Intelligent control system and method of UAV flight path based on AI algorithm

Abstract

The present invention discloses an intelligent control system and method for a UAV flight path based on an AI algorithm, relates to the field of UAVs, and solves the problem that UAVs lack the ability to cope with emergencies. The system comprises an airframe self-check module, an anti-interference control module, an obstacle perception module, an avoidance execution module, a path analysis module, and a path correction module. The airframe self-check module is used to perform a shaft smoothness test and a blade tilt dead zone test on the airframe of the UAV, the anti-interference control module is used to perform intelligent control on the anti-interference strength of the UAV during flight; the obstacle perception module is used to perceive obstacles during the flight of the UAV, the avoidance execution module is used to control the avoidance behavior of the UAV, the path analysis module is used to perform real-time optimization on the flight route of the UAV, and the path correction module corrects and optimizes the current flight path of the UAV. The present invention realizes autonomous intelligent obstacle avoidance and path decision optimization of the UAV.

Description

Technical Field

[0001] The present invention belongs to the field of unmanned aerial vehicles (UAVs), and relates to UAV flight path control technology, and specifically to an intelligent control system and method for UAV flight paths based on an AI algorithm. Background Art

[0002] A drone is an unmanned aircraft. With the development of technology and the reduction of costs, drones are widely used in the civil field, construction and infrastructure field, disaster response, and security and monitoring fields. Due to the diversity of application scenarios and the uncertainties in the scenarios, the current application fields have high professional requirements for drones' rapid response, efficient data collection, and flexible response to emergencies.

[0003] Under the current technological background, the application scenarios of drones are becoming more and more diverse. Most existing drones rely on remote control by drone pilots or fly according to pre-input flight paths. They lack the ability to respond to sudden environments and obstacles that suddenly appear. Therefore, how to achieve autonomous intelligent obstacle avoidance and path decision optimization of drones is the current problem.

[0004] To this end, we propose an intelligent control system and method for UAV flight path based on AI algorithm. Summary of the invention

[0005] In view of the deficiencies in the prior art, the object of the present invention is to provide an intelligent control system and method for a UAV flight path based on an AI algorithm.

[0006] The technical problems to be solved by the present invention are:

[0007] How to achieve autonomous intelligent obstacle avoidance and path decision optimization for drones.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] On the one hand, an AI algorithm-based intelligent control system for a UAV flight path includes an airframe self-check module, an environmental monitoring module, an anti-disturbance control module, an obstacle perception module, an avoidance execution module, a path analysis module, a path correction module, a control terminal, and a processor; the control terminal is used to upload an initial flight path; the airframe self-check module is used to perform a shaft smoothness test and a blade tilt dead zone test on the airframe of the UAV, and generate a test abnormality signal or a test pass signal and send it to the control terminal via the processor; the control terminal is connected to a display, and the display is used to display a test abnormality signal or a test pass signal;

[0010] When performing flight work according to the initial flight path, the environment monitoring module is used to obtain environmental interference data and signal interference data of the UAV during flight, and the environmental interference data is sent to the path analysis module and the signal interference data is sent to the anti-interference control module via the processor;

[0011] The anti-interference control module is used to intelligently control the anti-interference strength of the UAV during flight; the obstacle perception module is used to perceive obstacles during the flight of the UAV, generate horizontal avoidance signals or lifting avoidance signals, and obtain the obstacle avoidance response level, and the horizontal avoidance signal or lifting avoidance signal and the obstacle avoidance response level are sent to the avoidance execution module via the processor; the avoidance execution module is used to control the avoidance behavior of the UAV;

[0012] The obstacle perception module is also used to obtain path obstacle data of the UAV during flight, and send the path obstacle data of the UAV to the path analysis module through the processor; the path analysis module is used to optimize the flight route of the UAV in real time, generate a path re-planning signal and send it to the path correction module through the processor; the path correction module corrects and optimizes the current flight path of the UAV.

