Robot flight adsorption method based on intelligent sensing control

Abstract

The invention discloses a robot flight adsorption method based on intelligent sensing control. The method comprises the steps that firstly, the flying adsorption robot obtains an operation task instruction, and the operation state of the robot is obtained in real time based on a flying adsorption robot sensor system; recognizing working plane parameters of the robot based on a multi-sensor intelligent sensing system; then aircraft target working plane adsorption is completed based on a hierarchical control structure; and finally, adsorption force intelligent compensation control is carried out based on a self-adaptive pressure control mode, and stable adsorption control is achieved. According to the flight adsorption method, the adsorption strategy can be automatically adjusted according to the condition of the surface needing to be adsorbed, the adsorption pressure can also be automatically adjusted, and the flight adsorption method has good environment adaptability and adsorption capacity and is wide in application.

Description

Technical Field

[0001] The invention relates to a robot flight adsorption method and relates to the field of flying robots, in particular to a robot flight adsorption method based on intelligent perception control. Background Art

[0002] Large bridges, represented by cross-sea bridges, are an important part of transportation facilities and one of the key links in the arteries of transportation. Cross-sea bridges are located at sea, with harsh working environments, large volumes and structures, and complex system components, which poses great challenges to the operation and maintenance of bridges. Existing technical equipment is not only complex to operate, has high work intensity, and high safety risks, but also has a large number of dead zones and blind spots, which cannot be effectively detected and maintained, thus bringing serious safety hazards to the long-term safe and stable operation of bridges. Flying and adsorbable robots are a relatively effective means of bridge maintenance and are receiving more and more attention.

[0003] A current knob-type suction cup is equipped with a ground monitoring station and radio frequency communication. Due to the use of a knob-type suction cup, a fixing device needs to be installed at the landing point of the drone. A current self-adjusting adsorption component (adsorption plate, self-adjusting push rod, adjustment turn block, reset spring and docking absorber) allows the drone to dock directly on a vertical or inclined wall, but the adsorption force cannot be adaptively adjusted according to the wall conditions. A current flight control method for an aircraft to adsorb or detach from a target object, the adsorption object can be a stationary wall, or a moving object, and the aircraft is adjusted so that the aircraft and the adsorption target are relatively stationary.

[0004] The flying adsorption method in the prior art has at least the following defects: 1. The parameters of the target working surface, an important influencing factor, are ignored in the control process of the flying adsorption; 2. The control method and the working mode are relatively single, and the control method of the flying adsorption robot cannot be adjusted according to the state of the working surface; 3. The adsorption force is fixed and cannot be flexibly adjusted according to the working state of the robot. Summary of the invention

[0005] In order to solve the problems existing in the background technology, the present invention provides a robot flight adsorption method based on intelligent perception control.

[0006] The technical solution adopted by the present invention is:

[0007] The robot flight adsorption method based on intelligent perception control of the present invention comprises:

[0008] Step 1: When the flying adsorption robot is flying, obtain the adsorption operation task instruction.

[0009] Step 2: Based on the sensor system mounted on the flying suction robot, the current operating status of the flying suction robot is obtained in real time.

[0010] Step three: Identify the working surface parameters of the flying suction robot based on the multi-sensor intelligent perception system mounted on the flying suction robot.

[0011] Step 4: Based on the current working state and working surface parameters of the flying adsorption robot, a hierarchical control structure is used to control the flying adsorption robot to perform adsorption on the target working surface according to the adsorption working task instruction.

[0012] Step 5: During the process of the flying adsorption robot adsorbing the target working surface, the expert system reasoning and adaptive pressure control method are used to perform intelligent compensation control of the adsorption force of the flying adsorption robot to achieve stable adsorption control until the flying adsorption of the flying adsorption robot is completed.

[0013] In the step 1, when the flying adsorption robot is flying, the flying adsorption robot receives the adsorption operation task instruction from the control terminal by wired or wireless means, and the adsorption operation task instruction is specifically the operation task instruction for the flying adsorption robot to fly to the target adsorption position.

[0014] In step 2, the sensor system includes but is not limited to one or a combination of an inertial navigation system INS (Inertial Navigation System), a gyrocompass, a GPS (Global Positioning System) locator, an altimeter, a barometer and a laser radar.

