Obstacle-crossing wheel type climbing robot with contraction module and control method of obstacle-crossing wheel type climbing robot

Abstract

The invention relates to an obstacle-crossing wheel type climbing robot with a contraction module and a control method of the obstacle-crossing wheel type climbing robot. The robot comprises the contraction module, a double-chain-wheel plate, a climbing module and an obstacle-crossing buffering unit. The contraction module utilizes a double-shaft motor to drive a chain to contract and release, so that the dynamic adjustment on the climbing wall surface pressure and the climbing range in the obstacle crossing process is realized; the obstacle crossing buffer unit and the double-chain-wheel plate are arranged on the climbing module, and the double-chain-wheel plate is connected with the climbing module through the obstacle crossing buffer unit; according to the robot, on the basis of a data driving control algorithm and a controlled quantity compensation algorithm, the climbing and obstacle crossing process without the overturning tendency is achieved through a large and small driving wheel structure, an obstacle crossing buffering unit is arranged between a double-chain-wheel plate and a main body plate, and a space for adjusting the posture is reserved for obstacle crossing; and the stability and rapidity of the high-load climbing robot during obstacle crossing are ensured.

Description

Technical Field

[0001] This invention belongs to the field of climbing robot technology, and relates to an obstacle-crossing wheeled climbing robot with a retractable module and its control method, and more particularly to an obstacle-crossing wheeled climbing robot with a retractable module and no tendency to tip over and its control method. Background Technology

[0002] Large, tall structures such as bridge piers and wind turbine towers are prone to surface cracks, coating peeling, or internal defects during long-term service, necessitating regular and close-range inspection and maintenance. Traditional methods relying on manual suspended platforms or scaffolding are inherently inefficient, pose high safety risks, and limit the scope of work. Therefore, the use of climbing robot technology has emerged as an important alternative. However, this technology still faces a series of significant technical challenges in practical applications, especially when dealing with the complex environments of large, tall structures.

[0003] Specifically, existing climbing robots have significantly insufficient adaptive climbing capabilities on complex surfaces. The surfaces of bridge piers and towers are not flat, often containing various uneven obstacles such as vertical or curved facades, rust protrusions, and bolt welds. Currently, most wheeled robots rely on single methods such as magnetic adsorption or negative pressure adsorption for climbing. Their rigid wheels struggle to dynamically conform to changing surface contours during movement, directly leading to slippage or adsorption failure during climbing, making it difficult to guarantee reliability.

[0004] Secondly, there are limitations in the active obstacle-crossing mechanisms and control strategies. When encountering continuous stepped obstacles such as tower flanges, or randomly distributed obstacles, existing technologies mostly employ pre-planned fixed obstacle-crossing actions. This static control strategy lacks the ability to adjust the trajectory based on real-time contact force feedback, resulting in high energy consumption and slow dynamic response during obstacle crossing, making it unsuitable for the needs of tall structural pole surfaces. Simultaneously, existing high-load climbing robots have a high center of gravity and large inertia, making them prone to instability and overturning due to sudden changes in the center of gravity during obstacle crossing. Traditional static stabilization strategies (such as low-speed movement) are insufficient for dynamic scenarios, while complex dynamic adjustment algorithms place extremely high demands on the control system, especially when the load distribution is uneven, making stability even more difficult to guarantee.

[0005] In summary, current climbing robot technology faces significant technical bottlenecks in areas such as adaptability to complex surfaces, real-time obstacle crossing, and integrated operation. To meet the urgent need for efficient, safe, and unmanned inspection and maintenance of large, tall structures, it is imperative to develop a climbing robot that combines the ability to avoid tipping over with dynamic obstacle crossing capabilities, possessing both strong dynamic obstacle crossing ability and an effective anti-tumble mechanism, in order to fundamentally solve the aforementioned problems. Summary of the Invention

[0006] In view of this, in order to solve the problems of insufficient adaptability on complex surfaces, static rigidity of obstacle-crossing strategies, and poor dynamic stability under high loads that are common in existing climbing robots, which affect their applicable scenarios, this invention provides a wheeled climbing robot with a retractable module capable of overcoming obstacles and its control method. This climbing robot achieves dynamic obstacle-crossing capability under high loads and without overturning tendency on complex and towering structures through the anti-tipping mechanical design of large and small drive wheels, dynamic group control algorithm based on obstacle encounter characteristics, and PID compensation control that integrates pressure and tilt angle feedback. This significantly improves the adaptability, stability and efficiency of climbing operations.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] An obstacle-crossing wheeled climbing robot with retractable modules includes four climbing modules and two retractable modules spaced two climbing modules apart and symmetrically arranged.

[0009] The climbing module includes a main plate and two long connecting rods and two short connecting rods hinged at the four corners of both ends of the main plate. The free ends of the long and short connecting rods are respectively rotatably connected to a small drive wheel and a large drive wheel with flexible treads. Motor end caps for housing the climbing motor are installed on the outside of the long and short connecting rods. The output shaft of the climbing motor is fixedly connected to a motor pulley. The outer sides of the shafts of the small and large drive wheels are fixedly installed with drive pulleys that are connected to the motor pulleys via synchronous belt transmission. The climbing motor drives the small drive wheel and the large drive wheel respectively through the motor pulleys, drive pulleys and synchronous belts.

[0010] The shrinking module includes a dual-axis motor, motor bevel gears fixedly connected to the output shaft of the dual-axis motor on both sides, and a rotating shaft bevel gear meshing with the motor bevel gears. A rotating shaft chain gear is fixedly installed in the middle of the rotating shaft where the rotating shaft bevel gears are located. A chain meshing with the rotating shaft chain gear is fixedly installed on one tooth of the rotating shaft chain gear. The chain connects the climbing module and the shrinking module in a circumferential manner, and the structure to be tested is encircled by the entire structure for climbing and testing.

[0011] Furthermore, the climbing module is equipped with an obstacle-crossing buffer unit and a double sprocket plate. The double sprocket plate is connected to the climbing module through the obstacle-crossing buffer unit. The double sprocket plate includes a sprocket frame mounted on the side of the main plate away from the tension spring and rotatably connected to the main plate, and a chain gear rotatably mounted on the sprocket frame and meshing with the chain through the sprocket shaft. The obstacle-crossing buffer unit includes four compression springs connected between the main plate and the sprocket frame.