[0013] Furthermore, the environmental interference data includes the real-time environmental wind resistance, real-time environmental temperature, real-time environmental humidity and the real-time altitude of the drone during flight;

[0014] Signal interference data includes the center distance, center interference intensity and interference coverage radius of each potential interference source in the drone's corresponding signal detection area;

[0015] The path obstacle data includes the avoidance buffer distance, lateral avoidance distance, longitudinal avoidance distance between the UAV and the obstacle, and the real-time flight speed of the UAV;

[0016] The path obstacle data includes the number of obstacles in the obstacle perception area corresponding to the drone, as well as the obstacle specifications, obstacle distance and obstacle spacing of each obstacle.

[0017] Furthermore, the test steps of the shaft smoothness test specifically include:

[0018] After starting the propeller, monitor the shaft amplitude and shaft temperature of the propeller on the drone;

[0019] The shaft amplitude of the propeller on the drone is compared with the test amplitude critical value, and at the same time, the shaft temperature of the propeller on the drone is compared with the test temperature critical value;

[0020] If the value of the shaft body amplitude is greater than the test amplitude or the value of the shaft body temperature is greater than the test temperature critical value, a test abnormality signal is generated;

[0021] If the value of the shaft amplitude is less than or equal to the test amplitude and the value of the shaft temperature is less than or equal to the test temperature critical value, a blade tilt dead zone test is performed.

[0022] Furthermore, the test steps of the blade tilt dead zone test specifically include:

[0023] A plane rectangular coordinate system is established with the center of the spiral wing as the coordinate origin, the positive direction of the longitudinal axis is taken as the initial position, and the inclination angle of the rotation axis is set to the maximum value for inclination;

[0024] With the rotating shaft as the center, rotate the rotating shaft clockwise for one circle and monitor the jamming during the rotation process;

[0025] During the rotation of the rotating shaft, if a jam occurs, a test abnormality signal is generated; if no jam occurs, the tilt angle of the rotating shaft is adjusted and the test process is repeated;

[0026] When the value of the tilt angle of the rotating axis returns to zero, if no jamming occurs, a test pass signal is generated.

[0027] Furthermore, the control process of the anti-disturbance control module is as follows:

[0028] Obtain the center distance and interference coverage radius of each potential interference source in the UAV's corresponding signal detection area;

[0029] Compare the value of the center distance with the value of the interference coverage radius;

[0030] If the value of the center distance is greater than the value of the interference coverage radius, the corresponding potential interference source is screened out;

[0031] If the value of the center distance is less than or equal to the value of the interference coverage radius, the corresponding potential interference source is marked as an actual interference source and proceeds to the next step;

[0032] Obtain the center distance, interference coverage radius and center interference intensity of the actual interference source in the corresponding signal detection area of ​​the drone;

[0033] Calculate the interference intensity of each actual interference source in the UAV’s corresponding signal detection area;

[0034] The interference strength of each actual interference source is compared to obtain the maximum interference strength in the UAV's corresponding signal detection area;

[0035] Compare the maximum interference intensity in the signal detection area corresponding to the drone with the interference intensity threshold;

[0036] If the value of the interference intensity is less than or equal to the first interference intensity threshold, the anti-interference level of the UAV is adjusted to the first anti-interference level;

[0037] If the value of the interference intensity is greater than the first interference intensity threshold and less than or equal to the second interference intensity threshold, the anti-interference level of the UAV is adjusted to the second anti-interference level;

[0038] If the interference intensity value is greater than the third interference intensity threshold, the anti-interference level of the UAV is adjusted to the third anti-interference level;

[0039] Among them, the values ​​of the first interference intensity threshold and the second interference intensity threshold are both greater than zero, the first interference intensity threshold is less than the second interference intensity threshold, the anti-interference performance of the first anti-interference level is lower than the anti-interference performance of the first anti-interference level, and the anti-interference performance of the second anti-interference level is lower than the anti-interference performance of the third anti-interference level.

