CN117245911A - Aerial 3D printing robot and printing method thereof - Google Patents

Abstract

The invention discloses an aerial 3D printing robot and a printing method thereof, and relates to the technical fields of drones and 3D printing. The 3D printing robot includes a drone body, a mechanical arm, a 3D printing system and an automatic control system. The 3D printing system includes a system A feeding device and a damping printing nozzle are left. The damping printing nozzle includes a base shell. The upper part of the base shell is connected to a flange. The flange is provided with a through hole that matches the second feeding hose. The bottom of the base shell is connected to the funnel-shaped wind shield. A shell cover is provided at the junction of the base shell and the funnel-shaped wind shield. A laser ranging sensor is connected to the lower side of the shell cover. The present invention adopts the above-mentioned aerial 3D printing robot and its printing method, uses a drone as a carrier, gets rid of the space limitations of 3D printing, and is equipped with a tethered 3D printing device; it can realize additive manufacturing of large objects and buildings. Manufactured and can be used to repair walls and high-altitude objects.

Description

An aerial 3D printing robot and its printing method

Technical field

The present invention relates to the technical fields of drones and 3D printing, and in particular to an aerial 3D printing robot and a printing method thereof.

Background technique

Traditional 3D printing technology usually prints with a single degree of freedom in a fixed area, which limits the size and complexity of objects. In addition, traditional 3D printing materials are usually relatively small and suitable for manufacturing precision parts, but are not suitable for the manufacturing of large objects. As manufacturing technology continues to advance, there is an increasing need to be able to print in multiple degrees of freedom. For example, in the construction field, there is a need to be able to print in different directions to enable additive manufacturing of large buildings.

As a technology with great potential, additive manufacturing of buildings has attracted widespread attention in the construction field. However, the traditional construction process requires a lot of labor and time, and the use of 3D printing technology can achieve faster, more efficient and precise building manufacturing while reducing waste.

Construction robots equipped with robotic arm repair work mechanisms can replace construction workers to complete some high-temperature and hazardous-environment work, which can greatly reduce the incidence of accidents, reduce construction costs, and improve work efficiency. However, when repairing buildings, there are often the following technical difficulties:

(1) Impact of downwash airflow: In the process of aerial 3D printing by drones, the negative impact of downwash airflow on the printing process is a key technical challenge. This downwash airflow is generated by the rotating propeller of the drone, which may disturb the surrounding environment, thereby affecting the precise positioning, adhesion and accuracy of 3D printing materials, ultimately affecting the printing quality and performance of the final product.

In traditional 3D printing technology, the printing process takes place in a relatively stable environment, and materials are deposited on a fixed platform, making positioning and attachment relatively easy to control. However, aerial 3D printing introduces drone motion and airflow interference, making the printing process more complex.

(2) End nozzle collision problem: During the aerial 3D printing process, the control accuracy and error of the drone may cause the end nozzle to contact and collide with the surrounding environment (such as walls or the ground), which may cause serious problems, including Damaged nozzles, damaged print materials, and overall poor print quality.

Contents of the invention

The purpose of the present invention is to provide an aerial 3D printing robot and its printing method, in order to solve the two major problems of downwash airflow influence and end nozzle collision during aerial 3D printing. UAVs are used as carriers to get rid of the space limitations of 3D printing and are equipped with tethered 3D printing devices. The robot can implement additive manufacturing of large objects and buildings, and can be used to repair walls and high-altitude objects.

In order to achieve the above purpose, the present invention provides an aerial 3D printing robot, which includes a drone body, a mechanical arm, a 3D printing system and an automatic control system. The robotic arm is arranged on the lower side of the drone body, and the automatic control system The control system is installed on the upper side of the drone body;

The 3D printing system includes a tethered feeding device and a damping printing nozzle. The tethered feeding device includes a glue dispenser. The glue dispenser communicates with an airborne storage device provided on one side of the UAV body through a first feeding hose. The material bottles are connected, and the bottom of the airborne material storage bottle is provided with a quick-connect socket. The quick-connect socket is connected to the damping printing nozzle through a second feeding hose, and the other end of the dispensing machine is connected through a third feeding hose. Connected to the feeder; the upper part of the airborne material storage bottle is equipped with an upper fixed ring, and the lower part of the airborne material storage bottle is equipped with a lower fixed ring;

The damping printing nozzle includes a base shell. The upper part of the base shell is connected to a flange. The flange is provided with a through hole matching the second feeding hose. The bottom of the base shell is connected to the base shell. The funnel-shaped windshield is connected to each other. A shell cover is provided at the junction of the base shell and the funnel-shaped windshield. A laser ranging sensor is connected to the lower side of the shell cover. A spring is provided inside the base shell. The spring is sleeved on the nozzle. The upper end of the nozzle is provided with a quick-plug connector. The bottom end of the nozzle is provided with a film pressure sensor. The nozzle passes through the middle of the shell cover, and the bottom of the nozzle is located at the bottom of the windshield. lower side.

Preferably, the robotic arm includes a Delta robotic arm and a three-axis serial robotic arm. The Delta robotic arm includes a top plate connected to the UAV body. A steering gear is provided on the underside of the top plate, and a steering gear is arranged on the outside of the steering gear. Steering gear housing, the output shaft of the steering gear is connected to the upper end of the upper arm through the shaft column, the lower end of the upper arm is connected to the upper end of the lower arm, and the lower end of the lower arm is connected to the annular chassis;

The three-axis series manipulator includes a base joint connected to the bottom of the annular chassis, an elbow joint is connected to the lower side of the base joint, and a wrist joint is connected to the lower side of the elbow joint.