[0012] Furthermore, a bottom support disc is fixedly installed on the outer side of the main plate, and a top support disc is fixedly installed on the sprocket frame on the opposite side of the main plate. Compression springs are fixedly installed between the top support disc and the bottom support disc. A set of double sprocket plate support elements is fixedly installed on the sprocket frame on the opposite side of the compression springs. A set of climbing module support elements that cooperate with the double sprocket plate support elements is fixedly installed on the main plate. The double sprocket plate support elements and the climbing module support elements have through holes for screws to pass through. The climbing module achieves rotational movement around the screws through the obstacle-crossing buffer unit.

[0013] Furthermore, connecting rods for fixing tension springs are provided between the two long connecting rods and the two short connecting rods in the middle; limit blocks are fixedly installed at the hinge points of the two long connecting rods, the two short connecting rods and the main plate, and pressure sensors are attached to the side of the limit blocks that contacts the long connecting rods.

[0014] Furthermore, the shrink module also includes a housing, and the two ends of the rotating shaft containing the bevel gear are fixed in the housing through shaft holes and bearings.

[0015] A control method for an obstacle-crossing wheeled climbing robot with a retractable module employs dynamic group control. Based on obstacle encounter characteristics, each robot component is grouped, where each robot component refers to a combination module consisting of a double-sprocket plate, an obstacle-crossing buffer unit, and a climbing module. According to the grouping scenario, two retractable modules are assigned control weights W to the four robot components, respectively. A / W B The four robot units are labeled P1, P2, P3, and P4, and the two shrinking modules are labeled A and B, respectively.

[0016] Based on the robot's single-chip obstacle encounter state O(k)(4×1 dimensional obstacle identification vector, O i =1 indicates that the i-th module encountered an obstacle and the tilt angle deviation. Define 4 core grouping scenarios:

[0017] Scenario 1, Single Front Obstacle Encounter G1: P1 encounters an obstacle, A leads the obstacle crossing, B assists in preventing rollover; Shrink module control weight: W A =diag([0.8,0.5,0.5,0.5]); W B =diag([0.2,0.5,0.5,0.5]); Triggers the G1 grouping condition: O=[1,0,0,0] and ;

[0018] Scenario 2, Dual Front Obstacle Encounter G2: Obstacle encounter P1+P2, A leads obstacle crossing, B assists in anti-rollover; Shrink module control weight: W A =diag([0.7,0.7,0.5,0.5]); W B=diag([0.3,0.3,0.5,0.5]); triggers the G2 grouping condition: O=[1,1,0,0] and , ;

[0019] Scenario 3, Cross-Domain Obstacle Encounter G3: P2+P3 obstacle encounter, dual contraction with weighted coordination to prevent tilting; contraction module control weight: W A =diag([0.5,0.5,0.5,0.5]); W B =diag([0.5,0.5,0.5,0.5]); Triggers the G3 grouping condition: O=[0,1,1,0] and , ;

[0020] Scenario 4, Full Module Obstacle Encounter G4: Overall obstacle crossing, dual contraction for balanced force output; Contraction module control weight: W A =diag([0.5,0.5,0.5,0.5]); W B =diag([0.5,0.5,0.5,0.5]); Triggers the G4 grouping condition: O=[1,1,1,1] and all .

[0021] Furthermore, for the control of the contraction module's control variables, the robot employs a data-driven, model-free adaptive algorithm; firstly, the robot's full-state variables are defined... Where α(k) refers to the tilt angle of the four climbing modules. This refers to the pressure sensor values ​​of the four climbing modules;

[0022] Secondly, define the PG estimation criterion function:

[0023]

[0024] Therefore, the recursive formula for the PG estimation algorithm is obtained:

[0025]

[0026] in, It is an 8×2 dimensional pseudo-gradient matrix; (2×1 vector, L=1);

[0027] initial value

[0028] The PG matrix is ​​corrected based on the grouping weights: = (W A W A )· +diag(W B WB )· .

[0029] Among them, diag(W) A W A ): An 8×8 diagonal matrix, formed by W A (4×1) Repeated twice to form the group weights to an 8-dimensional state.

[0030] Furthermore, tension and Define a multi-objective criterion function that covers state tracking and dual-tension smoothness for all four climbing modules; dual-input multi-objective criterion function:

[0031] in, It is the expected value of all variables in the module; State tracking error term (scalar); , : Scalar, tension smoothing factor; ||T A (k)-T A (k-1)‖²、‖T B (k)-T B (k-1)‖²: Bi-tension smoothing term (both are scalars);

[0032] The derivation of the dual-tension control law from the above formula yields:

[0033]

[0034]

[0035] in, : The first column (8×1 vector) represents For all states The impact; : The second column (8×1 vector) represents For all states The impact; : 2×1 vector (ΔT = Current tension - Historical tension).

[0036] Furthermore, after obtaining the basic control quantities for obstacle crossing, compensation for the robot's anti-tipping control quantities is still required. Calculate the tilt angle deviation of the i-th module and determine whether compensation is triggered:

[0037]

[0038] when When the value is greater than 0, tilt angle constraint compensation is initiated.

[0039] Furthermore, the control unit first calculates the overturning compensation control deviation. :

[0040]

[0041] Among them, the first item middle: This represents the difference between the real-time pressure value and the expected pressure value on each robot segment, depending on the degree of overturning of each robot segment. It needs to be multiplied by a coefficient. When a single robot segment tilts significantly, the tilt angle... Since the value is relatively small, the weight of its control effect needs to be increased, therefore its control error is multiplied by a larger coefficient. By summing up the pressure deviations of all individual robot components, deviations with high control effect weights and larger deviations can increase the control effect, i.e., the contraction and release of the contraction module; the second item For the tilt angle compensation term, the excessive tilt angle deviation is converted into an incremental deviation in the pressure control (the greater the deviation, the stronger the compensation). Then, a PID control algorithm is used to adaptively compensate the control output. In the PID controller, the proportional coefficient K... P Integral coefficient K I With differential coefficient K D It is set based on the system's control effect and debugging experience.

[0042] Output quantity in the overturning angle compensation control algorithm for:

[0043]

[0044] Through the PID control algorithm described above, the shrinking module can compensate for the corresponding control input based on the pressure deviation and overturning angle of each robot segment, ensuring that the robot has no tendency to overturn.

