, a multi-rotor wall-climbing robot based on autonomous path planning is proposed based on actual situations. Summary of the Invention

[0006] The purpose of the present invention is to solve the existing practical technical problems and provide a multi-rotor wall-climbing robot based on autonomous path planning compared with the existing technology.

[0007] The object of the present invention can be achieved by the following technical solutions: a multi-rotor wall-climbing robot based on autonomous path planning comprises a wall-climbing frame, wherein the upper and lower ends of the wall-climbing frame are respectively equipped with an adjustable wall-climbing wheel assembly and a dual-degree-of-freedom rotor assembly, wherein the adjustable wall-climbing wheel assembly comprises a U-shaped frame embedded in the bottom of the wall-climbing frame, wherein the middle bottom end of the U-shaped frame is equipped with two main wall-climbing wheels distributed left and right and adapted to the angles thereof, and a pair of main wall-climbing wheels are symmetrically distributed in the front and rear directions and have two groups of slave wall-climbing wheels, and the two groups of slave wall-climbing wheels are linked and installed through a gear rotation mechanism;

[0008] The double-degree-of-freedom rotor assembly comprises two groups of rotating arms rotatably mounted in the front-rear direction of the wall-climbing frame and rotatably driven by a synchronous adjusting mechanism, each group of rotating arms is provided with two rotating arms distributed left and right, and a flight frame is rotatably driven and mounted between the top ends of the two rotating arms by a rotating motor, and a pair of helical propellers distributed left and right are mounted on the flight frame;

[0009] The wall-climbing frame is provided with an image acquisition camera and a wall recognition sensor embedded and mounted on the top wall in the advancing direction of the wall-climbing frame, and a control panel connected in signal with the image acquisition camera and the wall recognition sensor and autonomously planning a path for linkage control of the adjustable wall-climbing wheel set and the double-degree-of-freedom rotor assembly is further mounted in the wall-climbing frame.

[0010] Further, a fixed frame for rotatably mounting the main wall-climbing wheels is fixedly mounted at the middle of the bottom end of the U-shaped frame, a transmission chain structure for synchronously rotatably driving the front and rear main wall-climbing wheels is mounted in the fixed frame, and drive motors for rotatably driving the pair of main wall-climbing wheels are provided on the outside of the fixed frame.

[0011] Further, the gear rotating mechanism comprises a pair of supporting frames rotatably mounted for the slave wall-climbing wheels, one end of the pair of supporting frames away from the slave wall-climbing wheels is rotatably mounted on the inner wall of the U-shaped frame in an upwardly inclined manner, the top ends of the pair of supporting frames are rotatably connected by a transmission meshing gear set and a transmission motor, the pair of supporting frames drive the slave wall-climbing wheels to rotate in the same direction or in opposite directions, the pair of main wall-climbing wheels are centrally arranged, providing core driving force, and in combination with the adjustable slave wall-climbing wheels on the two sides, the wall-climbing posture for coping with different forms of wall surfaces, such as curved surfaces and concave-convex wall surfaces, can be adjusted by adjusting the wall-climbing angle between the slave wall-climbing wheels and the main wall-climbing wheels, so that flexible movement on different wall surfaces can be achieved.

[0012] Further, the synchronous adjusting mechanism comprises a pair of pitch shafts fixedly mounted between the lower end walls of the pair of rotating arms distributed left and right, the outer ends of the pair of pitch shafts penetrate through the outer end walls of the wall-climbing frame, drive boxes are fixedly connected to the outer end walls of the wall-climbing frame, a gear and a rack meshing transmission structure acting between the end portions of the two pitch shafts is mounted in the drive boxes, and a rotating motor two rotatably driving one of the gears is fixedly mounted on the outer end of the drive box, so that the two groups of rotating arms are synchronously rotated in the same direction or in opposite directions around the respective pitch shafts, and the two groups of rotating arms are always in a parallel state.

