阅读说明：本技术 一种自主感知移动的航天器表面爬行机器人 (Spacecraft surface crawling robot capable of autonomously sensing and moving ) 是由 张育林 张斌斌 向澳 郑明月 于 2021-08-06 设计创作，主要内容包括：本发明公开了一种自主感知移动的航天器表面爬行机器人,该爬行机器人包括机械结构和控制系统两部分。机械结构包括十八个舵机及其连接机构组成的六个多自由度腿足和机器人平台主体,平台主体安装有蓄电池,由继电器开关、升压器、静电吸附膜构成的静电吸附模块,以及内置IMU传感器的深度相机,爬行机器人基于静电吸附原理实现在航天器表面的吸附。控制系统采用嵌入式微处理器及扩展接口板等硬件,通过感知-规划-控制实现自主移动。同时可配合机器人步态控制吸附力的有无,减小机器人抬腿所需的力矩。最后提出一种动力学仿真方法计算机器人所需吸附力大小,并验证了机器人爬行能力。本发明实现了机器人在航天器表面的吸附、爬行、自主感知和移动。(The invention discloses an autonomous sensing and moving spacecraft surface crawling robot, which comprises a mechanical structure and a control system. The mechanical structure comprises eighteen steering engines and six multi-degree-of-freedom legs and feet formed by connecting mechanisms of the steering engines and a robot platform main body, a storage battery, an electrostatic adsorption module formed by a relay switch, a booster and an electrostatic adsorption film and a depth camera with an IMU sensor arranged inside are arranged in the platform main body, and the crawling robot realizes adsorption on the surface of the spacecraft based on the electrostatic adsorption principle. The control system adopts hardware such as an embedded microprocessor, an expansion interface board and the like, and realizes autonomous movement through perception, planning and control. Meanwhile, the device can be matched with the gait of the robot to control the adsorption force, so that the moment required by the robot to lift the legs is reduced. Finally, a dynamic simulation method is provided for calculating the required adsorption force of the robot and verifying the crawling capability of the robot. The invention realizes the adsorption, crawling, autonomous perception and movement of the robot on the surface of the spacecraft.)

1. A spacecraft surface crawling robot capable of autonomously sensing movement is characterized by comprising a mechanical structure and a control system;

the robot mechanical structure includes:

a. the robot comprises a robot bottom plate, a middle plate, a robot foot end mechanism, a robot foot, a robot control system and a robot control system, wherein the robot bottom plate is provided with a plurality of robot feet;

b. the robot platform comprises a robot platform main body consisting of a plurality of functional modules, a storage battery, an electrostatic adsorption module and a depth camera, wherein the storage battery is arranged on a robot middle plate and used for supplying power to a robot system; the robot platform main body is wrapped by a closed shell, and the solar sheet is arranged on the shell;

the robot control system includes:

a. the control system hardware consists of an embedded microprocessor and an expansion interface board, wherein the embedded microprocessor is used for receiving perception information and running a visual mapping algorithm and a path planning algorithm in real time; the expansion interface board is used for connecting the microprocessor and other functional hardware, and comprises a depth camera, a bus steering engine and an electrostatic adsorption module;

b. the control system has the main functions of sensing and reconstructing a surrounding environment map, carrying out path planning and six-foot inverse kinematics calculation, controlling the rotation angle of a steering engine of the robot, and simultaneously matching with the gait of the robot to control the adsorption force, so that the adsorption and autonomous movement of the robot on the surface of the spacecraft are realized.

2. The autonomous mobile sensing spacecraft surface crawling robot of claim 1, wherein the electrostatic adsorption module is a charge-discharge circuit consisting of an electrostatic adsorption film, a relay switch and a booster; the input of the booster is connected with the storage battery, the output of the booster is connected with six relay switches for controlling charging, the output of each relay switch for controlling charging is respectively connected with an electrostatic adsorption film and a relay switch for controlling discharging, and the other ends of the electrostatic adsorption film and the relay switch for controlling discharging are grounded; if the relay switch for controlling charging is closed and the relay switch for controlling discharging is opened, the adsorption film is charged, and if the relay switch for controlling charging is opened and the relay switch for controlling discharging is closed, the adsorption film is discharged.