[0045] Final control quantity: ; ;

[0046] Through the aforementioned data-driven control and control quantity compensation algorithms, the contraction module can stably contract and release the chain, ensuring that the robot can stably climb the wall and overcome obstacles without the tendency to tip over, thus guaranteeing the stability and speed of the robot's climbing and obstacle crossing.

[0047] The beneficial effects of this invention are as follows:

[0048] 1. The obstacle-crossing wheeled climbing robot with a retractable module disclosed in this invention utilizes an obstacle-crossing buffer unit as its core mechanism during obstacle crossing. When the drive wheel (small or large drive wheel) contacts an obstacle, the obstacle's reaction force is transmitted through a linkage, driving the entire climbing module to rotate around a screw in the obstacle-crossing buffer unit. This compresses a compression spring on one side, allowing the drive wheel to lift and cross the obstacle. This process provides crucial attitude adjustment space for obstacle crossing. After crossing the obstacle, the robot can quickly regain stable adhesion to the wall surface under the combined action of the compression spring's restoring force and the retractable module's real-time control of the chain's extension and retraction. This combination of mechanical buffering and active control effectively avoids the instantaneous loss of adhesion or impact caused by the retractable module's response delay, thus significantly ensuring the speed and stability of the obstacle crossing process.

[0049] 2. The obstacle-crossing wheeled climbing robot with a retractable module disclosed in this invention uses drive wheels of different radii to ensure that the line connecting the axles of the two wheels always forms an acute angle with the normal direction of the wall. This design creates a natural anti-tipping moment. When the robot tends to tilt forward, the design uses gravity to resist tipping; while when the robot tends to tilt excessively backward, the design provides a certain amount of anti-tipping space. This passive anti-tipping mechanism based on physical structure does not rely on real-time calculations of complex algorithms, providing a basic and reliable stability guarantee for the robot's high-load operation, and exhibits higher robustness compared to robots that solely rely on control strategies.

[0050] 3. The control method for an obstacle-crossing wheeled climbing robot with a retractable module disclosed in this invention surpasses the traditional control mode of pre-planned fixed actions, achieving intelligent dynamic group control. The system dynamically identifies core group scenarios based on the real-time obstacle encounter status and tilt angle deviation of individual robot modules. For different scenarios, optimized allocation of control resources is achieved by assigning retractable modules and different control weight matrices. For example, when encountering an obstacle on one side, the nearest retractable module leads the obstacle crossing, while another module assists in preventing tipping; when encountering obstacles across domains or all modules, a dual-module weighted collaborative strategy is adopted. This dynamic decision-making based on real-time sensor data enables the robot to adaptively cope with pole surface terrain, maintaining precise and stable control of wall adhesion during obstacle crossing.

[0051] 4. The control method for the obstacle-crossing wheeled climbing robot with a retractable module disclosed in this invention, to further enhance anti-tipping capability, introduces a PID compensation control algorithm that integrates multi-sensor information on the basis of data-driven control. This algorithm defines a comprehensive control deviation, calculates this comprehensive deviation using the PID algorithm, outputs a compensation quantity, and adds it to the basic control quantity. This feedforward-feedback composite control strategy can proactively and predictively suppress tipping tendencies, significantly improving the robot's anti-interference capability and overall stability under complex climbing and dynamic obstacle-crossing conditions.

[0052] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0053] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0054] Figure 1 This is a structural schematic diagram of the obstacle-crossing wheeled climbing robot with a retractable module according to the present invention.

[0055] Figure 2 For the present invention Figure 1 A schematic diagram of the climbing module;

[0056] Figure 3 For the present invention Figure 1 Schematic diagram of the structure of the double sprocket plate;

[0057] Figure 4 For the present invention Figure 1 A schematic diagram of the structure of the shrink module;

[0058] Figure 5 For the present invention Figure 1 Schematic diagram of the structure of the obstacle crossing buffer unit;

[0059] Figure 6 This is a schematic diagram illustrating the process of the climbing module overcoming obstacles in this invention.

[0060] Figure 7 This is a schematic diagram illustrating the process of the climbing module descending over obstacles in this invention;

[0061] Figure 8 This is a schematic diagram of the overturning angle in this invention;

[0062] Figure 9This is a schematic diagram showing the markings of the individual pieces and the retraction module of the climbing robot in this invention.

[0063] Reference numerals: 1-Climbing module; 1-1 Flexible tread; 1-2 Small drive wheel; 1-3 Drive pulley; 1-4 Synchronous belt; 1-5 Long connecting rod; 1-6 Motor pulley; 1-7 Climbing motor; 1-8 Motor end cover; 1-9 Limiting block; 1-10 Pressure sensor; 2-Double sprocket plate; 2-1 Sprocket frame; 2-2 Sprocket shaft; 2-3 Chain gear; 3-Retracting module; 3-1 Shaft; 3-2 Shaft chain gear; 3-3 Shaft bevel gear; 3-4 Dual-axis motor; 3-5 Motor bevel gear; 3-6 Chain; 3-7 Housing; 4 Obstacle-crossing buffer unit; 4-1 Top support disc; 4-2 Compression spring; 4-3 Bottom support disc; 4-4 Double sprocket plate support element; 4-5 Climbing module support element; 4-6 Nut; 4-7 Screw. Detailed Implementation

[0064] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0065] like Figure 1 The illustrated obstacle-crossing wheeled climbing robot with retractable modules, taking the climbing robot climbing on a structure with a circular end face such as a bridge pier or wind turbine tower as an example, specifically includes four climbing modules 1 that climb the surface of the bridge pier or wind turbine tower. A retractable module 3 is placed every two climbing modules 1, with the two retractable modules 3 positioned opposite each other. The four climbing modules 1 are circumferentially connected by chains 3-6 and the two retractable modules 3, achieving a circumferentially even distribution to completely encircle the surface of the bridge pier or wind turbine tower. The climbing modules 1 and retractable modules 3 are the components through which the climbing robot directly climbs. The double sprocket plate 2 is connected to the climbing modules 1 through an obstacle-crossing buffer unit 4, and directly connected to the retractable modules 3 via chains 3-6 and chain gears 2-3. The rotation of the chain gears 2-3 can counteract the lateral force when the retractable modules 3 retract.