Preferably, the upper side of the flange is connected to the bottom of the wrist joint.

Preferably, the automatic control system includes a hardware part, which includes an airborne computer NUC, a PX4 flight control, an airborne control board and a ground control board. The flight control PX4 is responsible for the flight control of the drone, and the airborne control board Responsible for the control and data processing of the robotic arm and damping printing nozzle. The ground control panel controls the ground dispensing machine. The flight control PX4 and the airborne control panel communicate with the airborne computer NUC through USB. The ground control panel communicates with the airborne computer through wireless Bluetooth. Computer NUC connection communication.

Preferably, a visual camera is also provided on the side of the drone body.

The invention also provides an aerial 3D printing method, which includes the following steps:

S1. Obtain environmental information through the vision camera. The environmental information includes the position information of the target printing surface and surrounding obstacle information;

S2. By analyzing the motor speed and using the airflow analysis algorithm, calculate the V-shaped area of the downwash airflow, and control the robotic arm to position the damped printing nozzle in the V-shaped area of the downwash airflow;

S3. Obtain the distance between the end of the damping printing nozzle and the target printing surface through the laser ranging sensor, and obtain the pressure between the end of the damping printing nozzle and the target printing surface through the film pressure sensor;

S4. Based on the 3D model to be printed, comprehensively use the position information obtained by RTK and the environmental information detected by the vision camera, combined with the distance information obtained by the laser ranging sensor obtained in step S3, and the pressure information obtained by the film pressure sensor. Step S2 The V-shaped area position information of the downwash airflow obtained from the system is used to calculate the optimal printing path that adapts to the current environment based on the path planning algorithm.

Preferably, the airflow analysis algorithm includes:

S21. Obtain the speed information of the drone motor through the flight control system;

S22. Calculate the speed of the downwash airflow. The calculation formula is as follows:

;

in, represents the speed of the downwash airflow; N is the rotation speed of the drone motor; ρ is the air density in the current environment; c is the performance coefficient describing the blade when generating airflow, taking into account the shape and aerodynamic characteristics of the blade; K is the coefficient;

S23. Determine the direction of the downwash airflow based on the position, direction and angle of the drone's blades;

S24. According to the speed of the downwash airflow calculated in step S22, the direction of the downwash airflow calculated in step S23, and the geometry of the motor rotor, calculate the V-shaped area formed by the downwash airflow through the following formula:

;

Among them, h(x) represents the height of the downwash airflow, that is, the distance from the drone blades to the downwash airflow; D is the diameter of the motor rotor; x is the distance in the horizontal direction from the center position of the motor rotor;

S25. Combined with the current position and attitude information of the drone, by calculating the angles of each joint of the robotic arm, adjust the position of the end nozzle to above the V-shaped area of the downwash airflow to reduce the interference of the airflow on the material.

Preferably, the path planning algorithm includes:

S31. Data acquisition:

(1) Extract key information from the 3D model to be printed. Key information includes starting point, end point and obstacle information;

(2) Use the RTK system to obtain the location information of the drone;

(3) The vision camera detects the surrounding environment, including the position and posture of obstacles;

(4) Call the airflow analysis algorithm to calculate the position and range of the V-shaped optimal printing area formed by the downwash airflow;

S32. Combine the data obtained by the vision camera, laser ranging sensor, and film pressure sensor to construct an environment map and identify obstacles and safe areas;

S33. Calculate the spatial distance from the starting point to the end point through the distance measurement algorithm, and use the A* path search algorithm to search for the path from the starting point to the end point in the environment map; during the search process, consider avoiding obstacles and keeping the path as close as possible. Within the V-shaped area;

Distance measure:

;

Among them, d represents the distance; (x_1,y_1,z_1) is the coordinates of the starting point; (x_2,y_2,z_2) is the coordinates of the target point;

A* path search algorithm:

;

Among them, h(node) is a heuristic function used to estimate the cost from the node node to the target position; xtarget, ytarget and ztarget are the coordinates of the target position; xnode, ynode and znode are the coordinates of the node node ;

S34. According to the objective function, comprehensively consider the length of the path, the distance to the obstacle, the distance to the end nozzle and the pressure data, and evaluate and sort the paths obtained by searching;

;

Among them, F (path) is the objective function to evaluate the comprehensive performance of the path; w ₁ , w ₂ , w ₃ , and w ₄ are weight coefficients; obstacle_avoidance is a measure related to avoiding obstacles; nozzle_distance represents the distance between the end nozzle and the target printing surface ; pressure_data represents the pressure data exerted by the end nozzle; length(path) represents the path length;

S35. If the evaluation finds that there is a better path, perform path optimization and smooth the path through interpolation method;

S36. During the printing process, monitor the data of the laser ranging sensor and the film pressure sensor in real time. If the data exceeds the threshold, adjust the posture of the robot arm and adjust the end position of the damping printing nozzle. The adjustment formula for the end of the damping printing nozzle is as follows:

;

Among them, new_position represents the new position of the damping printing nozzle after adjustment; current_position represents the current position of the damping printing nozzle; Δposition represents the position change amount of the damping printing nozzle that needs to be adjusted.