[0013] Further, the control panel is internally provided with a wall surface information acquisition module, a decision engine module, a path planning module, an obstacle perception module and an obstacle avoidance control module;

[0014] The wall surface information acquisition module is used to acquire wall surface information and task information of a target area, the wall surface information includes wall surface type and wall surface curvature radius, and the wall surface type includes flat wall surface and curved wall surface, wherein the curved wall surface is divided into concave curved wall surface and convex curved wall surface;

[0015] The wall surface information is sent to a decision engine module, the wall surface type and task information are sent to a path planning module, the path planning module performs main path planning according to the wall surface information and the task information, an ascending path is generated and sent to the decision engine module, the decision engine module adjusts the wall climbing posture of the adjustable wall climbing wheel set according to the wall surface information, before the wall climbing posture is adjusted, the propeller wings are driven to fly away from the ground to hover at the edge of the target area, after the wall climbing posture is adjusted, the wall climbing mode is switched to work along the ascending path close to the wall surface, and the obstacle perception module is used to acquire obstacle information during the work along the ascending path close to the wall surface, the generated danger level signal is sent to an obstacle avoidance control module for path optimization according to evaluation and analysis of the obstacle information.

[0016] Further, the process that the path planning module performs main path planning according to the wall surface type and the task information includes: acquiring a target area and a highest point detection point coordinate of the target area, and generating a straight ascending path or a spiral ascending path according to the wall surface type.

[0017] Further, the decision engine module receives the wall surface information, when it is identified that the wall surface is a flat wall surface, the two groups of slave wall climbing wheels are adjusted so that the moving surfaces thereof are on the same plane as the moving surface of the main wall climbing wheel, when it is identified that the wall surface is a concave curved wall surface, the two groups of slave wall climbing wheels are reversely adjusted so that a wall climbing included angle between the main wall climbing wheel and the two groups of slave wall climbing wheels is adapted to the curvature radius of the concave curved wall surface, and the three-wheel tangent points share the concave curved surface, and when it is identified that the wall surface is a convex curved wall surface, the two groups of slave wall climbing wheels are forwardly adjusted so that the wall climbing included angle between the main wall climbing wheel and the two groups of slave wall climbing wheels is adapted to the curvature radius of the convex curved wall surface, and the three-wheel tangent points share the convex curved surface.

[0018] After the wall climbing posture is adjusted, the decision engine module acquires the corresponding ascending path, adjusts the flight postures of the two groups of propeller wings, quickly aligns the wall surface normal direction of the target area through transformation of the thrust vector of the propeller wings, and makes the main wall climbing wheel and the slave wall climbing wheels of the standard wall climbing posture keep the best contact angle with the wall surface.

[0019] Further, the obstacle perception module acquires the obstacle information through an image acquisition camera, including the size, shape and detection distance of the obstacle, compares the obstacle information with a historical collision risk database, and generates a first-level danger level control signal, a second-level danger level control signal and a third-level danger level control signal.

[0020] Further, when the third-level danger level control signal is received, the obstacle avoidance control module performs the original wall climbing action along the ascending path, when the second-level danger level control signal is received, the obstacle avoidance control module performs a wall climbing bypass action, and when the first-level danger level control signal is received, the obstacle avoidance control module performs a flight leapfrog action.

[0021] Compared with the prior art, the advantages of the present application are that:

[0022] 1. The scheme is aimed at the traditional multi-rotor wall climbing robot for structural improvement, the hybrid configuration of "double freedom rotor assembly + adjustable wall climbing wheel set" is constructed, the wall climbing posture of coping with different wall surfaces is obtained through the angle adjustment of multiple wall climbing wheels, the conventional single rotation rotor structure is set as a double freedom rotor assembly, in the wall climbing posture switching process, the continuous smooth transformation of the thrust vector is realized to smoothly adjust the pitch angle control, the seamless conversion of the thrust from "opposing gravity" to "providing wall sticking pressure" is realized, and the dynamic instability problem of the conventional single rotation scheme in the transition stage is effectively solved.

[0023] 2. The scheme is also based on the double freedom rotor assembly, the "flight + wall climbing" dual-mode switching characteristics bring significant advantages to path planning, and through obstacle perception fusion + hierarchical obstacle avoidance strategy, real-time path dynamic optimization is realized, and the operation efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 It is the overall structure schematic diagram of the application;

[0025] Figure 2 It is the bottom view of the combination of the wall climbing frame and the adjustable wall climbing wheel set of the application;

[0026] Figure 3 It is the structure schematic diagram of the adjustable wall climbing wheel set of the application;

[0027] Figure 4 It is the structure schematic diagram of the double freedom rotor assembly of the application;

[0028] Figure 5 It is the structure schematic diagram of the double freedom rotor assembly of the application;

[0029] Figure 6 It is the state schematic diagram of the "flight + wall climbing" dual-mode switching of the application;

[0030] Figure 7 It is the state schematic diagram of the "flight + wall climbing" dual-mode switching of the application;

[0031] Figure 8 It is the system principle block diagram in the second embodiment of the application.