[0066] like Figure 2 The climbing module 1 shown includes a main plate 1-11 and two long connecting rods 1-5 and two short connecting rods 1-13 respectively hinged at the four corners of both ends of the main plate 1-11. The two long connecting rods 1-5 and the two short connecting rods 1-13 are fixedly installed with limiting blocks 1-9 at the hinge points with the main plate 1-11 by pins. A pressure sensor 1-10 is attached to the side of the limiting block 1-9 that contacts the long connecting rod 1-5.

[0067] Two long connecting rods 1-5 and two short connecting rods 1-13 are connected to each other via connecting rod holes at their midpoints; a tension spring 1-12 is fixedly installed between the two connecting rods 1-14; the shaft of the small drive wheel 1-2 is connected to one end of the two long connecting rods 1-5 via bearings at both ends, and the shaft of the large drive wheel 1-15 is connected to the two short connecting rods 1-13 via bearings at both ends; a flexible tread 1-1 is embedded in the small drive wheel 1-2 and the large drive wheel 1-15; a climbing motor 1- is installed on the outer side of the two long connecting rods 1-5 and the two short connecting rods 1-13. The motor end cap 1-8 of 7 and the motor pulley 1-6 are fixedly connected to the output shaft of the climbing motor 1-7. The outer side of the shafts of the small drive wheel 1-2 and the large drive wheel 1-15 are fixedly installed with drive pulleys 1-3 corresponding to the motor pulley 1-6. The climbing motor 1-7 drives the small drive wheel 1-2 and the large drive wheel 1-15 respectively through the motor pulley 1-6, drive pulley 1-3 and synchronous belt 1-4. The motor pulleys 1-6, drive pulleys 1-3 and synchronous belt 1-4 of the small drive wheel 1-2 and the large drive wheel 1-15 are all installed on the same side of the robot piece.

[0068] like Figure 3 The double sprocket plate 2 shown includes a sprocket frame 2-1 mounted on the side of the main plate 1-11 away from the tension spring 1-12 and rotatably connected to the main plate 1-11, and two sets of four chain gears 2-3 rotatably mounted on the sprocket frame 2-1 via a sprocket shaft 2-2. Specifically, both ends of the sprocket shaft 2-2 are fixed in the sprocket frame 2-1 through shaft holes and bearings; the two sets of chain gears 2-3 are symmetrically fixed on the sprocket shaft 2-2, and the chain 3-6 meshes with the chain gears 2-3; when the shrinking module 3 shrinks the chain 3-6, one end of the chain 3-6 is fixed on one tooth of the chain gear 3-2 on the shrinking module shaft, and at the same time, the chain 3-6 meshes with the chain gears 2-3 of the double sprocket plate 2, driving the chain gears 2-3 of the double sprocket plate 2 to rotate.

[0069] like Figure 5The obstacle-crossing buffer unit 4 shown includes four compression springs 4-2 connected between the main body plate 1-11 and the sprocket frame 2-1. Specifically, four bottom support discs 4-3 are fixedly installed on the outer side of the main body plate 1-11, and four top support discs 4-1 are fixedly installed on the sprocket frame 2-1 opposite to the main body plate 1-11. The compression springs 4-2 with intermediate rods are fixedly installed between the top support discs 4-1 and the bottom support discs 4-3 respectively. A set of double sprocket plate support elements 4-4 is fixedly installed on the sprocket frame 2-1 opposite to the compression springs 4-2. A set of climbing module supports that cooperate with the double sprocket plate support elements 4-4 is fixedly installed on the main body plate 1-11. Components 4-5, 4-4, and 4-5 have interconnected shaft holes. Screws 4-7 pass through the shaft holes of 4-4 and 4-5 and are installed between them via the shaft holes. The threaded outer end of screw 4-7 is fixed with nuts 4-6. The top support disc 4-1 is connected to the sprocket frame 2-1 in the sprocket plate 2 and is connected to the intermediate rod and bottom support disc 4-3 via compression spring 4-2. The intermediate rod serves as a structural through-and connection. The climbing module 1 can rotate around screw 4-7 via obstacle-crossing buffer unit 4.

[0070] like Figure 4 The shrinking module 3 shown includes a dual-axis motor 3-4, motor bevel gear 3-5, shaft bevel gear 3-3, shaft 3-1, shaft chain gear 3-2, housing 3-7, and chain 3-6. The motor bevel gear 3-5, which is fixedly connected to the output shaft of the dual-axis motor 3-4 on both sides, meshes with the shaft bevel gear 3-3. The two ends of the shaft 3-1, where the shaft bevel gear 3-3 is located, are fixed in the housing 3-7 through shaft holes and bearings. The shaft chain gear 3-2, which meshes with the chain 3-6, is fixed on the shaft 3-1. When the shrinking module 3 shrinks the chain 3-6, the shaft chain gear 3-2 on the shaft 3-1 rotates actively, driving the chain 3-6 to shrink. The maximum shrinking length of the chain 3-6 is one revolution of the shaft chain gear 3-2.

[0071] like Figure 6As shown, the working principle of the obstacle-crossing wheeled climbing robot with a retractable module is as follows: When the robot encounters an obstacle during its ascent, the small drive wheel 1-2 in contact with the obstacle experiences pressure from the obstacle. At this time, the huge friction from the flexible tread 1-1 and the power transmitted by the climbing motor 1-7 allow the small drive wheel 1-2 to adhere tightly to the obstacle. The small drive wheel 1-2 drives the long connecting rod 1-5 to lift away from the obstacle. The long connecting rod 1-5 further drives the climbing module 1 to rotate away from the obstacle around the screw 4-7 in the obstacle-crossing buffer unit 4. The rotating climbing module 1 causes the two compression springs 4-2 near the long connecting rod 1-5 to compress slightly and the two compression springs 4-2 near the short connecting rod 1-13 to stretch slightly. After the small drive wheel 1-2 crosses the obstacle, the obstacle-crossing buffer unit 1-2... The four compression springs 4-2 of the obstacle-crossing unit 4 return to their original positions, and the robot continues to rise. When the large drive wheel 1-15 is subjected to the pressure of the obstacle, the huge friction from the flexible tread 1-1 and the power transmitted by the climbing motor 1-7 enable the large drive wheel 1-15 to adhere tightly to the obstacle. The large drive wheel 1-15 drives the short connecting rod 1-13 to lift away from the obstacle. The short connecting rod 1-13 further drives the climbing module 1 to rotate away from the obstacle around the screw 4-7 in the obstacle-crossing buffer unit 4. The rotating climbing module 1 causes the two compression springs 4-2 near the short connecting rod 1-13 to compress slightly and the two compression springs 4-2 near the long connecting rod 1-5 to stretch slightly. After the large drive wheel 1-15 crosses the obstacle, the four compression springs 4-2 of the obstacle-crossing buffer unit 4 return to their original positions.