[0040] Furthermore, the working process of the obstacle perception module is as follows:

[0041] The obstacle sensing area corresponding to the UAV is constructed based on the limit sensing distance of the sensing element. When an obstacle enters the obstacle sensing area, the avoidance buffer distance, lateral avoidance distance, longitudinal avoidance distance between the UAV and the obstacle and the real-time flight speed of the UAV are obtained.

[0042] Calculate the real-time approach rate of the obstacle, and compare the real-time approach rate of the obstacle with the approach rate threshold, and determine whether the avoidance response level of the corresponding obstacle is the first avoidance response level, the second avoidance response level, or the third avoidance response level; wherein the avoidance immediacy of the first avoidance response level is lower than the avoidance immediacy of the second avoidance response level, and the avoidance immediacy of the second avoidance response level is lower than the avoidance immediacy of the third avoidance response level;

[0043] Then, the translational avoidance displacement and the lifting avoidance displacement of the obstacle are calculated, and the numerical values ​​of the translational avoidance displacement and the lifting avoidance displacement of the obstacle are compared to generate a horizontal avoidance signal or a lifting avoidance signal.

[0044] Furthermore, the working process of the avoidance execution module is as follows:

[0045] If a horizontal avoidance signal is received, the drone is controlled to avoid obstacles laterally;

[0046] If the lift avoidance signal is received, the drone is controlled to avoid obstacles longitudinally;

[0047] When the obstacle avoidance response level is the first avoidance response level, there is no need to change the flight speed of the drone during the avoidance process;

[0048] When the obstacle avoidance response level is the second avoidance response level, the flight speed of the drone needs to be reduced during the avoidance process;

[0049] When the obstacle avoidance response level is the third avoidance response level, the drone needs to be controlled to perform emergency braking and avoid the obstacle in time during the avoidance process.

[0050] Furthermore, the analysis and optimization process of the path analysis module is specifically as follows:

[0051] Obtain the number of obstacles in the obstacle perception area corresponding to the drone, as well as the obstacle specifications and obstacle distances of each obstacle, and calculate the avoidance complexity coefficient in the obstacle perception area corresponding to the drone;

[0052] Then, the real-time environmental wind resistance, real-time environmental temperature, real-time environmental humidity and real-time altitude of the UAV during flight are obtained to calculate the environmental interference coefficient of the UAV during flight;

[0053] Compare the avoidance complexity coefficient of the drone in the corresponding obstacle perception area with the avoidance complexity critical value, and compare the environmental interference coefficient of the drone during flight with the environmental interference critical value;

[0054] If the avoidance complexity coefficient of the drone in the corresponding obstacle perception area is less than or equal to the avoidance complexity critical value, and the environmental interference coefficient of the drone during flight is less than or equal to the environmental interference critical value, no operation will be performed;

[0055] If the avoidance complexity coefficient of the drone in the corresponding obstacle perception area is greater than the avoidance complexity critical value or the environmental interference coefficient of the drone during flight is greater than the environmental interference critical value, a path replanning signal is generated.

[0056] Furthermore, the correction measures of the path correction module specifically include:

[0057] If a path re-planning signal is received, the drone's avoidance strategy is adjusted to a long-distance detour to avoid the obstacle group;

[0058] If no path re-planning signal is received, the drone's avoidance strategy is adjusted to short-distance displacement to pass through obstacles.

[0059] The second aspect is the intelligent control method of the UAV flight path based on AI algorithm, which specifically includes:

[0060] Step S101, uploading the initial flight path, and performing a shaft smoothness test and a blade tilt dead zone test on the drone body, generating a test abnormality signal or a test pass signal;

[0061] Step S102: if a test pass signal is generated, the environmental interference data and the signal interference data of the drone are obtained, and the anti-interference strength of the drone is intelligently adjusted;

[0062] Step S103, simultaneously sensing obstacles during the flight of the UAV, generating a horizontal avoidance signal or a vertical avoidance signal, and obtaining an avoidance response level of the obstacle, and controlling the avoidance behavior of the UAV in combination with the horizontal avoidance signal, the vertical avoidance signal or the avoidance response level;

[0063] Step S104, finally obtaining the path obstacle data of the UAV, optimizing the flight route of the UAV in real time, generating a path re-planning signal, and correcting and optimizing the current flight path of the UAV according to the path re-planning signal.