[0015] In the step three, the multi-sensor intelligent perception system includes one or a combination of image sensors, distance sensor arrays, attitude sensors and position sensors.

[0016] In the step three, the working surface parameters of the flying adsorption robot include the roughness and inclination angle of the working surface.

[0017] In the step four, the flying adsorption robot performs adsorption of the target working surface by specifically controlling the flying adsorption robot from the free flight stage to the stable adsorption and docking stage, such as the stage of approaching the target working surface, the stage of hovering in front of the target working surface, the stage of moving the adsorption device, the stage of adsorption and docking, etc.

[0018] In the step 5, when the adaptive pressure control method is used, intelligent compensation control of the adsorption force of each adsorption device on the flying adsorption robot is performed simultaneously.

[0019] The beneficial effects of the present invention are:

[0020] 1. The present invention identifies the working surface parameters of the robot based on a multi-sensor intelligent perception system, thereby realizing refined control of the robot's flight adsorption operation.

[0021] 2. The present invention adopts a hierarchical control structure to complete the adsorption process control of the aircraft target operation surface. The multi-layer structure cooperates with each other to jointly complete the switching of various operation modes and the flight adsorption operation trajectory control.

[0022] 3. When the robot is adsorbed and docked, the present invention adopts an adaptive pressure control method to perform intelligent compensation control of the adsorption force according to the operating state parameters to achieve stable adsorption control.

[0023] The flying adsorption method of the present invention can automatically adjust the adsorption strategy according to the surface conditions to be adsorbed, and can also automatically adjust the adsorption pressure. It has good environmental adaptability and adsorption capacity and is widely used. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a flow chart of the method of the present invention;

[0025] Figure 2 It is a schematic diagram of the process of identifying target working surface parameters by the multi-sensor intelligent perception system of the present invention;

[0026] Figure 3 It is a schematic diagram of the process of completing the adsorption of the target working surface of the aircraft by the hierarchical control structure of the present invention;

[0027] Figure 4 It is a schematic diagram of the intelligent compensation control of adsorption force based on the combination of expert system reasoning and adaptive pressure control of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail in conjunction with the accompanying drawings. This description introduces specific embodiments consistent with the principles of the present invention by way of example and not limitation, and the description of these embodiments is sufficiently detailed to enable those skilled in the art to practice the present invention, and other embodiments may be used and the structure of each element may be changed and / or replaced without departing from the scope and spirit of the present invention. Therefore, the following detailed description should not be understood in a restrictive sense.

[0029] like Figure 1 As shown, the robot flight adsorption method based on intelligent perception control of the present invention is specifically as follows:

[0030] Step 1: When the flying adsorption robot is flying, obtain the adsorption operation task instruction. When the flying adsorption robot is flying, the flying adsorption robot receives the adsorption operation task instruction from the control terminal in a wired or wireless manner. The adsorption operation task instruction is specifically the operation task instruction for the flying adsorption robot to fly to the target adsorption position.

[0031] Step 2: Based on the sensor system mounted on the flying adsorption robot, the current operation status of the flying adsorption robot is obtained in real time. The sensor system includes but is not limited to one or a combination of an inertial navigation system INS, a gyrocompass, a GPS locator, an altimeter, a barometer and a laser radar.

[0032] Step 3: Identify the working surface parameters of the flying adsorption robot based on the multi-sensor intelligent perception system mounted on the flying adsorption robot. The multi-sensor intelligent perception system includes one or a combination of image sensors, distance sensor arrays, attitude sensors, and position sensors. The working surface parameters of the flying adsorption robot include the roughness and inclination of the working surface.

[0033] Step 4: Based on the current working state and working surface parameters of the flying adsorption robot, a hierarchical control structure is used to control the flying adsorption robot to perform adsorption on the target working surface according to the adsorption operation task instruction. The flying adsorption robot performs adsorption on the target working surface by specifically controlling the flying adsorption robot from the free flight stage to the stable adsorption and docking stage, such as the stage of approaching the target working surface, the stage of hovering in front of the target working surface, the stage of moving the adsorption device, the stage of adsorption and docking, etc.

[0034] Step 5: During the process of the flying adsorption robot adsorbing the target working surface, the adsorption force of the flying adsorption robot is intelligently compensated and controlled using the expert system reasoning and the adaptive pressure control method to achieve stable adsorption control until the flying adsorption of the flying adsorption robot is completed. When the adaptive pressure control method is used, the adsorption force of each adsorption device on the flying adsorption robot is intelligently compensated and controlled at the same time.