Therefore, the present invention adopts the above-mentioned aerial 3D printing robot and its printing method, and its technical effects are as follows:

(1) Solve the problem of downwash airflow affecting printing. The 3D printing robot needs to be close to the target operating area. The air flow blown down by the drone's blades will blow away the materials. At the same time, it will cause ground effect when flying close to the ground. In the present invention, a funnel-shaped windshield is installed at the end of the nozzle to guide the blown airflow to the surroundings to form a stable airflow environment under the windshield. Through the aerial 3D printing method, the robotic arm is controlled so that the end nozzle is in a V-shaped area with weak downwash airflow. The combination of the two prevents the downwash airflow and ground effect caused by the rotation of the blades from blowing away the 3D printing material, protecting the 3D printing material. Printing proceeds normally;

(2) Solve the problem of collision between the nozzle end and the target surface. Set up a damping printing nozzle. The end nozzle adopts a spring damping design. During the printing process, the nozzle is very close to the ground, and there will be errors in the robot's control. This damping design can protect the nozzle when it hits the ground; during the 3D printing process, The laser ranging sensor measures the distance between the bottom of the nozzle and the target printing surface and is used for data input of the printing control algorithm and anti-collision algorithm: the film pressure sensor measures the contact pressure at the bottom of the nozzle and is used for data input of the anti-collision algorithm through aerial 3D printing Method to prevent the collision between the end nozzle and the target surface;

(3) The design of the quick-plug interface has the advantage of being easy to disassemble, and the storage tube can be easily replaced;

(4) The tethered extrusion method can solve the problem of 3D printing material life and power source. The solution of using an airborne motor has too small pressure and the motor itself is too heavy;

(5) 3D printing materials are specially configured materials, and their solidification time can be adjusted according to the proportion;

(6) Through the aerial 3D printing method, the problem of downwash air blowing away materials and the collision of the nozzle end with the target surface during the printing process is solved. UAVs can replace workers to complete operations in high-risk environments, avoid high-altitude accidents, and improve work efficiency. efficiency.

The technical solution of the present invention will be further described in detail below through the accompanying drawings and examples.

Description of drawings

Figure 1 is a schematic structural diagram of an aerial 3D printing robot;

Figure 2 is a partial front view of the aerial 3D printing robot;

Figure 3 is a cross-sectional view of the damping nozzle;

Figure 4 is a flow chart of the aerial 3D printing method;

Figure 5 is the algorithm diagram of the membrane pressure sensor;

Figure 6 is a diagram of the airflow analysis algorithm;

Figure 7 is a diagram of the path planning algorithm.

Reference signs

1. UAV body;

2. Robotic arm; 21. Delta robotic arm; 211. Top plate; 212. Steering gear; 213. Steering gear housing; 214. Upper arm; 215. Lower arm; 216. Ring chassis; 22. Three-axis series mechanical arm; 221. Base joint; 222, elbow joint; 223, wrist joint;

3. 3D printing system; 31. Tethered feeding device; 311. Dispensing machine; 312. First feeding hose; 313. Airborne storage bottle; 314. Second feeding hose; 315. Third feeding hose ; 316. Feeder; 317. Upper fixed ring; 318. Lower fixed ring; 32. Damping printing nozzle; 321. Base shell; 322. Flange plate; 323. Funnel-shaped windshield; 324. Shell cover; 325 , laser ranging sensor; 326, spring, 327, nozzle; 328, film pressure sensor;

4. Automatic control system; 401. Airborne computer NUC; 402. PX4 flight control; 403. Airborne control panel; 404. Ground control panel.

Detailed ways

The technical solution of the present invention will be further described below through the drawings and examples.

Unless otherwise defined, technical terms or scientific terms used in the present invention shall have the usual meaning understood by a person with ordinary skill in the field to which the present invention belongs.

Embodiment 1

As shown in Figure 1, an aerial 3D printing robot includes a drone body 1, a robotic arm 2, a 3D printing device 3 and an automatic control system 4. This embodiment uses a quad-rotor drone as a carrier to achieve aerial 3D printing tasks. There is also a visual camera (not shown in the picture) on the side of the drone body, which is used to obtain surrounding environment information.

As shown in Figure 2, the robotic arm 2 consists of a Delta robotic arm 21 and a three-axis series robotic arm 22, forming a six-degree-of-freedom control platform. The Delta manipulator 21 and the three-axis tandem manipulator 22 constitute a posture movement unit. The top end of the Delta manipulator 21 is connected to the UAV body 1, and the bottom end is connected to the three-axis tandem manipulator 22; the manipulator unit has a drive unit. Movement drive servo. Both robotic arm structures are common technical means in the field, and their control methods have been optimized through research on 3D printing scenarios.

The design structure of the Delta robotic arm 21 includes a top plate 211, a steering gear housing 213, a steering gear 212, an upper arm 214, a lower arm 215 and an annular chassis 216. The top plate 211 serves as the base of the Delta robotic arm 21 and is connected and fixed with the bottom of the UAV body 1 through nylon studs. The top plate 211 provides a stable support and connection platform for the Delta robotic arm 21.

There are three delta robotic arms. Three servo housings 213 are respectively fixed on the top plate 211. A servo 212 is installed inside the servo housing 213. The servo housing 213 is used to carry and fix the servo 212 to ensure the accuracy of the movement of the robotic arm. sex. Each steering gear 212 is connected to the upper arm 214 through its output shaft. The upper arm 214 is connected to the steering gear output shaft through an aluminum alloy shaft column. The lower arm 215 is connected to the upper arm 214 through a ball hinge. The annular chassis 216 is located at the bottom of the Delta robotic arm 21 , connected to the lower end of the lower arm 215 through a ball hinge.