[0064] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0065] 1. The present invention first uploads the initial flight path to the processor through the control terminal, and performs a shaft smoothness test and a blade tilt dead zone test on the drone body through the body self-check module, and then uses the environmental monitoring module to obtain the signal interference data of the drone and send it to the anti-interference control module, and then uses the anti-interference control module to intelligently control the anti-interference strength of the drone, and then uses the obstacle perception module to perceive the obstacles in the flight process of the drone, generates a horizontal avoidance signal or a lifting avoidance signal and sends it to the avoidance execution module, and obtains the obstacle avoidance response level and sends it to the avoidance execution module, and finally controls the avoidance behavior of the drone through the avoidance execution module. The present invention realizes autonomous intelligent obstacle avoidance of the drone.

[0066] 2. The present invention obtains the environmental interference data of the UAV through the environmental monitoring module and sends it to the path analysis module, and uses the obstacle perception module to obtain the path obstacle data of the UAV and sends it to the path analysis module. The path analysis module then analyzes the flight route of the UAV, generates a path re-planning signal and sends it to the path correction module. Finally, the current flight path of the UAV is corrected and optimized by the path correction module. The present invention realizes the path decision optimization of the UAV. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0068] Figure 1 is the overall system block diagram of the present invention;

[0069] Figure 2 This is a physical demonstration diagram of the machine body self-check module in the present invention;

[0070] Figure 3 This is a schematic diagram of the operation of the machine body self-check module in the present invention;

[0071] Figure 4 This is a test principle diagram of the environmental monitoring module in the present invention;

[0072] Figure 5 This is a working principle diagram of the obstacle perception module in the present invention;

[0073] Figure 6 This is a schematic diagram of the operation of the obstacle perception module in the present invention;

[0074] Figure 7 The present invention is a flow chart of the method.

[0075] In the figure: 1. Helical wing; 11. Helical wing No. 1; 12. Helical wing No. 2; 13. Helical wing No. 3; 14. Helical wing No. 4; 2. Rotation axis. DETAILED DESCRIPTION

[0076] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0077] Example 1: Please refer to Figure 1-Figure 6 As shown, the present invention provides a technical solution: an intelligent control system for the flight path of a UAV based on an AI algorithm, comprising an aircraft self-check module, an environmental monitoring module, an anti-disturbance control module, an obstacle perception module, an avoidance execution module, a path analysis module, a path correction module, a control terminal and a processor;

[0078] In the actual working process, the obstacle avoidance plan of the drone during flight is decided through the built-in AI algorithm of the drone. The AI ​​algorithm also makes path decision based on the flight status of the drone and the interference of the surrounding environment;

[0079] The control terminal uploads the initial flight path to the processor. It should be specifically noted that the UAV involved in this solution needs to input the initial flight path in advance during the pre-flight preparation stage. Unless there are special circumstances, the UAV will strictly follow the initial flight path to complete the flight mission during the flight.

[0080] See also Figure 2 and Figure 3 As shown, the body self-check module is used to perform a pre-flight self-check on the body of the drone. The self-check process is as follows:

[0081] like Figure 2 As shown, according to the order of the numbers of the rotors 1, the operating status of each rotor 1 on the UAV is tested one by one. The test items include a shaft smoothness test and a blade tilt dead zone test. The rotor 1 includes a No. 1 rotor 11, a No. 2 rotor 12, a No. 3 rotor 13 and a No. 4 rotor 14.