[0035] The specific embodiments of the present invention are as follows:

[0036] The method of the present invention can be applied or combined with a variety of flying adsorption equipment, such as a bridge inspection flying adsorption robot, on which an adsorption system, an intelligent sensing system, a control system and a visual monitoring system are mounted. The adsorption system of the robot adopts negative pressure adsorption technology, and the adsorption system includes: a centrifugal fan, a pipe, a suction cup group, a pressure sensor, an adsorption support arm, etc. The adsorption support arm can be extended and turned under the action of the mechanism, so that it can adapt to docking on cement surfaces with different inclination angles.

[0037] In step three, the multi-sensor intelligent perception system includes an image-based surface roughness detection unit, an ultrasonic array sensor-based working surface inclination measurement unit, and an airborne strapdown attitude sensing unit.

[0038] like Figure 2As shown in the figure, it is the process of multi-sensor intelligent perception system identifying the robot's operating parameters. The recognition of the roughness of the operating surface is realized by image processing. First, the bridge surface state is divided into five levels: smooth, moderate, rough, very rough, and uneven. The collected bridge deck structure surface photos are classified and labeled. At the same time, the samples are multiplied by illumination transformation and projection transformation to form a sample library. Then the labeled images are sent to a CNN (Convolutional Neural Networks) convolutional neural network with a relatively simple structure, and the network is trained, so that a neural network that divides the operating surface state into 5 categories can be established. In the actual operation process, the images collected by the video device are extracted and sent to the trained neural network, so that the qualitative classification of the operating docking surface can be obtained.

[0039] The real-time detection of the inclination angle of the working surface is achieved by ultrasonic distance measurement. During operation, the four ultrasonic sensors on board point to the working surface, so that the four distance parameters of the robot platform relative to the working surface can be measured. According to the difference of the four distance parameters, the inclination angle of the working surface relative to the robot platform can be calculated. Combining the pointing angle of the sensor and the posture of the robot platform can be further converted into the inclination angle of the working surface relative to the earth coordinate system.

[0040] The posture detection of the robot main platform is achieved through the strapdown inertial navigation system, gyrocompass and GPS, which can measure the robot's pitch, roll and heading parameters, and then use the extended Kalman filter (EKF) for data fusion to finally obtain the robot's posture.

[0041] The robot's mechanism rotation angle is controlled according to the robot's suction cup and bracket, and finally the robot's operating status and working surface parameters are obtained based on the above.

[0042] In step 4, if Figure 3 As shown in the figure, the hierarchical control structure includes the task management layer, the operation planning and decision-making layer, and the process control layer. The task management layer receives task instructions, the operation planning and decision-making layer generates the control instructions of the underlying mechanism according to the task and environmental parameters, and the process control layer completes the control of each underlying mechanism. The three-layer structure cooperates with each other to complete the switching of multiple operation modes and the control of the flight adsorption operation trajectory.

[0043] In terms of operation mode, the robot can be divided into bridge bottom operation mode, acute angle slope operation mode, obtuse angle slope operation mode and vertical surface operation mode. On each operation surface, it can be divided into arrival and docking stage, stable operation stage, movement stage within the operation surface and flight off the operation surface stage. In addition, when there is insufficient adsorption force or significant abnormal interference, a safety disposal stage is required. The hierarchical control structure used is as follows: Figure 4As shown in the figure, it is mainly divided into task management layer, operation planning and decision layer and process control layer. The task management layer accepts remote control instructions and emergency disposal instructions based on safety observation at the same time, sets the tasks that need to be performed at present, and manages the operation tasks, including emergency, docking, flying away and walking. The operation planning and decision layer plans and makes decisions on the parameters of the flight control system and the vacuum adsorption system according to the tasks, state parameters and environmental parameters that need to be performed at present, and forms the motion control instructions (flight speed, direction, lifting force / ascending and descending speed) and target pressure, working state (running, stopping) and suction cup motion control instructions of the vacuum adsorption system at the same time, so as to carry out airport action planning, operation path planning, adsorption force decision and platform attitude decision. The bottom actuator includes the adsorption force control unit with centrifugal fan as the actuator, the suction cup extension / steering control unit and the flight rotor control unit, so as to carry out ultrasonic ranging bracket control, rotor flight control, detection device motion control, suction cup rotation control, operation device motion control, rack motion control and adsorption control, etc. The three-layer structure cooperates with each other to complete the switching of multiple operation modes and flight adsorption operation trajectory control.