3. The autonomous sensing and moving spacecraft surface crawling robot of claim 1, wherein the control system can realize charging and discharging of the adsorption film according to the gait requirements of the robot so as to control the adsorption force; before the robot lifts the legs, the adsorption film discharges, the electrostatic adsorption force disappears, the moment required by the robot for lifting the legs is reduced, and compared with other adsorption modes, the risk of desorbing the spacecraft due to vibration is reduced; before the robot falls on the leg, the adsorption film is charged, and the robot can quickly adsorb after falling on the leg.

4. The spacecraft surface crawling robot capable of autonomously sensing and moving according to claim 1, wherein a solar sheet is attached to a robot shell and used for collecting solar energy and converting the solar energy into electric energy to supplement power for a storage battery, so that the power supply requirement of the robot capable of autonomously working for a long time is guaranteed.

5. The autonomous perceptive mobile spacecraft surface crawling robot of claim 1, wherein the control system implements a sensory-decision-controlled closed-loop autonomous mobile control, and the implementation procedure comprises: the control system receives the RGB image, the depth image and the IMU information input by the depth camera, calculates to obtain a grid network map, and carries out robot pose estimation; running a path planning algorithm according to the map, the pose and the target information to obtain a collision-free optimal path; performing hexapod inverse kinematics calculation according to the path to obtain a steering engine rotation angle, controlling the steering engine to rotate, and feeding back to pose estimation;

firstly, calculating a grid network map, wherein a world coordinate point corresponding to a current depth image is required, and the calculation formula is as follows:

wherein u and v are arbitrary coordinate points in an image coordinate system, and u₀、v₀Is the central coordinate, x, of the image_w、y_w、z_wRepresenting three-dimensional coordinate points in a world coordinate system; z is a radical of_cA z-axis value representing the camera coordinates, i.e. the distance of the object to the camera; r and T are respectively a 3x3 rotation matrix and a 3x1 translation matrix of the external reference matrix; f is the focal length of the camera, and 1/dx and 1/dy respectively represent the number of pixel points in unit distance; when the origins of the camera coordinate system and the world coordinate system coincide, further simplification can result:

after point clouds corresponding to the depth images are obtained, converting the point clouds into voxel grid networks in an octree format, and obtaining grid network maps; the pose information is obtained by integrating the acceleration and the angular velocity acquired by the IMU sensor;

running a graph search path planning algorithm according to the map, the pose and the target information to search and obtain a collision-free optimal path reaching the target, and calculating the angle of each rotation of the steering engine through the six-foot inverse kinematics of the robot after obtaining the optimal path; the principle of the six-legged robot inverse kinematics calculation is as follows:

establishing a single-foot local coordinate system according to a D-H method, taking a point where a steering engine is connected with a robot main body platform as a starting point, and taking a conversion matrix of adjacent joint coordinate systems asThe standard form is:

wherein a is_i-1Represents the length of the previous joint; d_iRepresents the offset between joint i-1 and joint i; theta_iThe included angle of two joints, namely the joint angle is shown; alpha is alpha_i-1The torsion angle of the joint i-1 is represented and is an angle from the axis of the joint i-1 to the joint i around the common vertical line of the two joints; m represents the mth foot;

the D-H parameters for a single foot are shown in the following table:

bar number α_i-1 a_i-1 d_i θ_i 1 0 0 0 θ₁ 2 π/2 l₁ 0 θ₂ 3 0 l₂ 0 θ₃ 4 0 l₃ 0 0

Wherein the rod 1 is a rigid body formed by connecting legs and feet, a steering engine of the main platform of the robot and a connecting piece, and the length of the rod is l₁(ii) a The rod 2 is a rigid body composed of a fixed part and steering engines at two ends of the fixed part, and the length of the rod is l₂(ii) a The rod 3 is a foot end mechanism with a length of l₃(ii) a The rod 4 is a contact point of the foot end mechanism and the ground;