[0082] In this embodiment, a quad-wing UAV is preferably used as the UAV for the flight mission. In the actual work process, other models and styles of UAVs can be selected according to actual flight requirements and environmental requirements;

[0083] The shaft smoothness test is specifically to monitor the shaft amplitude and shaft temperature of the rotor 1 on the drone after starting the rotor 1. In this embodiment, the shaft amplitude and shaft temperature of the rotor 1 on the drone are obtained by a vibration sensor and a temperature sensor provided at the shaft of each rotor 1 on the drone;

[0084] The shaft body amplitude of the rotor 1 on the UAV is compared with the test amplitude critical value, and at the same time, the shaft body temperature of the rotor 1 on the UAV is compared with the test temperature critical value;

[0085] If the value of the shaft body amplitude is greater than the test amplitude or the value of the shaft body temperature is greater than the test temperature critical value, a test abnormality signal is generated;

[0086] If the value of the shaft body amplitude is less than or equal to the test amplitude and the value of the shaft body temperature is less than or equal to the test temperature critical value, then perform subsequent tests;

[0087] It is understandable that if the rotating shaft position of the spiral wing 1 has abnormal vibration and abnormal heating, it means that the rotating shaft has abnormal friction or collision during the rotation process, which makes the flight mission have safety hazards;

[0088] See also Figure 3 As shown, the blade tilt dead zone test is specifically to test whether there is a tilt dead zone on the blade of the rotor 1. The test steps are as follows:

[0089] Step A1, establishing a plane rectangular coordinate system with the center of the spiral wing 1 as the coordinate origin, taking the positive direction of the longitudinal axis as the initial position, and setting the inclination angle of the rotating axis 2 to the maximum value for inclination;

[0090] Step A2, taking the rotating shaft 2 as the center, rotating the rotating shaft 2 clockwise for one circle, and monitoring the jamming during the rotation process;

[0091] Step A3, during the rotation of the rotating shaft 2, if a jam occurs, a test abnormality signal is generated; if no jam occurs, the inclination angle of the rotating shaft 2 is adjusted and the test process is repeated;

[0092] Step A4, when the value of the tilt angle of the rotating shaft 2 returns to zero, if no jamming occurs, a test pass signal is generated;

[0093] The machine body self-check module sends a test abnormality signal or a test pass signal to the processor. If the processor receives the test abnormality signal, the test abnormality signal is sent to the control terminal. If the processor receives the test normal signal, no operation is performed;

[0094] The control terminal is connected to a display, and the display is used to display a test abnormality signal or a test pass signal. If the test abnormality signal is displayed, it reminds the operator that the drone body has an abnormality. If the test pass signal is displayed, the drone has no abnormality and can perform flight work;

[0095] After the quad-wing UAV completes the pre-flight self-check, the flight work is performed and the flight starts according to the initial flight path in the processor. The environmental monitoring module is used to obtain the environmental interference data of the UAV during flight, and the environmental interference data includes the real-time environmental wind resistance, real-time environmental temperature, real-time environmental humidity and real-time ground height of the UAV during flight;

[0096] The real-time environmental wind resistance is specifically the resistance encountered by the drone during flight. In actual operation, the value of the real-time environmental wind resistance is affected by the environmental wind speed and the windward area of ​​the drone. The real-time height above the ground is the height difference between the position of the drone and the ground. In this embodiment, the real-time environmental wind resistance, real-time environmental temperature and real-time environmental humidity are obtained through the meteorological sensor integrated in the drone, and the real-time height above the ground of the drone is obtained through the wireless positioning element.

[0097] The environment monitoring module sends the environmental interference data of the UAV during flight to the processor, and the processor sends the environmental interference data of the UAV during flight to the path analysis module;

[0098] See also Figure 4 As shown, the environmental monitoring module is also used to obtain signal interference data when the drone is flying. In this embodiment, the drone is equipped with a signal detection element. A signal detection area is constructed based on the detection distance of the signal detection element. The drone monitors the signal interference data in the signal detection area in real time through the signal detection element. The signal interference data includes the center distance, center interference intensity and interference coverage radius of each potential interference source in the signal detection area of ​​the drone.