[0044] In step five, intelligent compensation control of the adsorption force can be performed based on a combination of expert system reasoning and adaptive pressure control to achieve stable adsorption control. In order to keep the robot stable during adsorption and docking, intelligent compensation control of the adsorption force is performed based on a combination of expert system reasoning and adaptive pressure control to achieve stable adsorption control. The bridge deck working surface of the flying adsorption robot can be divided into four categories according to the inclination angle: bottom surface, acute angle inclined surface, vertical vertical surface, and obtuse angle inclined surface, that is, including vertical surface, bottom surface and inclined surface. According to the roughness of the bridge wall, it can be divided into five levels: smooth, moderate, relatively rough, very rough, and uneven. Different inclination states and surface states have great differences in the demand for adsorption force.

[0045] First, based on the test and force balance analysis of the robot under different working modes and working surface conditions, the expert experience with working inclination angle and surface state as input is sorted out to construct a knowledge base. In the actual operation process, taking the control of two suction cups as an example, according to the working parameters identified by the sensor system and the image system, the two vacuum adsorption pressures that need to be controlled are obtained from the knowledge base based on the knowledge reasoning method as the input of the centrifugal fan control. The pressure sensor is installed in the negative pressure chamber of the adsorption device to measure the actual pressure of the negative pressure chamber in real time. The process control unit compares the actual pressure with the target pressure. According to the pressure difference, the adaptive closed-loop control method is used to adjust the adsorption force, that is, the pressure difference is input into the adaptive controller for processing, and then the driver controls the centrifugal pump to control the two suction cups separately.

[0046] Therefore, the present invention completes the flying adsorption of the bridge inspection flying adsorption robot through the intelligent perception control method.

Claims

1. A robot flight adsorption method based on intelligent perception control, characterized in that: include: Step 1: When the flying adsorption robot is flying, obtain an adsorption operation task instruction; Step 2: obtaining the current operation status of the flying suction robot in real time based on a sensor system mounted on the flying suction robot; Step 3: Identify the working surface parameters of the flying suction robot based on the multi-sensor intelligent perception system carried on the flying suction robot; Step 4: Based on the current working state and working surface parameters of the flying adsorption robot, a hierarchical control structure is used to control the flying adsorption robot to perform adsorption on the target working surface according to the adsorption working task instruction; Step 5: During the process of the flying adsorption robot adsorbing the target working surface, an adaptive pressure control method is used to perform intelligent compensation control of the adsorption force of the flying adsorption robot until the flying adsorption robot completes the flight adsorption.

2. The robot flight adsorption method based on intelligent perception control according to claim 1 is characterized in that: In the step 1, when the flying adsorption robot is flying, the flying adsorption robot receives the adsorption operation task instruction from the control terminal by wired or wireless means, and the adsorption operation task instruction is specifically the operation task instruction for the flying adsorption robot to fly to the target adsorption position.

3. The robot flight adsorption method based on intelligent perception control according to claim 1 is characterized in that: In the step 2, the sensor system includes one or a combination of an inertial navigation system INS, a gyrocompass, a GPS locator, an altimeter, a barometer and a laser radar.

4. The robot flight adsorption method based on intelligent perception control according to claim 1 is characterized in that: In the step three, the multi-sensor intelligent perception system includes one or a combination of image sensors, distance sensor arrays, attitude sensors and position sensors.

5. The robot flight adsorption method based on intelligent perception control according to claim 1 is characterized in that: In the step three, the working surface parameters of the flying adsorption robot include the roughness and inclination angle of the working surface.

6. The robot flight adsorption method based on intelligent perception control according to claim 1 is characterized in that: In the step 4, the flying adsorption robot performs adsorption on the target working surface by specifically controlling the flying adsorption robot from the free flight stage to the stable adsorption and docking stage.

7. The robot flight adsorption method based on intelligent perception control according to claim 1 is characterized in that: In the step 5, when the adaptive pressure control method is used, intelligent compensation control of the adsorption force of each adsorption device on the flying adsorption robot is performed simultaneously.

Images