the conversion matrix from the foot end point coordinate system to the leg and foot and robot main body platform connecting point coordinate system is as follows:

the post-deployment calculation yields:

wherein c is_i＝cosθ_i，s_i＝sinθ_i，c_ij＝cos(θ_i+θ_j)，s_ij＝sin(θ_i+θ_j)；

According to the transformation matrix, the coordinate of a point, which is arranged under the global coordinate system and is connected with the robot main body platform, is (x)₀、y₀、z₀) The coordinates of the end points of the robot foot are (x, y and z), and the joint angle theta is solved_iThe inverse kinematics equation of (a) is as follows:

the joint angle is the angle of the robot steering engine which needs to rotate, and the control system is connected to the steering engine through the expansion interface board and then controls the steering engine to rotate by the corresponding angle, so that the complete autonomous movement control of perception-decision-control is realized.

6. An autonomous perceptive mobile spacecraft surface crawling robot according to claim 1, characterized in that the robot adsorption and mobility capabilities are dynamically simulated; calculating by using multi-body dynamics simulation software to obtain the adsorption force and the friction force required by the robot to move on the surface of the spacecraft, and simultaneously verifying the feasibility of the moving gait of the robot; the main simulation process is as follows:

(1) simplifying a robot model, only keeping a rotating joint and a main mass rigid body, and importing the model into multi-body dynamics simulation software;

(2) defining the rotating pair, the moving pair, the contact force with the wall surface, the adsorption force and the quality parameters of all parts according to the actual form of the robot;

(3) setting the simulation environment as a gravity-free environment to simulate a space environment;

(4) calculating the rotation angle of a steering engine according to the gait design of the robot and the inverse kinematics of the robot, and taking the rotation angle as a control function in simulation;

(5) applying a pretightening force to the foot end of the robot, then running simulation, calculating to obtain the friction force required by the foot end, and determining the area and voltage of the adsorption film according to the actual friction force performance of the electrostatic adsorption film;

(6) and observing whether the movement of the center of mass of the robot is carried out according to a preset moving direction and a preset step, and verifying the moving capability of the robot.

Technical Field

The invention belongs to the field of mobile robots, and particularly relates to a spacecraft surface crawling robot capable of autonomously sensing movement.

Background

The robot on-orbit service is a main component of space infrastructure and is the key for developing the space frontier. The on-orbit service concept was first proposed in the 60's of the 20 th century and then implemented in many of the key tasks of the last century. In fact, the robot on-orbit service is of great significance to the current aerospace industry of China, and with the deepening of the construction and operation work of the space station of China, the robot on-orbit service such as monitoring and maintenance of the outside of the space station is necessarily involved. These inspections and repairs, if done by the astronaut, can pose a risk to the health of the astronaut's life. And for a general spacecraft, a spacecraft is not equipped, and the spacecraft can be monitored and maintained only by relying on the on-orbit service of the robot.

The main applications of the robot on-orbit service include: in-orbit maintenance, spacecraft docking, in-orbit parking, in-orbit filling, in-orbit transportation, spacecraft external surface inspection, space rescue and removal of orbital debris.

The system has important significance for most spacecrafts such as on-orbit maintenance, spacecraft surface inspection, space rescue and the like. In fact, statistics have shown that in the last decade, on average 100 satellites (from 78 to 130) were transmitted per year, and that most of them do not present any significant problems in performing the mission. However, a small fraction of them have different degrees of anomalies or even failures, even in recent years in-orbit failures have outweighed launch failures. In the past, spacecraft could only be updated with software to maintain the performance of an anomalous spacecraft, and failure of a spacecraft would mean scrapping. This cumulatively results in billions of dollars of loss.