[0072] like Figure 7As shown, when the robot encounters an obstacle during its descent, the large drive wheel 1-15 in contact with the obstacle experiences pressure from the obstacle. At this time, the significant friction from the flexible tread 1-1 and the power transmitted by the climbing motor 1-7 allow the large drive wheel 1-15 to adhere tightly to the obstacle. The large drive wheel 1-15 then lifts the short connecting rod 1-13 away from the obstacle. The short connecting rod 1-13 further drives the climbing module 1 to rotate around the screw 4-7 in the obstacle-crossing buffer unit 4, moving away from the obstacle. The rotating climbing module 1 causes two compression springs 4-2 near the short connecting rod 1-13 to compress slightly, and two compression springs 4-2 near the long connecting rod 1-5 to stretch slightly. After the large drive wheel 1-15 passes the obstacle, the obstacle-crossing buffer unit 4... The four compression springs 4-2 reset; the robot continues to descend; when the small drive wheel 1-2 is subjected to the pressure of the obstacle, the huge friction from the flexible tread 1-1 and the power transmitted by the climbing motor 1-7 enable the small drive wheel 1-2 to adhere tightly to the obstacle; the small drive wheel 1-2 drives the long connecting rod 1-5 to lift away from the obstacle; the long connecting rod 1-5 further drives the climbing module 1 to rotate away from the obstacle around the screw 4-7 in the obstacle-crossing buffer unit 4. The rotating climbing module 1 causes the two compression springs 4-2 near the long connecting rod 1-5 to compress slightly, and causes the two compression springs 4-2 near the short connecting rod 1-13 to stretch slightly; after the small drive wheel 1-2 crosses the obstacle, the four compression springs 4-2 of the obstacle-crossing buffer unit 4 reset.

[0073] When the climbing module 1 adheres to the wall, the retraction module 3 retracts and releases the chain 3-6 to maintain the sensing value of the pressure sensor 1-10 at the desired pressure value, thus enabling the climbing module 1 to stably adhere to the wall. When the climbing module 1 overcomes obstacles, it needs a certain obstacle-crossing space to pass over the obstacle. If the retraction module 3 cannot release the chain 3-6 in time to provide obstacle-crossing space, the climbing module 1 obtains obstacle-crossing space by compressing the compression spring 4-2 in the obstacle-crossing buffer unit 4 through the reaction force of the obstacle. The pressure sensor 1-10 detects an increase in pressure value. The retraction module 3 releases the chain 3-6, restoring the pressure value to the desired pressure value. At the same time, the compression spring 4-2 of the obstacle-crossing buffer unit 4 resets, and the released chain 3-6 provides obstacle-crossing space. Therefore, the obstacle-crossing buffer unit 4 can provide space for obstacle crossing, ensuring the stability and speed of the climbing and obstacle-crossing robot.

[0074] like Figure 8As shown, the radius difference between the large drive wheel 1-15 and the small drive wheel 1-2 constitutes the robot's anti-tipping design. Let α be the angle between the extension of the line connecting the center points of the small drive wheel 1-2 and the large drive wheel 1-15 and the wall. α is always greater than 0°. If α appears closer to the small drive wheel 1-2 and farther from the large drive wheel 1-15, the robot has no tendency to tip over. If α does not exist, the extension of the line connecting the center points of the two wheels is parallel to the wall, and the robot is on the edge of tipping tendency. If α appears closer to the large drive wheel 1-15 and farther from the small drive wheel 1-2, the robot has a tendency to tip over, affecting the overall stability of the robot.

[0075] The obstacle-crossing wheeled climbing robot employs dynamic group control for overall obstacle crossing. Based on obstacle encounter characteristics, each robot component is grouped, where each robot component refers to a combined module consisting of the double-sprocket plate 2, the obstacle-crossing buffer unit 4, and the climbing module 1. According to the grouping scenario, two contraction modules are assigned control weights W to the four robot components respectively. A / W B .like Figure 9 As shown, the four robot units are labeled P1, P2, P3, and P4, and the two shrinking modules are labeled A and B.

[0076] Based on the robot's obstacle encounter status O (k) (4×1 dimensional obstacle identification vector, O) i =1 indicates that the i-th module encountered an obstacle and the tilt angle deviation. Define 4 core grouping scenarios:

[0077] Scenario 1, Single Front Obstacle Encounter G1: P1 encounters an obstacle, A leads the obstacle crossing, B assists in preventing rollover; Shrink module control weight: W A =diag([0.8,0.5,0.5,0.5]); W B =diag([0.2,0.5,0.5,0.5]); Triggers the G1 grouping condition: O=[1,0,0,0] and ;

[0078] Scenario 2, Dual Front Obstacle Encounter G2: Obstacle encounter P1+P2, A leads obstacle crossing, B assists in anti-rollover; Shrink module control weight: W A =diag([0.7,0.7,0.5,0.5]); W B =diag([0.3,0.3,0.5,0.5]); triggers the G2 grouping condition: O=[1,1,0,0] and , ;

[0079] Scenario 3, Cross-Domain Obstacle Encounter G3: P2+P3 obstacle encounter, dual contraction with weighted coordination to prevent tilting; contraction module control weight: W A =diag([0.5,0.5,0.5,0.5]); W B =diag([0.5,0.5,0.5,0.5]); Triggers the G3 grouping condition: O=[0,1,1,0] and , ;

[0080] Scenario 4, Full Module Obstacle Encounter G4: Overall obstacle crossing, dual contraction for balanced force output; Contraction module control weight: W A =diag([0.5,0.5,0.5,0.5]); W B =diag([0.5,0.5,0.5,0.5]); Triggers the G4 grouping condition: O=[1,1,1,1] and all .