[0099] like Figure 4 As shown in the figure, the center distance is the spatial distance between the UAV and the center position of the potential interference source, the center interference intensity is the interference intensity at the center position of the potential interference source, and the interference coverage radius reflects the influence range of the potential interference source. In the actual working process, the potential interference sources include but are not limited to radio transmitters, wireless signaling towers, wireless base stations, etc.

[0100] The environmental monitoring module sends the signal interference data of the UAV during flight to the processor, and the processor sends the signal interference data of the UAV during flight to the anti-interference control module;

[0101] The anti-interference control module is used to intelligently control the anti-interference strength of the drone during flight. The control process is as follows:

[0102] Obtain the center distance and interference coverage radius of each potential interference source in the UAV's corresponding signal detection area;

[0103] Compare the value of the center distance with the value of the interference coverage radius;

[0104] If the value of the center distance is greater than the value of the interference coverage radius, the corresponding potential interference source is screened out;

[0105] If the value of the center distance is less than or equal to the value of the interference coverage radius, the corresponding potential interference source is marked as an actual interference source and proceeds to the next step;

[0106] Then, the center distance YLr, interference coverage radius YRr and center interference intensity YQr of the actual interference source in the UAV corresponding signal detection area are obtained, where r is the actual interference source serial number, the upper limit of r is m, and the value of m is equal to the number of actual interference sources;

[0107] The interference intensity RQr of each actual interference source in the UAV's corresponding signal detection area is calculated according to the formula. The specific formula is as follows:

[0108] ;

[0109] The interference strength of each actual interference source is compared to obtain the maximum interference strength in the UAV's corresponding signal detection area;

[0110] Compare the maximum interference intensity in the signal detection area corresponding to the drone with the interference intensity threshold;

[0111] If the value of the interference intensity is less than or equal to the first interference intensity threshold, the anti-interference level of the UAV is adjusted to the first anti-interference level;

[0112] If the value of the interference intensity is greater than the first interference intensity threshold and less than or equal to the second interference intensity threshold, the anti-interference level of the UAV is adjusted to the second anti-interference level;

[0113] If the interference intensity value is greater than the third interference intensity threshold, the anti-interference level of the UAV is adjusted to the third anti-interference level;

[0114] The values ​​of the first interference intensity threshold and the second interference intensity threshold are both greater than zero, the first interference intensity threshold is less than the second interference intensity threshold, the anti-interference performance of the first anti-interference level is lower than the anti-interference performance of the first anti-interference level, and the anti-interference performance of the second anti-interference level is lower than the anti-interference performance of the third anti-interference level;

[0115] See also Figure 5 and Figure 6 As shown, the obstacle sensing module is used to sense obstacles during the flight of the drone. The working process of the module is as follows:

[0116] In this embodiment, the drone is equipped with sensing elements, including but not limited to laser radar, infrared sensor, visual sensor, ultrasonic sensor, etc.;

[0117] like Figure 5 As shown, based on the limit sensing distance of the sensing element, the obstacle sensing area corresponding to the UAV is constructed. Specifically, a circular area is constructed with the UAV as the center and the limit sensing distance of the sensing element as the radius, and is named the obstacle sensing area;

[0118] like Figure 6 As shown, when an obstacle enters the obstacle sensing area, the path obstacle data of the drone is obtained through the sensing element mounted on the drone body. The path obstacle data includes the avoidance buffer distance GH, the lateral avoidance distance HG, the longitudinal avoidance distance ZG between the drone and the obstacle, and the real-time flight speed FS of the drone, wherein the lateral avoidance distance is the horizontal displacement distance when the drone avoids the obstacle, and the longitudinal avoidance distance is the vertical displacement distance when the drone avoids the obstacle;

[0119] The real-time obstacle closing rate ZV is calculated according to the formula, which is as follows:

[0120] ZV = (GH-GH') / t-FS;

[0121] Among them, t is the interval between the previous moment and the next moment, t is a fixed value, and the real-time approach rate can be regarded as a component speed of the obstacle in the direction of approaching the drone;