The structure of the three-axis serial manipulator 22 has three rotating joints, namely a base joint 221, an elbow joint 222 and a wrist joint 223. These joints work together to achieve complex positioning and movement. The base joint 221 (first axis) is located on the bottom surface of the annular chassis 216 and is connected to the fixed base of the robotic arm. It allows the robotic arm to rotate around a vertical axis in a horizontal plane and is used to control the rotation direction of the robotic arm. The elbow joint 222 (second axis) is connected to the end of the base joint 221 to allow the robot arm to rotate in a plane perpendicular to the base joint 221. This joint allows the arm to raise and lower vertically. The wrist joint 223 (third axis) is connected to the end of the elbow joint 222, allowing the robot arm to rotate in a plane perpendicular to the elbow joint 222. The wrist joint 223 controls the positioning and direction of the end effector of the robotic arm.

Through the combination of the Delta manipulator 21 and the three-axis series manipulator 22, six degrees of freedom control of the posture movement unit is achieved. The delta manipulator 21 is responsible for movement at X, Y, and Z positions, and the three-axis series manipulator 22 is responsible for movement at pitch, roll, and yaw angles. This combination can simultaneously improve the stability of end printing and meet the printing needs in different angle scenarios, especially suitable for solving situations where walls have different tilt angles.

In the design of the UAV body, a quad-rotor UAV was chosen as the carrier platform, which is an advanced technology widely used in this field. The core of the UAV carrier is the onboard computer NUC401 and PX4 flight control 402. The onboard computer NUC401 is responsible for the computing tasks of upper-layer applications, communications, planning and perception algorithms, which provides a solid foundation for the intelligence and efficiency of the system. PX4 flight control 402 focuses on calculating the control algorithm of the drone to ensure the safety and stability of the flight.

The onboard computer NUC401 and the PX4 flight control 402 are directly connected via USB and achieve efficient communication through the mavlink protocol. In addition, a 10000mAh6S battery is selected as the energy supply, and power tethering technology can be used as needed to meet the power needs of different flight missions.

The choice of drones brings multi-dimensional flexibility to aerial 3D printing robots. Its vertical take-off and landing and hover capabilities allow it to fly freely in three-dimensional space, allowing it to easily reach different heights and locations. This feature effectively overcomes the space limitations of traditional 3D printing machines and facilitates a wider range of printing tasks.

The 3D printing system 3 is a key component, including a tethered feeding device 31 and a damped printing nozzle 32, which is designed to achieve accurate supply and fine ejection of materials.

The tethered feeding device 31 includes a feeder 316, a dispensing machine 311, a first feeding hose 312, a second feeding hose 314, a third feeding hose 315 and an onboard storage bottle 313. These components work closely together to ensure Accuracy and stability of material supply. The feeder 316 and the dispensing machine 311 are powered by a 220V power supply and placed on the ground. The two are connected through a third feeding hose 315. The feeder 316 is responsible for providing the required pressure, while the dispensing machine 311 can control the pressure. Size and extrusion switches. The two are connected to the airborne material storage bottle 313 on the drone body 1 in a tethered manner through the feeding hose, so that the material flows from the airborne material storage bottle 313 to the damping printing nozzle 32 .

The onboard storage bottle 313 is a cylindrical container, the upper end of which is equipped with threads and can be installed with a lid. The lower end is conical in shape, and a small outlet extends from the end to serve as a discharge port. Through this structure, the second feeding hose 314 can be connected to the damping printing nozzle 32 below through the quick-plug interface of the discharge port. The airborne storage bottle 313 is fixed on the side of the drone body through two 3D printed fixing rings. The outer diameter of the upper fixing ring 317 is slightly larger than that of the airborne material storage bottle 313, and the airborne material storage bottle 313 passes through the upper fixing ring 317 from bottom to top, thereby preventing tipping. The lower fixed ring 318 is slightly smaller than the outer diameter of the airborne material storage bottle 313, and supports the bottom of the airborne material storage bottle 313 to maintain stability. The top cover of the onboard storage bottle 313 has a quick-plug interface, which can be connected to the dispensing machine 311 on the ground through the second feeding hose 314. Its operation buffers the material during the material supply process to ensure that the material can flow evenly when it is extruded from the printing nozzle, thereby ensuring printing quality.

The ground control panel 404 and the feeder 316 are electrically connected through relays. The ground control panel 404 can control the switch of the feeder and the pressure of the dispensing machine 311. The control panel also communicates wirelessly with the onboard computer NUC401 in the automatic control system via Bluetooth to achieve remote control. In this way, after receiving the control instructions from the onboard computer NUC401, the microcontroller can control the on and off of the relay, thereby controlling the switch of the feeder 316 and realizing real-time remote control of the operation of the 3D printing device.