[0081] For controlling the shrinkage module's control variables, the robot employs a data-driven, model-free adaptive algorithm. First, the robot's full-state variables are defined. Where α(k) refers to the tilt angle of the four climbing modules. This refers to the pressure sensor values ​​of the four climbing modules.

[0082] Secondly, define the PG estimation criterion function:

[0083]

[0084] Therefore, the recursive formula for the PG estimation algorithm is obtained:

[0085]

[0086] in, It is an 8×2 dimensional pseudo-gradient matrix; (2×1 vector, L=1);

[0087] initial value

[0088] The PG matrix is ​​corrected based on the grouping weights:

[0089] =diag(W A W A )· +diag(W B W B )· .

[0090] Among them, diag(W) A W A): An 8×8 diagonal matrix, formed by W A (4×1) Repeated twice to construct, mapping the group weights to an 8-dimensional state; for example, in scenario G1, W A The P1 weight is 0.8. middle" "Element amplification strengthens the control effect of A on P1."

[0091] For the tension of the dual-shrinkage module and T B (k) Define a multi-objective criterion function that covers state tracking and dual-tension smoothness for all four climbing modules. Dual-input multi-objective criterion function:

[0092] in, It is the expected value of all variables in the module; State tracking error term (scalar); , : Scalar, tension smoothing factor; ||T A (k)-T A (k-1)‖²、‖T B (k)-T B (k-1)‖²: Bi-tension smoothing term (both are scalars).

[0093] The derivation of the dual-tension control law from the above formula yields:

[0094]

[0095]

[0096] in, : The first column (8×1 vector) represents For all states The impact; : The second column (8×1 vector) represents For all states The impact; : 2×1 vector (ΔT = Current tension - Historical tension).

[0097] After obtaining the basic control quantities for obstacle crossing, compensation for the robot's anti-tipping control is still required. Calculate the tilt angle deviation of the i-th module and determine whether compensation is triggered:

[0098]

[0099] when When the value is greater than 0, tilt angle constraint compensation is initiated.

[0100] The control unit first calculates the overturning compensation control deviation e(t):

[0101]

[0102] Among them, the first item middle: This represents the difference between the real-time pressure value and the expected pressure value on each robot segment, depending on the degree of overturning of each robot segment. It needs to be multiplied by a coefficient. When a single robot segment tilts significantly, the tilt angle... Since the value is relatively small, the weight of its control effect needs to be increased, therefore its control error is multiplied by a larger coefficient. By summing up the pressure deviations of all individual robot components, deviations with high control effect weights and larger deviations can increase the control effect, i.e., the contraction and release of the contraction module; the second item For the tilt angle compensation term, the excessive tilt angle deviation is converted into an incremental deviation in the pressure control (the greater the deviation, the stronger the compensation). Then, a PID control algorithm is used to adaptively compensate the control output. In the PID controller, the proportional coefficient K... P Integral coefficient K I With differential coefficient K D It is set based on the system's control effect and debugging experience.

[0103] Output quantity in the overturning angle compensation control algorithm for:

[0104]

[0105] Through the PID control algorithm described above, the shrinking module can compensate for the corresponding control input based on the pressure deviation and overturning angle of each robot segment, ensuring that the robot has no tendency to overturn.

[0106] Final control quantity: ; ;

[0107] Through the aforementioned data-driven control and control quantity compensation algorithms, the contraction module can stably contract and release the chain, ensuring that the robot can stably climb the wall and overcome obstacles without the tendency to tip over, thus guaranteeing the stability and speed of the robot's climbing and obstacle crossing.

[0108] The climbing method of this obstacle-crossing wheeled climbing robot with a retractable module is as follows:

[0109] S1. After removing the two chains 3-6 from one side of one of the shrinking modules 3, place the robot on the rod surface, and then fix the chains 3-6 to the rotating shaft chain gear 3-2 of the shrinking module 3.

[0110] S2, the retraction module 3 retracts the chain 3-6, causing the long connecting rod 1-5 and the short connecting rod 1-13 to reach the limit block 1-9, thereby allowing the small drive wheel 1-2 and the large drive wheel 1-15 to stably adhere to the pole surface under the action of the tension spring 1-12.

[0111] S3. The obstacle-crossing wheeled climbing robot with retractable module 3 climbs or descends on the pole surface; when the small drive wheel 1-2 and the large drive wheel 1-15 rotate synchronously, they drive the obstacle-crossing wheeled climbing robot with retractable module 3 to climb or descend.

[0112] The process by which this obstacle-crossing wheeled climbing robot with a retractable module climbs and overcomes obstacles on the pole is as follows:

[0113] S1. The eight climbing motors 1-7 start simultaneously and rotate in the same direction, driving the four small drive wheels 1-2 and the four large drive wheels 1-15 to rotate synchronously in the same direction, and the robot begins to climb.

[0114] S2. When the robot encounters an obstacle during its climbing motion, the small drive wheel 1-2, which comes into contact with the obstacle, experiences resistance from the obstacle. This resistance causes significant deformation on the flexible tread 1-1, generating substantial friction. This friction, along with the power from the climbing motor 1-7, drives the small drive wheel 1-2 to adhere to the obstacle. The small drive wheel 1-2 further drives the long connecting rod 1-5, which in turn drives the main body plate 1-11, causing the climbing module 1 to rotate around the screw 4-7. This rotation of the climbing module 1 further compresses the two compression springs 4-2 near the small drive wheel 1-2 and releases the two compression springs 4-2 near the large drive wheel 1-15. A compression spring 4-2, passing through the main body plate 1-11 and the limiting block 1-9, further increases the pressure value obtained by the pressure sensor 1-10 at the limiting block 1-9. The increased pressure value obtained by the pressure sensor 1-10 exceeds the expected pressure value, causing the contraction module 3 to release the chain 3-6. The released chain 3-6 drives the chain gear 2-3 on the double sprocket plate 2 to rotate, increasing the distance between the double sprocket plate 2 and the wall surface. The double sprocket plate 2 drives two release springs near the small drive wheel 1-2, and compresses two compression springs 4-2 near the large drive wheel 1-15. When the pressure value obtained by the pressure sensor 1-10 equals the expected pressure value, the contraction module 3 stops releasing the chain 3-6.