And the on-orbit maintenance provides a repair opportunity for hardware faults of the spacecraft, so that the service life of the spacecraft can be effectively prolonged. A typical example is a Habo Space Telescope (HST) that has undergone five repairs, including the replacement of the circuit board. The spacecraft surface inspection can evaluate the health condition of the spacecraft to find hidden risks, and the space rescue is to actively rescue and remove faults of the spacecraft which has failed and cannot respond. The on-orbit service can effectively reduce the failure rejection rate of the spacecraft and has good economic benefit.

However, the in-orbit service currently used is usually provided by one or more large mechanical arms which are arranged on the spacecraft, and the mode has the advantages that the mechanical arm has a large movement range and high freedom degree, and can perform various in-orbit service tasks such as capture, maintenance and the like. However, the mechanical arm is expensive in manufacturing, mounting and launching cost and difficult to control, only few spacecrafts have on-orbit service capability, and the mechanical arm mainly provides service (such as a space station) for large spacecrafts. Meanwhile, the on-orbit service based on the mechanical arm needs to dock and capture the spacecraft first, so that the difficulty is high for the non-cooperative spacecraft with unknown information, and the risk is high.

Another idea is to use a smart and small space crawling robot for on-orbit service. The robot capable of crawling on the surface of the spacecraft reduces the on-orbit service cost and difficulty, and can complete detection of key parts of the spacecraft and maintenance and replacement of fault parts. However, the existing crawling robot for space application needs manual operation to complete tasks, cannot autonomously sense the environment to perform collision-free path planning and movement, and limits the application occasions.

Disclosure of Invention

The invention aims to provide an autonomous sensing and moving spacecraft surface crawling robot aiming at the defects of the prior art, which can realize adsorption on the surface of a spacecraft and intelligent autonomous crawling, does not need manual control to avoid obstacles, and can be widely applied to on-orbit service tasks such as autonomous detection, autonomous maintenance and the like on the surface of the spacecraft; meanwhile, the electrostatic adsorption module is optimized to be matched with the gait of the robot for adsorption, so that the reliability of the robot is enhanced.

The purpose of the invention is realized by the following technical scheme: a spacecraft surface crawling robot capable of autonomously sensing movement comprises a mechanical structure and a control system;

the robot mechanical structure includes:

a. the robot comprises a robot bottom plate, a middle plate, a robot foot end mechanism, a robot foot, a robot control system and a robot control system, wherein the robot bottom plate is provided with a plurality of robot feet;

b. the robot platform comprises a robot platform main body consisting of a plurality of functional modules, a storage battery, an electrostatic adsorption module and a depth camera, wherein the storage battery is arranged on a robot middle plate and used for supplying power to a robot system; the robot platform main body is wrapped by a closed shell, and the solar sheet is arranged on the shell;

the robot control system includes:

a. the control system hardware consists of an embedded microprocessor and an expansion interface board, wherein the embedded microprocessor is used for receiving perception information and running a visual mapping algorithm and a path planning algorithm in real time; the expansion interface board is used for connecting the microprocessor and other functional hardware, and comprises a depth camera, a bus steering engine and an electrostatic adsorption module;

b. the control system has the main functions of sensing and reconstructing a surrounding environment map, carrying out path planning and six-foot inverse kinematics calculation, controlling the rotation angle of a steering engine of the robot, and simultaneously matching with the gait of the robot to control the adsorption force, so that the adsorption and autonomous movement of the robot on the surface of the spacecraft are realized.

Furthermore, the electrostatic adsorption module is a charge-discharge circuit consisting of an electrostatic adsorption film, a relay switch and a booster; the input of the booster is connected with the storage battery, the output of the booster is connected with six relay switches for controlling charging, the output of each relay switch for controlling charging is respectively connected with an electrostatic adsorption film and a relay switch for controlling discharging, and the other ends of the electrostatic adsorption film and the relay switch for controlling discharging are grounded; if the relay switch for controlling charging is closed and the relay switch for controlling discharging is opened, the adsorption film is charged, and if the relay switch for controlling charging is opened and the relay switch for controlling discharging is closed, the adsorption film is discharged.