The damped printing nozzle 32 is a key component in the 3D printing system. As shown in Figure 3, the damping printing nozzle 32 includes a base shell 321, a flange 322, a shell cover 324, a nozzle 327, a spring 326, a funnel-shaped wind shield 323, a quick-plug connector, a film pressure sensor 328 and a laser ranging Sensor 325. The base shell 321 is the main structure of the damping printing that damps the printing nozzle 32 and has multiple important functions. A flange 322 is provided on the top for assembly and connection between the damping printing nozzle 32 and the three-axis series robotic arm to achieve precise positioning. There is a small hole in the middle for the feeding hose to pass through and the printing material is introduced into the nozzle 327. Below the flange 322 is a cylindrical shell body with a length of 70 mm and an inner diameter of 19 mm. The bottom has a matching hollow shell cover 324 to make the entire nozzle structure closed and stable.

The nozzle 327 is located in the base shell 321, and the end just passes through the hollow shell cover. The top of the nozzle 327 is a quick-plug interface for inserting the feeding hose. The main part of the nozzle 327 is conical in design, and the inner diameter of the discharge port is larger than the feed port. Small, this design helps guide flow so material flows out better, ensuring even spraying. The spring 326 is located in the base shell 321 and is coated on the top of the nozzle 327. The upper end of the spring 326 is in contact with the base shell 321, and the lower end is in contact with the outer ring of the nozzle 327. The existence of the spring 326 effectively reduces the vibration of the nozzle and improves printing stability. The length of the spring 326 plus the length of the outer ring of the nozzle 327 makes it slightly longer than the base shell, thereby forming a damping effect and slowing down the up and down movement of the nozzle. The inner diameter of the cylindrical shell is slightly larger than the outer diameter of the outer ring of the nozzle. The hollow shell cover 324 forms a limit so that the nozzle 327 can move up and down in the base shell 321 to prevent the nozzle 327 from contacting the ground during the printing process. Damage to the device. The funnel-shaped windshield 323 is located at the bottom of the nozzle and has a diameter of 120mm. It protects the ejected material from the interference of motor downwash airflow and ground effect, ensuring the stability of printing.

The film pressure sensor 328 is installed at the end of the nozzle 327 for monitoring the contact pressure at the bottom of the nozzle 327 so as to adjust the position of the nozzle and protect the nozzle when necessary. The laser ranging sensor 325 is installed next to the nozzle 327 and is used to measure the distance between the nozzle 327 and the printing surface, thereby achieving precise control of the printing height and position. These two sensors are connected to the onboard control board 403 to achieve real-time acquisition of sensor measurement data, thereby ensuring stability and accuracy during the printing process.

The 3D printing material uses a mixture of polydimethylsiloxane 1700 and 184. Its viscosity and plasticity can be adjusted by changing the ratio. The mixing ratio of 4:1 is suitable for the 4mm feeding hose used in this example.

The automatic control system 4 combines multiple hardware and software components to realize the autonomous control and operation of drones and 3D printing devices. The hardware part of the system includes the airborne computer NUC401, PX4 flight control 402, airborne control board 403 and ground Control board 404, the software part of the system includes communication, control, application, planning and sensing algorithms. The onboard computer NUC401 is responsible for high-level mission planning, perception algorithms and application execution. This is the core decision-making unit of the system and will achieve autonomous decision-making and control based on sensor data and external instructions. PX4 flight control 402 is responsible for the flight control and attitude adjustment of the drone, which is a key part to ensure the stable flight of the drone. It receives data from sensors and controls the motor to maintain the required flight state. The onboard control board 403 is responsible for the control and data processing tasks of the robotic arm and nozzle, including the movement of the robotic arm, the control of the nozzle, and the collection and processing of sensor data. The ground control panel 404 controls the operation of the ground feeder and the glue dispensing machine, and starts and stops the glue dispensing machine and adjusts related parameters.

Embodiment 2

As shown in Figure 4, an aerial 3D printing method can solve the two major problems of material blown by downwash airflow and collision between the nozzle end and the target surface during the printing process. This method controls the robotic arm to keep the end nozzle in a V-shaped area with weak downwash airflow, and uses the laser distance sensor to measure the distance between the end nozzle and the target printing surface and the end nozzle pressure data measured by the film pressure sensor to prevent the end nozzle. In collision with the target surface, the target information is recognized by the vision camera to plan the printing path, and finally the position of the end nozzle is controlled to complete the aerial 3D printing work.

The specific steps of the aerial 3D printing method are as follows:

S1. Environmental perception and analysis: Comprehensive perception of the surrounding environment through the use of a series of environmental sensors, including laser ranging sensors, film pressure sensors and visual perception systems. The laser distance sensor is used to measure the distance between the end nozzle and the target printing surface, and the film pressure sensor is used to measure the pressure exerted by the end nozzle on the target surface. At the same time, the vision camera can identify the target printing location and surrounding obstacles. The data acquired by these sensors are processed and analyzed by specific algorithms to provide accurate information on the state of the environment.

S2. Airflow analysis: Conduct a detailed analysis of the intensity and direction of the downwash airflow. Since the position, direction, and angle of the UAV blades are fixed, and the motor speed and airflow intensity show a clear correspondence, the downwash airflow will form a cone-shaped expansion pattern downward from the motor rotor. In this pattern, a V-shaped area is formed above the intersection of the two airflows, which is the most suitable area for operation. Through detailed analysis of the motor speed, the V-shaped area of the downwash airflow is determined, and the end nozzle is placed in this area, thereby effectively reducing the interference of the downwash airflow on the material.