[0032] Explanation of reference numerals in the drawing:

[0033] 1. Wall climbing frame;

[0034] 2. Adjustable wall climbing wheel set; 21. U-shaped frame; 22. Fixed frame; 23. Main wall climbing wheel; 24. Support frame; 25. Slave wall climbing wheel;

[0035] 3, dual-degree-of-freedom rotor assembly; 31, rotating arm; 32, flight frame; 33, propeller wing; 34, rotating motor 1; 35, drive box; 36, rotating motor 2; 37, pitch shaft;

[0036] 4, image acquisition camera. DETAILED DESCRIPTION

[0037] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application; obviously, the described embodiments are only part of the embodiments of the present application, but not all the embodiments; based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0038] Embodiment one: the present application relates to a multi-rotor wall climbing robot based on path autonomous planning, please refer to Figures 1-3 , comprising a wall climbing frame 1, and an adjustable wall climbing wheel set 2 and a dual-degree-of-freedom rotor assembly 3 are respectively installed at the upper and lower ends of the wall climbing frame 1, the adjustable wall climbing wheel set 2 comprises a U-shaped frame 21 embedded and installed in the bottom of the wall climbing frame 1;

[0039] Two main wall climbing wheels 23 distributed left and right and angle-adapted with the U-shaped frame 21 are installed at the bottom end of the middle part of the U-shaped frame 21, a fixed frame 22 for the rotational installation of the main wall climbing wheels 23 is fixedly installed at the bottom end of the middle part of the U-shaped frame 21, a transmission chain structure for the synchronous rotational driving of the front and rear main wall climbing wheels 23 is installed inside the fixed frame 22, and a driving motor for the rotational driving of the pair of main wall climbing wheels 23 is arranged on the outside of the fixed frame 22;

[0040] Two groups of slave wall climbing wheels 25 are symmetrically distributed in the front and rear directions of the pair of main wall climbing wheels 23, the two groups of slave wall climbing wheels 25 are linkage-installed through a gear rotating mechanism, the gear rotating mechanism comprises a pair of supporting frames 24 symmetrically distributed in front and back and used for the rotational installation of the slave wall climbing wheels 25, one end of the pair of supporting frames 24 away from the slave wall climbing wheels 25 is rotationally installed on the inner wall of the U-shaped frame 21 in an upward and inclined manner, and the top ends of the pair of supporting frames 24 are linkage-installed through a transmission meshing gear set and a transmission motor, so that the pair of supporting frames 24 drive the slave wall climbing wheels 25 to rotate in the same direction or in the opposite direction;

[0041] The pair of main wall climbing wheels 23 are arranged in the middle, provide core driving force, and are combined with the adjustable slave wall climbing wheels 25 on the two sides, so that the wall-clamping angle between the slave wall climbing wheels 25 and the main wall climbing wheels 23 is adjusted, so that the wall climbing posture for coping with different forms of wall surfaces, such as curved surfaces and concave-convex wall surfaces, can be obtained, and flexible movement on different wall surfaces can be adapted.

[0042] Please refer to Figure 1 and Figure 4, the double-degree-of-freedom rotor assembly 3 comprises two groups of rotating arms 31 rotatably mounted in the front-rear direction of the wall climbing frame 1 and rotatably driven by a synchronous adjusting mechanism, each group of rotating arms 31 is provided with two rotating arms 31 distributed left and right, and the top ends of the two rotating arms 31 are rotatably driven and mounted with a flight frame 32 by a rotating motor 34, and the flight frame 32 is mounted with a pair of propellers 33 distributed left and right;

[0043] The synchronous adjusting mechanism comprises a pitch shaft 37 fixedly mounted between the lower end walls of the pair of rotating arms 31 distributed left and right, the outer ends of the pair of pitch shafts 37 penetrate to the outer end wall of the wall climbing frame 1, and the outer end wall of the wall climbing frame 1 is fixedly connected with a drive box 35, the inside of the drive box 35 is mounted with a gear and rack meshing transmission structure acting between the end portions of the two pitch shafts 37, and the outer end of the drive box 35 is fixedly mounted with a rotating motor 36 for rotatably driving one of the gears;