Furthermore, the control system can realize the charging and discharging of the adsorption film according to the gait requirements of the robot so as to control the adsorption force; before the robot lifts the legs, the adsorption film discharges, the electrostatic adsorption force disappears, the moment required by the robot for lifting the legs is reduced, and compared with other adsorption modes, the risk of desorbing the spacecraft due to vibration is reduced; before the robot falls on the leg, the adsorption film is charged, and the robot can quickly adsorb after falling on the leg.

Furthermore, the solar piece is attached to the shell of the robot and used for collecting solar energy and converting the solar energy into electric energy to supplement electricity for the storage battery, and the power supply requirement that the robot can work independently for a long time is guaranteed.

Further, the control system realizes the sensing-decision-control closed-loop autonomous mobile control, and the realization process comprises the following steps: the control system receives the RGB image, the depth image and the IMU information input by the depth camera, calculates to obtain a grid network map, and carries out robot pose estimation; running a path planning algorithm according to the map, the pose and the target information (artificially designated targets) to obtain a collision-free optimal path; performing hexapod inverse kinematics calculation according to the path to obtain a steering engine rotation angle, controlling the steering engine to rotate, and feeding back to pose estimation;

firstly, calculating a grid network map, wherein a world coordinate point corresponding to a current depth image is required, and the calculation formula is as follows:

wherein u and v are arbitrary coordinate points in an image coordinate system, and u₀、v₀Is the central coordinate, x, of the image_w、y_w、z_wRepresenting three-dimensional coordinate points in a world coordinate system; z is a radical of_cA z-axis value representing the camera coordinates, i.e. the distance of the object to the camera; r and T are respectively a 3x3 rotation matrix and a 3x1 translation matrix of the external reference matrix; f is the focal length of the camera, and 1/dx and 1/dy respectively represent the number of pixel points in unit distance; when the origins of the camera coordinate system and the world coordinate system coincide, further simplification can result:

after point clouds corresponding to the depth images are obtained, converting the point clouds into voxel grid networks in an octree format, and obtaining grid network maps; the pose information is obtained by integrating the acceleration and the angular velocity acquired by the IMU sensor;

running a graph search path planning algorithm-A algorithm according to the map, the pose and the target information, searching to obtain a collision-free optimal path reaching the target, and calculating the angle of each rotation of the steering engine through six-foot inverse kinematics of the robot after obtaining the optimal path; the principle of the six-legged robot inverse kinematics calculation is as follows:

establishing a single-foot local coordinate system according to a D-H method, taking a point where a steering engine is connected with a robot main body platform as a starting point, and taking a conversion matrix of adjacent joint coordinate systems asThe standard form is:

wherein a is_i-1Represents the length of the previous joint; d_iRepresents the offset between joint i-1 and joint i; theta_iThe included angle of two joints, namely the joint angle is shown; alpha is alpha_i-1The torsion angle of the joint i-1 is represented and is an angle from the axis of the joint i-1 to the joint i around the common vertical line of the two joints; m represents the mth foot;

the D-H parameters for a single foot are shown in the following table:

bar number	αi-1	ai-1	di	θi
					1	0	0	0	θ1
2	π/2	l1	0	θ2
					3	0	l2	0	θ3
4	0	l3	0	0

Wherein the rod 1 is a rigid body formed by connecting legs and feet, a steering engine of the main platform of the robot and a connecting piece, and the length of the rod is l₁(ii) a The rod 2 is a rigid body composed of a fixed part and steering engines at two ends of the fixed part, and the length of the rod is l₂(ii) a The rod 3 is a foot end mechanism with a length of l₃(ii) a The rod 4 is a contact point of the foot end mechanism and the ground;

the conversion matrix from the foot end point coordinate system to the leg and foot and robot main body platform connecting point coordinate system is as follows:

the post-deployment calculation yields:

wherein c is_i＝cosθ_i，s_i＝sinθ_i，c_ij＝cos(θ_i+θ_j)，s_ij＝sin(θ_i+θ_j)；

According to the transformation matrix, the coordinate of a point, which is arranged under the global coordinate system and is connected with the robot main body platform, is (x)₀、y₀、z₀) The coordinates of the end points of the robot foot are (x, y and z), and the joint angle theta is solved_iThe inverse kinematics equation of (a) is as follows:

the joint angle is the angle of the robot steering engine which needs to rotate, and the control system is connected to the steering engine through the expansion interface board and then controls the steering engine to rotate by the corresponding angle, so that the complete autonomous movement control of perception-decision-control is realized.

Further, performing dynamic simulation on the adsorption and movement capacity of the robot; because the electrostatic adsorption force is in positive correlation with the area of the adsorption membrane and the voltage, and the adsorption force is in direct proportion with the area of the membrane under the condition of high-voltage fixation, the adsorption force needs to be reduced as much as possible within an allowable range, so that the occupied space of the adsorption membrane is reduced; in fact, under the space microgravity environment, a large adsorption force is not needed, but when the adsorption force provided by the high-voltage electrostatic adsorption film needs to overcome the influence of single leg lifting, attitude disturbance and micro-vibration on the adsorption reliability and stability of the robot, friction force is needed when the robot moves on a spacecraft, and the friction force provided by the adsorption force can meet the requirements in the classic gaits of advancing, turning and the like;

calculating by using multi-body dynamics simulation software to obtain the adsorption force and the friction force required by the robot to move on the surface of the spacecraft, and simultaneously verifying the feasibility of the moving gait of the robot; the main simulation process is as follows:

(1) simplifying a robot model, only keeping a rotating joint and a main mass rigid body, and importing the model into multi-body dynamics simulation software;

(2) defining the rotating pair, the moving pair, the contact force with the wall surface, the adsorption force and the quality parameters of all parts according to the actual form of the robot;

(3) setting the simulation environment as a gravity-free environment to simulate a space environment;

(4) calculating the rotation angle of a steering engine according to the gait design of the robot and the inverse kinematics of the robot, and taking the rotation angle as a control function in simulation;

(5) applying a pretightening force to the foot end of the robot, then running simulation, calculating to obtain the friction force required by the foot end, and determining the area and voltage of the adsorption film according to the actual friction force performance of the electrostatic adsorption film;

(6) and observing the movement of the center of mass of the robot, and verifying the movement capability of the robot according to the preset movement direction and step.

The invention has the beneficial effects that:

1. the invention relates to an autonomous movement control system designed for a crawling robot, which realizes sensing-planning-control closed-loop autonomous movement control, comprises a visual mapping algorithm, a path planning algorithm and inverse kinematics calculation, controls a steering engine to rotate, and finally realizes autonomous obstacle avoidance and movement on the surface of a spacecraft.

2. The robot structure uses the principle of electrostatic adsorption as an adsorption force source, so that the structural design difficulty of the robot is simplified; meanwhile, the electrostatic adsorption module adopts the design that one adsorption film is matched with one charging circuit and one discharging circuit, so that the control system is allowed to adjust the adsorption force according to the gait, the moment required by leg lifting is reduced, and the risk of vibration separation from the spacecraft is reduced.

3. The invention provides a dynamic simulation method for a crawling robot, which can quickly calculate the minimum adsorption force required by the actual robot when the robot crawls on the surface of a spacecraft and realize the selection of the optimal adsorption film area.

4. The invention integrates a booster, a relay switch, an embedded microprocessor, an expansion board, a storage battery and the like required by the space crawling robot into a robot main body with the size of 12cm by 10.5cm by 10cm, thereby reducing the manufacturing and transmitting cost and lowering the on-orbit service use threshold.