S3. Collision protection: Use the distance data obtained by the laser ranging sensor to monitor the distance between the end nozzle and the target printing surface in real time. During the printing process, precise robotic arm control is used to fine-tune the position of the end nozzle, thereby accurately controlling the printing distance and ensuring a safe distance from the target surface. In this way, the collision between the end nozzle and the target surface is effectively prevented, ensuring the smooth progress of the printing process.

S4. Path planning: Based on the 3D model to be printed, comprehensively use the position information obtained by RTK and the environmental information detected by the visual camera, combined with the distance information of the laser ranging sensor, the pressure information of the film pressure sensor, and the downwash airflow obtained in the above steps. With relatively weak V-shaped area position information, the path planning algorithm calculates the optimal printing path that adapts to the current environment, ensuring the safety and stability of the printing path.

S5. Real-time control and adjustment: During the entire printing process, changes in sensor data and visual perception information are continuously monitored. According to the preset algorithm decision, the attitude of the robot arm and the position of the end nozzle can be flexibly adjusted. Such real-time adjustments can not only ensure the maintenance of printing quality, but also effectively avoid potential collision risks, keeping the entire printing process in a safe and high-quality state.

Among them, the specific processes of the film pressure sensor, air flow analysis algorithm, and path planning algorithm are as follows:

(1) Thin film pressure sensor fitting algorithm

As shown in Figure 5, it is the algorithm diagram of the membrane pressure sensor.

Data collection: First, collect a series of different force values known to be exerted on the membrane pressure sensor, and record the corresponding sensor output values. This data will be used to train the algorithm.

Data preprocessing: Preprocess the collected data, including noise removal, normalization and smoothing. This will help improve the accuracy and stability of the fitting algorithm.

Fitting model selection: Select a mathematical model of the relationship between the output of the membrane pressure sensor and the actual applied force using a one-variable linear model.

;

Among them : y is the dependent variable (predicted value) ; dependent variable value.

Fitting algorithm: Use mathematical models to fit preprocessed data. Use regression analysis to find the best fitting parameters so that the model can accurately predict the relationship between sensor output and actual applied force:

;

Among them: y is the dependent variable (predicted value); x is the independent variable (input feature); β ₀ is the intercept, indicating the dependent variable value when the independent variable is 0; β ₁ is the slope, indicating the effect of the independent variable on the dependent variable The degree of influence; ε is the error term, which represents random errors that cannot be fully explained by the model;

Model validation: Use a part of the data that has not been used in training for model validation. Evaluate the accuracy and generalization ability of the fitting algorithm by comparing the sensor output predicted by the model to the actual measurements.

Adjustment and optimization: If the verification results are not satisfactory, you can adjust the selection of the fitting model or the algorithm parameters, and retrain and verify to obtain better performance.

Real-time application: Embed the fitted algorithm into the data processing flow of the membrane pressure sensor. In practical applications, the output measured by the sensor will be converted into the actual applied force value through the algorithm, and the resulting force value will eventually be input into the aerial 3D printing method.

(2) Air flow analysis algorithm

Figure 6 shows the air flow analysis algorithm diagram. The specific analysis process is as follows:

First, obtain the motor speed information: Get the speed information of the drone motor, which can be obtained through the flight control system or sensor.

Calculate the downwash airflow speed: Calculate the intensity of the downwash airflow based on the formula of motor speed and related airflow properties. This may involve factors such as air density, blade shape, etc.

;

in, represents the speed of the downwash airflow; N is the rotation speed of the drone motor; ρ is the air density in the current environment, which is affected by factors such as temperature and air pressure; c is the performance coefficient describing the blade when generating airflow, taking into account the propeller Leaf shape and aerodynamic properties. K is the coefficient.

Analyze the direction of airflow: Determine the main direction of the downwash airflow based on the position, direction and angle of the drone's blades.

Determine the V-shaped area: Based on the calculated airflow intensity and direction, as well as the geometry of the motor rotor, the V-shaped area formed by the downwash airflow is determined through mathematical models and geometric calculations.

;

Among them, h(x) represents the height of the downwash airflow, that is, the distance from the UAV blades to the downwash airflow; D is the diameter of the motor rotor; x is the distance in the horizontal direction from the center of the motor rotor.

Adjust the nozzle position: Combining the current position and attitude information of the drone, by calculating the angles of each joint of the robotic arm, adjust the position of the end nozzle to above the V-shaped area of the downwash airflow to reduce the interference of the airflow on the material.

(3) Path planning algorithm

Figure 7 shows the path planning algorithm diagram.

data collection:

Extract key information from the 3D model to be printed, such as starting point, end point and obstacle information.

Use the RTK system to obtain the precise location information of the drone.

The visual perception system detects the surrounding environment, including the position and posture of obstacles.

The above airflow analysis algorithm is called to calculate the position and range of the V-shaped optimal printing area formed by the downwash airflow.

Environmental analysis: Combining data from the visual perception system and data from environmental sensors to build an environment map and identify obstacles and safe areas.

Path search: Calculate the spatial distance from the starting point to the end point through the distance measurement algorithm, and use the A* path search algorithm to search for the path from the starting point to the end point in the environment map. During the search process, consider avoiding obstacles and keeping the path within the V-shaped area as much as possible.

Distance measure:

;

Among them, d represents the distance; (x_1, y_1, z_1) is the coordinates of the starting point; (x_2, y_2, z_2) is the coordinates of the target point;

Path search algorithm:

;

Among them, h(node) is a heuristic function used to estimate the cost from the node node to the target position; xtarget, ytarget and ztarget are the coordinates of the target position; xnode, ynode and znode are the coordinates of the node node ;

Path evaluation: According to the objective function, the path length, distance to obstacles, end nozzle distance and pressure data are comprehensively considered to evaluate and sort the searched paths.