[0044] Please refer to Figures 5-7 , the two groups of rotating arms 31 are realized to be synchronously rotated forward or reversely around the respective pitch shafts 37, so that the two groups of rotating arms 31 are always in a parallel state, the synchronous adjusting mechanism realizes the same rotation of the two groups of propellers 33, and is responsible for accurately aligning the target pitch angle, and the propeller 33 self-rotation structure finely adjusts the thrust vector direction to offset the interference moment.

[0045] The embodiment improves the traditional single fixed wheel moving structure of the wall climbing robot, and sets an adjustable wall climbing wheel group with three wheels in a plane, and adjusts the angles of the three wheels to obtain a wall climbing posture suitable for different wall surfaces, which is not only suitable for flat wall surfaces, but also suitable for curved wall surfaces;

[0046] And for the traditional rotor wall climbing robot, the conventional single rotating rotor structure is set as a double-degree-of-freedom rotor assembly, in the wall climbing process, the continuous smooth transformation of the thrust vector is realized by the flexible adjustment of the double-degree-of-freedom rotor, so as to stably adjust the pitch angle control, through the dynamic matching of the wheel group angle and the rotor thrust, the three-point contact in a circle is realized, which is a substantial breakthrough, realizes the seamless conversion of the thrust from “opposing gravity” to “providing wall sticking pressure”, and effectively solves the problem of dynamic instability in the transition stage of the traditional single rotating scheme of the rotor structure as a whole.

[0047] Embodiment two: please refer to Figure 1 , the image acquisition camera 4 and the wall surface recognition sensor are embedded and mounted on the top wall of the wall climbing frame 1 in the advancing direction, and the control panel connected with the image acquisition camera 4 and the wall surface recognition sensor and automatically planning the path of the adjustable wall climbing wheel group 2 and the double-degree-of-freedom rotor assembly 3 is also mounted in the wall climbing frame 1, and the wall surface recognition sensor can be a line laser profiler.

[0048] Please refer to Figure 8The control panel is internally provided with a wall information acquisition module, a decision engine module, a path planning module, an obstacle perception module and an obstacle avoidance control module.

[0049] The wall information acquisition module is configured to acquire wall information and task information of the target area, the wall information including wall type and wall curvature radius, the wall type including flat wall and curved wall, the curved wall including concave curved wall and convex curved wall, and the wall information being sent to the decision engine module and the wall type and the task information being sent to the path planning module.

[0050] The path planning module is configured to plan a main path according to the wall type and the task information, generate an ascending path and send the ascending path to the decision engine module, adjust a wall climbing posture of the adjustable wall climbing wheel set 2 according to the wall information, fly to the edge of the target area and hover above the edge before the wall climbing posture is adjusted, and switch to a wall climbing mode to work along the ascending path close to the wall after the wall climbing posture is adjusted.

[0051] The obstacle perception module is configured to acquire obstacle information during the working along the ascending path close to the wall, evaluate and analyze the obstacle information, send a generated danger level signal to the obstacle avoidance control module, and optimize the path.

[0052] The process of planning the main path by the path planning module according to the wall type and the task information includes: acquiring the target area and the highest point detection point coordinates of the target area, the target area and the highest point detection point coordinates of the target area being summarized as the task information, generating a straight ascending path or a spiral ascending path according to the wall type, specifically, generating the straight ascending path for the flat wall, and generating the spiral ascending path for the curved wall.

[0053] The process of adjusting the wall climbing posture of the adjustable wall climbing wheel set 2 according to the wall information includes:

[0054] The decision engine module receives the wall information, adjusts the front and rear two groups of slave wall climbing wheels 25 so that the moving surfaces of the front and rear two groups of slave wall climbing wheels 25 and the moving surface of the main wall climbing wheel 23 are on the same plane when it is identified that the wall is a flat wall, reversely adjusts the two groups of slave wall climbing wheels 25 so that a wall climbing angle between the two groups of slave wall climbing wheels 25 and the main wall climbing wheel 23 is adapted to the curvature radius of the concave curved wall, and the three-wheel tangent points are on the concave curved surface when it is identified that the wall is a concave curved wall.