[0115] S3. After the small drive wheels 1-2 have completely crossed the obstacle, the robot continues to climb until the large drive wheels 1-15 are subjected to resistance from the obstacle. During the process of the large drive wheels 1-15 crossing the obstacle, the pressure value obtained by the pressure sensor 1-10 is maintained by the cooperation of each module, so as to achieve stable obstacle crossing of the robot.

[0116] S4. When the robot reaches the expected climbing height, the eight climbing motors 1-7 stop rotating, and the climbing motion stops.

[0117] The obstacle-crossing wheeled climbing robot with retraction module 3 descends and overcomes obstacles on the pole as follows:

[0118] S1, the eight climbing motors 1-7 start rotating in opposite directions at the same time, driving the four small drive wheels 1-2 and the four large drive wheels 1-15 to rotate synchronously in opposite directions, and the robot begins to descend.

[0119] S2. When the robot encounters an obstacle during its climbing motion, the large drive wheel 1-15, which comes into contact with the obstacle, experiences resistance from the obstacle. This resistance causes significant deformation on the flexible tread 1-1, resulting in substantial friction. This friction, along with the power from the climbing motor 1-7, drives the large drive wheel 1-15 to climb the obstacle. The large drive wheel 1-15 further drives the short connecting rod 1-13, which in turn drives the main body plate 1-11, causing the climbing module 1 to rotate around the screw 4-7. This rotation of the climbing module 1 further compresses the two compression springs 4-2 near the large drive wheel 1-15 and releases the two compression springs 4-2 near the small drive wheel 1-2. The two compression springs 4-2 on the side, through the main body plate 1-11 and the limiting block 1-9, further increase the pressure value obtained by the pressure sensor 1-10 at the limiting block 1-9. The pressure value obtained by the pressure sensor 1-10 increases, exceeding the expected pressure value, causing the contraction module 3 to start releasing the chain 3-6. The released chain 3-6 drives the chain gear 2-3 on the double sprocket plate 2 to rotate, increasing the distance between the double sprocket plate 2 and the wall. The double sprocket plate 2 drives the two release springs on the side near the large drive wheel 1-15, and drives the two compression springs 4-2 on the side near the small drive wheel 1-2 to compress. When the pressure value obtained by the pressure sensor 1-10 is equal to the expected pressure value, the contraction module 3 stops releasing the chain 3-6.

[0120] S3. After the large drive wheels 1-15 have completely crossed the obstacle, the robot continues to climb until the small drive wheels 1-2 are subjected to resistance from the obstacle. During the process of the small drive wheels 1-2 crossing the obstacle, the pressure value obtained by the pressure sensor 1-10 is maintained by the cooperation of each module, so as to achieve stable obstacle crossing of the robot.

[0121] S4. When the robot reaches the expected descent height, the eight climbing motors 1-7 stop rotating, and the descent stops.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A wheeled obstacle-crossing climbing robot with a retractable module, characterized in that, It includes four climbing modules (1) and two contraction modules (3) that are symmetrically arranged with two climbing modules (1) spaced apart; The climbing module (1) includes a main body plate (1-11) and two long connecting rods (1-5) and two short connecting rods (1-13) respectively hinged at the four corners of both ends of the main body plate (1-11). The free ends of the long connecting rods (1-5) and the short connecting rods (1-13) are respectively rotatably connected to a small drive wheel (1-2) and a large drive wheel (1-15) with flexible treads (1-1). Motor end caps (1-7) for placing the climbing motor (1-7) are installed on the outside of the long connecting rods (1-5) and the short connecting rods (1-13). -8), the output shaft of the climbing motor (1-7) is fixedly connected to the motor pulley (1-6). The outer side of the shafts of the small drive wheel (1-2) and the large drive wheel (1-15) are fixedly installed with drive pulleys (1-3) that are connected to the motor pulley (1-6) through the synchronous belt (1-4). The climbing motor (1-7) drives the small drive wheel (1-2) and the large drive wheel (1-15) respectively through the motor pulley (1-6), the drive pulley (1-3) and the synchronous belt (1-4). The shrinking module (3) includes a dual-axis motor (3-4), motor bevel gears (3-5) installed on both sides of the dual-axis motor (3-4) and fixedly connected to the output shaft of the dual-axis motor (3-4), and a rotating shaft bevel gear (3-3) meshing with the motor bevel gear (3-5). A rotating shaft chain gear (3-2) is fixedly installed in the middle of the rotating shaft (3-1) where the rotating shaft bevel gear (3-3) is located. A chain (3-6) meshing with the rotating shaft chain gear (3-2) is fixedly installed on one tooth of the rotating shaft chain gear (3-2). The chain (3-6) connects the climbing module (1) and the shrinking module (3) in a circumferential manner, and encircles the structure to be tested for climbing test.

2. The obstacle-crossing wheeled climbing robot as described in claim 1, characterized in that, The climbing module (1) is equipped with an obstacle-crossing buffer unit (4) and a double sprocket plate (2). The double sprocket plate (2) is connected to the climbing module (1) through the obstacle-crossing buffer unit (4). The double sprocket plate (2) includes a sprocket frame (2-1) installed on the side of the main plate (1-11) away from the tension spring (1-12) and rotatably connected to the main plate (1-11), and a chain gear (2-3) rotatably installed on the sprocket frame (2-1) through the sprocket shaft (2-2) and meshing with the chain (3-6). The obstacle-crossing buffer unit (4) includes four compression springs (4-2) connected between the main plate (1-11) and the sprocket frame (2-1).

3. The obstacle-crossing wheeled climbing robot as described in claim 2, characterized in that, A bottom support disc (4-3) is fixedly installed on the outside of the main plate (1-11). A top bearing disc (4-1) is fixedly installed on the sprocket frame (2-1) on the opposite side of the main plate (1-11). Compression springs (4-2) are fixedly installed between the top bearing disc (4-1) and the bottom support disc (4-3). A set of double sprocket plate support elements (4-4) is fixedly installed on the sprocket frame (2-1) on the opposite side of the compression springs (4-2). A set of climbing module support elements (4-5) that cooperate with the double sprocket plate support elements (4-4) is fixedly installed on the main plate (1-11). The double sprocket plate support elements (4-4) and the climbing module support elements (4-5) have through holes for screws (4-7) to pass through. The climbing module (1) achieves rotational movement with screws (4-7) as the axis through the obstacle-crossing buffer unit (4).