;

Among them, F(path) represents the objective function for evaluating the comprehensive performance of the path; length(path) represents the path length; w ₁ , w ₂ , w ₃ , and w ₄ are weight coefficients; obstacle_avoidance is a measure related to avoiding obstacles; nozzle_ distance represents the distance between the end nozzle and the target printing surface; pressure_data represents the pressure data exerted by the end nozzle, which is used to prevent the collision between the end nozzle and the target surface. The smaller the pressure data, the safer it is.

Path optimization: If the evaluation finds that there is a better path, path optimization can be performed to smooth the path through interpolation and other methods to avoid sharp angles and instability.

Real-time adjustment: During the printing process, the data of the laser distance sensor and film pressure sensor are monitored in real time. If the data exceeds the threshold, that is, the end nozzle is too close or the pressure is too high, the robot arm posture is adjusted immediately and the end nozzle position is adjusted to avoid collisions and maintain a stable printing process.

;

Among them, new_position represents the new position after adjustment; current_position is the position of the current end nozzle; Δposition is the position change that needs to be adjusted.

In addition, the specific process of aerial 3D printing of the present invention is as follows:

(1) File sending and starting: The user sends the object file to be printed to the onboard computer NUC, and then starts the drone, feeder, dispensing machine and ground control panel. These devices are a key part of enabling aerial 3D printing.

(2) Printing path planning: The onboard computer NUC calculates the incoming item file, plans the printing path, and determines the movement trajectory of the 3D printing robot in the air.

(3) Flight control and positioning: The onboard computer NUC sends instructions to the PX4 flight control, which uses visual perception and other technologies to make the drone fly above the target printing area and ensure accurate positioning and flight control.

(4) Robotic arm adjustment: The onboard computer NUC sends instructions to the onboard control panel to control the rotation of the delta robot arm servo to adjust the position of the end series robot arm so that the nozzle is close to the construction area. Adjust the direction of the nozzle by controlling the rotation of the tandem robot arm motor to ensure that the nozzle is located on the normal line of the plane of the desired repair area.

(5) Printing material extrusion: The onboard computer NUC sends instructions to the ground control panel to start the feeder and dispensing machine. Through the buffering of the storage bottle, the 3D printing material is evenly extruded from the nozzle to achieve layer-by-layer printing of objects.

(6) Complete printing: Under the synergy of drones, robotic arms, 3D printing systems and automatic control systems, the entire 3D printing process is completed layer by layer according to the planned path.

Therefore, the present invention adopts the above-mentioned aerial 3D printing robot and its printing method, uses a drone as a carrier, gets rid of the space limitations of 3D printing, and is equipped with a tethered 3D printing device; this robot can realize printing of large objects and buildings. Perform additive manufacturing and can be applied to repair walls and high-altitude objects.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that: The technical solution of the present invention may be modified or equivalently substituted, but these modifications or equivalent substitutions cannot cause the modified technical solution to depart from the spirit and scope of the technical solution of the present invention.

Claims (8)