[0055] When the convex curve wall surface is identified, the two groups of from wall climbing wheels 25 are adjusted to form a wall climbing angle with the main wall climbing wheel 23 that matches the curvature radius of the convex curve wall surface, so that the three-wheel tangent point is also on the convex surface, to adjust the wall climbing posture of the adjustable wall climbing wheel group 2 by changing the size of the wall climbing angle. It needs to be added here that in order to improve the three-wheel common surface of different wall surfaces, the support frame 24 can be set as a telescopic structure to obtain mechanical constraint compensation when dealing with critical values, and through wheel group angle transformation, a three-point tangent surface is realized.

[0056] After the wall climbing posture is adjusted, the decision engine module obtains the corresponding rising path, adjusts the flight posture of the two groups of propeller wings 33, and quickly aligns the wall surface normal direction of the target area through the transformation of the thrust vector of the propeller wings 33. During the gradual adjustment and alignment process, the from wall climbing wheels 25 at the front end of the wall climbing frame 1 first contact the target detection area wall surface after continuous adjustment, so that the main wall climbing wheel 23 and the from wall climbing wheel 25 of the standard wall climbing posture maintain the best contact angle with the wall surface. Especially for curved wall surfaces, the risk of detachment caused by excessive curvature of the curved surface is avoided.

[0057] The obstacle perception module generates a control signal based on obstacle information evaluation and analysis, including: obtaining obstacle information through image acquisition camera 4, including obstacle size, shape and detection distance, comparing the obstacle information with the historical collision risk database to generate a first, second and third danger level control signal.

[0058] When receiving the third danger level control signal, the obstacle avoidance control module executes the original wall climbing action along the rising path, when receiving the second danger level control signal, it executes the wall climbing action, and when receiving the first danger level control signal, it executes the flight action. Based on real-time detection of dynamic obstacles, a hierarchical obstacle avoidance mechanism is implemented for real-time path dynamic optimization.

[0059] As shown above: The traditional multi-rotor wall climbing robot is improved in structure, and a hybrid configuration of "double-freedom rotor assembly + adjustable wall climbing wheel group" is constructed. The climbing posture for different wall surfaces is obtained by adjusting the angle of multiple wall climbing wheels, and the conventional single-rotation rotor structure is set as a double-freedom rotor assembly. During the wall climbing process, the continuous and smooth transformation of the thrust vector is realized to smoothly adjust the pitch angle control. The technical core is to seamlessly combine the flight freedom and wall working advantages, effectively solve the dynamic instability problem of the traditional single-rotation scheme in the transition stage, and realize efficient and safe surface coverage detection.

[0060] Based on the double-freedom rotor assembly, the "flight + wall climbing" dual-mode switching characteristics bring significant advantages to path planning, and through obstacle perception fusion + hierarchical obstacle avoidance strategy, real-time path dynamic optimization is realized, and work efficiency is improved.

[0061] The above descriptions are merely specific embodiments of the present application, but the protection scope of the present application is not limited to this, and any technical personnel familiar with the technical field can make equivalent replacements or changes within the technical scope disclosed by the present application and according to the technical solutions and improvement concepts of the present application, which should be covered within the protection scope of the present application.

Claims

1. A multi-rotor wall-climbing robot based on autonomous path planning, comprising a wall-climbing frame (1), characterized in that: The upper and lower ends of the climbing frame (1) are respectively equipped with an adjustable climbing wheel assembly (2) and a dual-degree-of-freedom rotor assembly (3), wherein the adjustable climbing wheel assembly (2) comprises a U-shaped frame (21) embedded in the bottom of the climbing frame (1), and two main climbing wheels (23) distributed left and right and adapted to the angles thereof are installed at the bottom of the middle portion of the U-shaped frame (21), and two groups of slave climbing wheels (25) are symmetrically distributed in the front and rear directions of the pair of main climbing wheels (23), and the two groups of slave climbing wheels (25) are linked and installed through a gear rotation mechanism; The dual-degree-of-freedom rotor assembly (3) comprises two groups of rotating arms (31) rotatably mounted on the climbing frame (1) in the front-rear direction and driven by a synchronous adjustment mechanism, each group of rotating arms (31) is provided with two rotating arms distributed on the left and right, a flying frame (32) is installed between the top ends of the two rotating arms (31) and driven by a rotating motor (34), and a pair of propeller wings (33) distributed on the left and right are installed on the flying frame (32); The wall climbing frame (1) is embedded with an image acquisition camera (4) and a wall recognition sensor on the top wall in the forward direction thereof. A control panel is also installed inside the wall climbing frame (1) for connecting signals with the image acquisition camera (4) and the wall recognition sensor and for performing autonomous path planning for the linkage control of the adjustable wall climbing wheel assembly (2) and the dual-degree-of-freedom rotor assembly (3).