Drawings

Fig. 1 is an internal structure view of a crawling robot according to an embodiment of the present invention;

FIG. 2 is a structural diagram of a housing of a crawling robot provided by an embodiment of the invention;

FIG. 3 is a triangular gait schematic diagram of a hexapod robot

Fig. 4 is a schematic diagram of charging and discharging of an electrostatic adsorption film according to an embodiment of the present invention;

fig. 5 is a flow chart of autonomous movement control of the control system according to the embodiment of the present invention;

FIG. 6 is a diagram of a model of a single-foot kinematics provided by an embodiment of the present invention;

FIG. 7 is a diagram of a simulation model in Adams software according to an embodiment of the present invention;

FIG. 8 is a result of calculating a desired friction for a simulation according to an embodiment of the present invention;

fig. 9 is a result of calculating the displacement of the center of mass of the robot through simulation according to the embodiment of the present invention.

Detailed Description

The invention is described in further detail below with reference to the figures and specific examples.

The embodiment of the invention provides an autonomous sensing and moving spacecraft surface crawling robot, which comprises a mechanical structure and a control system;

as shown in fig. 1 and 2, the robot mechanical structure includes:

a. the robot comprises a robot body, a robot bottom plate 7, a robot middle plate 4, a robot foot end mechanism 1, a robot middle plate, a robot upper plate, a robot lower plate, a robot upper plate, a robot lower plate, a robot upper plates, a robot lower plate, a robot upper plate and a robot lower plate, wherein the six leg feet are formed by eighteen steering engines 6, are arranged between the robot bottom plate 7 and the robot middle plate 4, are distributed in a left-right-and symmetrical mode, one foot of the robot is formed by connecting piece 5 and one connecting piece and connecting piece 3 to connect the three steering machines, and form one foot end of the robot, and connect the robot foot end of the robot, and connect the robot foot end of the robot, connect the robot, and connect the robot, connect the robot foot end of the robot, connect the robot, and connect the robot, connect the foot end of the foot end mechanism 1, connect the robot, connect the foot end of the robot, connect the; in the embodiment, the steering engine 6 is a UART asynchronous half-duplex serial bus steering engine, two I/Os are arranged on the steering engine 6, the steering engines can be connected with each other through double interfaces, and finally one I/O is led out to be connected with the embedded microprocessor 11, so that the serial control of the steering engine is realized, and the occupation of a serial port on the embedded microprocessor 11 is simplified.

b. The robot platform comprises a robot platform main body consisting of a plurality of functional modules, a storage battery 10 arranged on a robot middle plate 4 and used for supplying power to a robot system, an electrostatic adsorption module consisting of an electrostatic adsorption film 15, a relay switch 8 and a booster 9, and a depth camera 13 arranged on a robot top plate 2, wherein an IMU sensor is arranged in the depth camera 13; the robot platform main body is wrapped by a closed shell 14, and a solar sheet 15 is arranged on the shell;

the robot control system includes:

a. the control system hardware is composed of an embedded microprocessor 11 and an expansion interface board 12, as shown in fig. 1, the embedded microprocessor is used for receiving perception information and running a visual mapping algorithm and a path planning algorithm in real time; the expansion interface board is used for connecting the microprocessor and other functional hardware, and comprises a depth camera, a bus steering engine and an electrostatic adsorption module;

b. the control system has the main functions of sensing and reconstructing a surrounding environment map, carrying out path planning and six-foot inverse kinematics calculation, controlling the rotation angle of a steering engine of the robot, and simultaneously matching with the gait of the robot to control the adsorption force, so that the adsorption and autonomous movement of the robot on the surface of the spacecraft are realized.

Specifically, the robot with the six-foot structure has the most efficient static movement characteristic compared with other foot type structures, when the robot uses a triangular gait as shown in fig. 3, three feet are taken as a group and alternately supported to advance, as shown in the figure, the three feet move to A ', B ' and C ' from A, B and C respectively, the robot can be ensured to move along a straight line, meanwhile, the three feet are taken as the supporting feet of the robot all the time, and the centroid of the robot falls between the triangles formed by the three feet, so that the robot is quite stable.