4. The obstacle-crossing wheeled climbing robot as described in claim 3, characterized in that, Two long connecting rods (1-5) and two short connecting rods (1-13) are respectively provided with connecting rods (1-14) for fixing and installing tension springs (1-12) in the middle; limit blocks (1-9) are fixedly installed at the hinge joints between the two long connecting rods (1-5), the two short connecting rods (1-13) and the main plate (1-11), and pressure sensors (1-10) are attached to the side of the limit block (1-9) that contacts the long connecting rod (1-5).

5. The obstacle-crossing wheeled climbing robot as described in claim 4, characterized in that, The shrink module (3) also includes a housing (3-7), and the two ends of the rotating shaft (3-1) where the rotating shaft bevel gear (3-3) is located are fixed in the housing (3-7) through shaft holes and bearings.

6. The control method for the obstacle-crossing wheeled climbing robot as described in any one of claims 1 to 5, characterized in that, Dynamic group control is adopted. Based on obstacle encounter characteristics, each robot unit is grouped. The robot unit refers to the combination module of double sprocket plate (2), obstacle crossing buffer unit (4) and climbing module (1). According to the grouping scenario, two contraction modules are assigned control weights W to the four robot units respectively. A / W B The four robot units are labeled P1, P2, P3, and P4, and the two shrinking modules are labeled A and B, respectively. Based on the robot's obstacle encounter status O (k) (4×1 dimensional obstacle identification vector, O) i =1 indicates that the i-th module encountered an obstacle and the tilt angle deviation. Define 4 core grouping scenarios: Scenario 1, Single Front Obstacle Encounter G1: P1 encounters an obstacle, A leads the obstacle crossing, B assists in preventing rollover; Shrink module control weight: W A =diag([0.8,0.5,0.5,0.5]); W B =diag([0.2,0.5,0.5,0.5]); Triggers the G1 grouping condition: O=[1,0,0,0] and ≥2°; Scenario 2, Dual Front Obstacle Encounter G2: Obstacle encounter P1+P2, A leads obstacle crossing, B assists in anti-rollover; Shrink module control weight: W A =diag([0.7,0.7,0.5,0.5]); W B =diag([0.3,0.3,0.5,0.5]); Triggers the G2 grouping condition: O=[1,1,0,0] and , ; Scenario 3, Cross-Domain Obstacle Encounter G3: P2+P3 obstacle encounter, dual contraction with weighted coordination to prevent tilting; contraction module control weight: W A =diag([0.5,0.5,0.5,0.5]); W B =diag([0.5,0.5,0.5,0.5]); triggers the G3 grouping condition: O=[0,1,1,0] and , ; Scenario 4, Full Module Obstacle Encounter G4: Overall obstacle crossing, dual contraction for balanced force output; Contraction module control weight: W A =diag([0.5,0.5,0.5,0.5]); W B =diag([0.5,0.5,0.5,0.5]); Triggers the G4 grouping condition: O=[1,1,1,1] and all .

7. The control method for the obstacle-crossing wheeled climbing robot as described in claim 6, characterized in that, For controlling the control variables of the contraction module, the robot adopts a data-driven model-free adaptive algorithm; firstly, the robot's full-state variables are defined. Where α(k) refers to the tilt angle of the four climbing modules. This refers to the pressure sensor values ​​of the four climbing modules; Secondly, define the PG estimation criterion function: Therefore, the recursive formula for the PG estimation algorithm is obtained: in, It is an 8×2 dimensional pseudo-gradient matrix; (2×1 vector, L=1); initial value The PG matrix is ​​corrected based on the grouping weights: =diag(W A W A )· +diag(W B W B )· ; Among them, diag(W) A W A ): An 8×8 diagonal matrix, formed by W A (4×1) Repeated twice to form the group weights to an 8-dimensional state.

8. The control method for the obstacle-crossing wheeled climbing robot as described in claim 7, characterized in that, tension and Define a multi-objective criterion function that covers state tracking and dual-tension smoothness for all four climbing modules; dual-input multi-objective criterion function: in, It is the expected value of all variables in the module; State tracking error term (scalar); : Scalar, tension smoothing factor; ||T A (k)-T A (k-1)‖²、‖T B (k)-T B (k-1)‖²: Bi-tension smoothing term (both are scalars); The derivation of the dual-tension control law from the above formula yields: in, : The first column (8×1 vector) represents For all states The impact; : The second column (8×1 vector) represents For all states The impact; : 2×1 vector (ΔT = Current tension - Historical tension).

9. The control method for the obstacle-crossing wheeled climbing robot as described in claim 8, characterized in that, After obtaining the basic control variables for obstacle crossing, compensation needs to be made for the robot's anti-tipping control variables; calculate the tilt angle deviation of the i-th module and determine whether compensation should be triggered: when When the value is greater than 0, tilt angle constraint compensation is initiated.

10. The control method for the obstacle-crossing wheeled climbing robot as described in claim 9, characterized in that, The control unit first calculates the overturning compensation control deviation. for: in, This indicates the serial number of the robot chip. Indicates the expected stress value. This represents the real-time pressure feedback value of the current sequence number of the robot chip. Indicates the tilt degree of the current sequence number of the robot chip; The tilt angle compensation term converts the excessive tilt angle deviation into an incremental deviation in pressure control. The tilt angle compensation coefficient is used, and a PID control algorithm is employed to adaptively compensate and control the output. : in, K is the input parameter for the shrink module system. P K is the proportionality coefficient. I K is the integral coefficient. D These are the differential coefficients; Through the PID control algorithm described above, the shrinking module can compensate for the corresponding control inputs based on the pressure deviation and tipping angle of each robot segment, ensuring that the robot has no tendency to tip over. Final control quantity: ; ; Through the aforementioned data-driven control and control quantity compensation algorithms, the contraction module can stably contract and release the chain, ensuring that the robot can stably climb the wall and overcome obstacles without the tendency to tip over, thus guaranteeing the stability and speed of the robot's climbing and obstacle crossing.

Images