2. The multi-rotor wall-climbing robot based on autonomous path planning according to claim 1, characterized in that: A fixing frame (22) for rotatably mounting the main wall-climbing wheels (23) is fixedly mounted at the middle of the bottom end of the U-shaped frame (21), and a driving motor for rotatably driving a pair of main wall-climbing wheels (23) is provided on the outside of the fixing frame (22).

3. The multi-rotor wall-climbing robot based on autonomous path planning according to claim 2, characterized in that: The gear rotation mechanism comprises a pair of support frames (24) symmetrically distributed front and back and used for rotating the slave wall-climbing wheel (25). The pair of support frames (24) are rotatably mounted on the inner wall of the U-shaped frame (21) away from one end of the slave wall-climbing wheel (25). The top ends of the pair of support frames (24) are connected to each other through a transmission meshing gear set and a transmission motor.

4. The multi-rotor wall-climbing robot based on autonomous path planning according to claim 1, characterized in that: The synchronous adjustment mechanism includes a pitch shaft (37) fixedly installed between the lower end walls of a pair of rotating arms (31) distributed on the left and right, the outer ends of the pair of pitch shafts (37) pass through the outer end wall of the climbing frame (1), and a drive box (35) is fixedly connected to the outer end wall of the climbing frame (1), a gear and rack meshing transmission structure acting between the ends of the two pitch shafts (37) is installed inside the drive box (35), and a rotating motor (36) for rotating one of the gears is fixedly installed at the outer end of the drive box (35).

5. The multi-rotor wall-climbing robot based on autonomous path planning according to claim 1, characterized in that: The control panel is internally provided with a wall information acquisition module, a decision engine module, a path planning module, an obstacle perception module and an obstacle avoidance control module; The wall information acquisition module is used to obtain the wall information and task information of the target area. The wall information includes the wall type and wall curvature radius. The wall type includes flat wall and curved wall. Curved wall is divided into concave curved wall and convex curved wall. The wall information is sent to the decision engine module, and the wall type and task information are sent to the path planning module. The path planning module performs main path planning according to the wall type and task information, generates an ascending path and sends it to the decision engine module. The decision engine module adjusts the climbing posture of the adjustable wall-climbing wheel group (2) according to the wall information, and switches to the wall-climbing mode to operate close to the wall along the ascending path. The obstacle perception module is used to obtain obstacle information, evaluate and analyze the obstacle information, generate a danger level signal and send it to the obstacle avoidance control module for path optimization.

6. The multi-rotor wall-climbing robot based on autonomous path planning according to claim 5, characterized in that: The path planning module obtains the target area and the coordinates of the highest point detection point in the target area, and generates a straight ascending path or a spiral ascending path according to the wall type.

7. The multi-rotor wall-climbing robot based on autonomous path planning according to claim 6, characterized in that: After adjusting the wall climbing attitude, the decision engine module obtains the corresponding ascending path, adjusts the flight attitude of the two sets of propeller wings (33), and quickly aligns the wall normal direction of the target area through the transformation of the propeller wings (33) thrust vector until the main climbing wheel (23) and the slave climbing wheel (25) maintain an optimal contact angle with the wall.

8. The multi-rotor wall-climbing robot based on autonomous path planning according to claim 7, characterized in that: The obstacle sensing module obtains obstacle information, including obstacle size, shape, and detection distance, through an image acquisition camera (4), and generates a first-level hazard level control signal, a second-level hazard level control signal, and a third-level hazard level control signal based on the obstacle information. When the obstacle avoidance control module receives the third-level hazard level control signal, it performs a wall climbing action along the ascending path; when it receives the second-level hazard level control signal, it performs a wall climbing and bypassing action; and when it receives the first-level hazard level control signal, it performs a flying over action.

Images