The rotation energy of the steering engine comes from a storage battery 10, the capacity of the storage battery 10 is 6000mAh, and the storage battery 10 supplies power for the robot system. The voltage of the secondary battery is converted from a direct current voltage of 7 v to a high voltage of 3 kv after passing through the booster 9, and then connected to the electrostatic adsorption film through a discharging and charging circuit as shown in fig. 4.

Fig. 4 shows a preferred implementation of the charging and discharging circuit, in which a 3 kv high-voltage input H is connected to electrostatic adsorption films M1, M2, M3, M4, M5, and M6 through relay switches S1, S3, S5, S7, S9, and S11, respectively, and the other end of the electrostatic adsorption film is grounded to G, which is a charging loop. The discharging circuit is formed by relay switches S2, S4, S6, S8, S10 and S12, electrostatic adsorption films M1, M2, M3, M4, M5, M6 and ground G.

When the relay switches S1, S3, S5, S7, S9, S11 are closed, the electrostatic adsorption film is charged to generate an adsorption force. When the relay switches S1, S3, S5, S7, S9, and S11 are opened and the relay switches S2, S4, S6, S8, S10, and S12 are closed, the electrostatic attraction film discharge attraction force disappears. Each relay is connected to the expansion interface board 12, so that the charging and discharging of each adsorption film can be independently controlled by independently controlling the on-off of the relay connected with the adsorption film, and the sufficient degree of freedom of the robot is ensured.

Whether the adsorption force exists or not is controlled through the charge-discharge loop, so that the moving gait of the robot can be matched in real time, and the adsorption force exists when the feet of the robot are in a supporting state; when the robot foot is in a moving state, the adsorption force disappears. The adsorption of the foot end to the spacecraft is realized based on the electrostatic adsorption principle, the moment required by the robot for lifting the legs is reduced, and the risk of desorbing the spacecraft due to vibration is reduced compared with other adsorption modes.

The autonomous movement control flow of the crawling robot is shown in fig. 5, and an Intel D435i depth camera is selected to simultaneously provide RGB image input, depth image input and IMU information input. Calculating a grid network map based on the depth image information, integrating the angular velocity and the acceleration measured by the IMU to obtain a pose estimation result, correcting IMU information errors by using RGB information, and artificially giving target information according to actual task requirements. The path planning algorithm adopts a heuristic A-x algorithm, which is a common path searching and graph traversing algorithm and has the advantages of complete probability and high searching speed. And performing robot inverse kinematics calculation after the optimal path is searched.

FIG. 6 is a kinematic model of a single foot of the robot, where l₁Is 57mm, l₂Is 78mm, l₃The rotating angle of the steering engine can be quickly obtained by combining the moving gait of the robot to carry out inverse kinematics calculation for 56 mm.

In order to calculate the minimum adsorption force required by the robot to adsorb on the surface of a spacecraft, a robot simplified model shown in fig. 7 is established in Adams multi-body dynamics simulation software, a rotating pair is defined for each steering engine rotating shaft, contact force and pre-pressure are defined between a foot end and a wall surface, and simulation of a straight moving gait is achieved according to the moving gait and inverse kinematics of the robot. The simulation contained seven movement cycles, each of which the robot moved 2cm forward. The frictional force required for the robot shown in fig. 8 to climb the wall surface was obtained, and (a) and (b) were frictional forces in the X and Y directions, respectively. And combining the calculation result with the performance of the adsorption film, and selecting the diameter of the adsorption film of the robot to be 60mm under the condition of high voltage of 3 kilovolts.

The dynamics simulation simultaneously verifies the moving performance of the robot, and the mass center moving of the robot in the main moving direction shown in the figure 9 is obtained. It can be seen that the robot mass center advances by 14cm in 7 movement cycles, and can stably perform a predetermined task.

The above are merely examples of the present invention, and are not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement and the like, which are not made by the inventive work, are included in the scope of protection of the present invention within the spirit and principle of the present invention.