1. An aerial 3D printing robot, characterized in that it includes a drone body, a robotic arm, a 3D printing system and an automatic control system. The robotic arm is arranged on the lower side of the drone body, and the automatic control system is configured On the upper side of the drone body; The 3D printing system includes a tethered feeding device and a damping printing nozzle. The tethered feeding device includes a glue dispenser. The glue dispenser communicates with an airborne storage device provided on one side of the UAV body through a first feeding hose. The material bottles are connected, and the bottom of the airborne material storage bottle is provided with a quick-connect socket. The quick-connect socket is connected to the damping printing nozzle through a second feeding hose, and the other end of the dispensing machine is connected through a third feeding hose. Connected to the feeder; the upper part of the airborne material storage bottle is equipped with an upper fixed ring, and the lower part of the airborne material storage bottle is equipped with a lower fixed ring; The damping printing nozzle includes a base shell. The upper part of the base shell is connected to a flange. The flange is provided with a through hole matching the second feeding hose. The bottom of the base shell is connected to the base shell. The funnel-shaped windshield is connected to each other. A shell cover is provided at the junction of the base shell and the funnel-shaped windshield. A laser ranging sensor is connected to the lower side of the shell cover. A spring is provided inside the base shell. The spring is sleeved on the nozzle. The upper end of the nozzle is provided with a quick-plug connector. The bottom end of the nozzle is provided with a film pressure sensor. The nozzle passes through the middle of the shell cover, and the bottom of the nozzle is located at the bottom of the windshield. lower side. 2. An aerial 3D printing robot according to claim 1, characterized in that the robotic arm includes a delta robotic arm and a three-axis series robotic arm, and the delta robotic arm includes a top plate connected to the drone body, A steering gear is arranged on the lower side of the top plate, and a steering gear housing is arranged on the outside of the steering gear. The output shaft of the steering gear is connected to the upper end of the upper arm through the shaft column, and the lower end of the upper arm is connected to the upper end of the lower arm. The lower end of the lower arm is connected to the annular chassis; The three-axis series manipulator includes a base joint connected to the bottom of the annular chassis, an elbow joint is connected to the lower side of the base joint, and a wrist joint is connected to the lower side of the elbow joint. 3. An aerial 3D printing robot according to claim 1, characterized in that the upper side of the flange is connected to the bottom of the wrist joint. 4. An aerial 3D printing robot according to claim 1, characterized in that the automatic control system includes a hardware part, and the hardware part includes an airborne computer NUC, a PX4 flight control, an airborne control panel and a ground control board, the flight control PX4 is responsible for the flight control of the drone, the airborne control board is responsible for the control and data processing of the robotic arm and damping printing nozzle, the ground control board controls the ground dispensing machine, the flight control PX4 and the airborne control board are connected via USB It communicates with the airborne computer NUC, and the ground control panel communicates with the airborne computer NUC through wireless Bluetooth. 5. An aerial 3D printing robot according to claim 1, characterized in that a visual camera is provided on the side of the drone body. 6. An aerial 3D printing method according to any one of claims 1 to 5, characterized in that it includes the following steps: S1. Obtain environmental information through the vision camera. The environmental information includes the position information of the target printing surface and surrounding obstacle information; S2. By analyzing the motor speed and using the airflow analysis algorithm, calculate the V-shaped area of the downwash airflow, and control the robotic arm to position the damped printing nozzle in the V-shaped area of the downwash airflow; S3. Obtain the distance between the end of the damping printing nozzle and the target printing surface through the laser ranging sensor, and obtain the pressure between the end of the damping printing nozzle and the target printing surface through the film pressure sensor; S4. Based on the 3D model to be printed, comprehensively use the position information obtained by RTK and the environmental information detected by the vision camera, combined with the distance information obtained by the laser ranging sensor obtained in step S3, and the pressure information obtained by the film pressure sensor. Step S2 The V-shaped area position information of the downwash airflow obtained from the system is used to calculate the optimal printing path that adapts to the current environment based on the path planning algorithm. 7. An aerial 3D printing method according to claim 6, characterized in that the airflow analysis algorithm includes: S21. Obtain the speed information of the drone motor through the flight control system; S22. Calculate the speed of the downwash airflow. The calculation formula is as follows: ; in, represents the speed of the downwash airflow; N is the rotation speed of the drone motor; ρ is the air density in the current environment; c is the performance coefficient describing the blade when generating airflow, taking into account the shape and aerodynamic characteristics of the blade; K is the coefficient; S23. Determine the direction of the downwash airflow based on the position, direction and angle of the drone's blades; S24. According to the speed of the downwash airflow calculated in step S22, the direction of the downwash airflow calculated in step S23, and the geometry of the motor rotor, calculate the V-shaped area formed by the downwash airflow through the following formula: ; Among them, h(x) represents the height of the downwash airflow, that is, the distance from the drone blades to the downwash airflow; D is the diameter of the motor rotor; x is the distance in the horizontal direction from the center position of the motor rotor; S25. Combined with the current position and attitude information of the drone, by calculating the angles of each joint of the robotic arm, adjust the position of the end nozzle to above the V-shaped area of the downwash airflow to reduce the interference of the airflow on the material. 8. An aerial 3D printing method according to claim 6, characterized in that the path planning algorithm includes: S31. Data acquisition: (1) Extract key information from the 3D model to be printed. Key information includes starting point, end point and obstacle information; (2) Use the RTK system to obtain the location information of the drone; (3) The vision camera detects the surrounding environment, including the position and posture of obstacles; (4) Call the airflow analysis algorithm to calculate the position and range of the V-shaped optimal printing area formed by the downwash airflow; S32. Combine the data obtained by the vision camera, laser ranging sensor, and film pressure sensor to construct an environment map and identify obstacles and safe areas; S33. Calculate the spatial distance from the starting point to the end point through the distance measurement algorithm, and use the A* path search algorithm to search for the path from the starting point to the end point in the environment map; during the search process, consider avoiding obstacles and keeping the path as close as possible. Within the V-shaped area; Distance measure: ; Among them, d represents the distance; (x_1,y_1,z_1) is the coordinates of the starting point; (x_2,y_2,z_2) is the coordinates of the target point; Path search algorithm: ; Among them, h(node) is a heuristic function used to estimate the cost from the node node to the target position; xtarget, ytarget and ztarget are the coordinates of the target position; xnode, ynode and znode are the coordinates of the node node ; S34. According to the objective function, comprehensively consider the length of the path, the distance to the obstacle, the distance to the end nozzle and the pressure data, and evaluate and sort the paths obtained by searching; ; Among them, F (path) is the objective function to evaluate the comprehensive performance of the path; w ₁ , w ₂ , w ₃ , and w ₄ are weight coefficients; obstacle_avoidance is a measure related to avoiding obstacles; nozzle_distance represents the distance between the end nozzle and the target printing surface Distance; pressure_data represents the pressure data exerted by the end nozzle; length(path) represents the path length; S35. If the evaluation finds that there is a better path, perform path optimization and smooth the path through interpolation method; S36. During the printing process, monitor the data of the laser ranging sensor and the film pressure sensor in real time. If the data exceeds the threshold, adjust the posture of the robot arm and adjust the end position of the damping printing nozzle. The adjustment formula for the end of the damping printing nozzle is as follows: ; Among them, new_position represents the new position of the damping printing nozzle after adjustment; current_position represents the current position of the damping printing nozzle; Δposition represents the position change amount of the damping printing nozzle that needs to be adjusted.