[0122] Compare the real-time closing rate of the obstacle with the closing rate threshold;

[0123] If the real-time closing rate is less than or equal to the first closing rate threshold, determining that the obstacle avoidance response level is the first avoidance response level;

[0124] If the real-time closing rate is greater than the first closing rate threshold and less than or equal to the second closing rate threshold, determining that the obstacle avoidance response level is the second avoidance response level;

[0125] If the real-time closing rate is less than or equal to the second closing rate threshold, determining that the obstacle avoidance response level is the third avoidance response level;

[0126] Wherein, the values ​​of the first approach rate threshold and the second approach rate threshold are both greater than zero, the first approach rate threshold is less than the second approach rate threshold, the avoidance immediacy of the first avoidance response level is lower than the avoidance immediacy of the second avoidance response level, and the avoidance immediacy of the second avoidance response level is lower than the avoidance immediacy of the third avoidance response level;

[0127] The obstacle avoidance displacement PG and elevation avoidance displacement SG are calculated according to the formula. The specific formula is as follows:

[0128] ;

[0129] ;

[0130] Compare the numerical values ​​of the obstacle's translational avoidance displacement and the elevation avoidance displacement;

[0131] If the value of the translation avoidance displacement is less than or equal to the lifting avoidance displacement, a horizontal avoidance signal is generated;

[0132] If the value of the translation avoidance displacement is greater than the lifting avoidance displacement, a lifting avoidance signal is generated;

[0133] The obstacle sensing module sends the horizontal avoidance signal or the lifting avoidance signal and the obstacle avoidance response level to the processor, and the processor sends the horizontal avoidance signal or the lifting avoidance signal and the obstacle avoidance response level to the avoidance execution module;

[0134] The avoidance execution module is used to control the avoidance behavior of the drone. The working process of the module is as follows:

[0135] If a horizontal avoidance signal is received, the drone is controlled to avoid obstacles laterally;

[0136] If the lift avoidance signal is received, the drone is controlled to avoid obstacles longitudinally;

[0137] When the obstacle avoidance response level is the first avoidance response level, there is no need to change the flight speed of the drone during the avoidance process;

[0138] When the obstacle avoidance response level is the second avoidance response level, the flight speed of the drone needs to be reduced during the avoidance process;

[0139] When the obstacle avoidance response level is the third avoidance response level, the drone needs to be controlled to perform emergency braking and avoid the obstacle in time during the avoidance process;

[0140] It is understandable that the hierarchical control of the UAV’s flight speed during the avoidance process is to reserve more avoidance space and ensure that the damage to the UAV during the avoidance process is minimized;

[0141] The obstacle perception module also obtains the path obstacle data of the drone during flight and sends it to the processor, and the processor sends the path obstacle data of the drone during flight to the path analysis module, the path obstacle data includes the number of obstacles in the obstacle perception area of ​​the drone and the obstacle specification, obstacle distance and obstacle spacing of each obstacle, wherein the obstacle specification is the space area occupied by the obstacle in the perspective of the drone, the obstacle distance is the distance between the obstacle and the drone, and the obstacle spacing is the minimum spacing between the obstacle and other obstacles;

[0142] The path analysis module is used to optimize the flight path of the UAV in real time. The analysis and optimization process is as follows:

[0143] Obtain the number of obstacles, obstacle specifications ZSi and obstacle spacing ZTi in the obstacle perception area of ​​the drone, where i is the sequence number of obstacles in the obstacle perception area, i is a non-zero natural number, and the upper limit of i is n, and the value of n is equal to the number of obstacles in the obstacle perception area;

[0144] The avoidance complexity coefficient GB of the drone in the obstacle perception area is calculated according to the formula. The specific formula is as follows:

[0145] ;

[0146] Among them, s1 and s2 are fixed value proportional coefficients, and the values ​​of s1 and s2 are both greater than zero. It can be understood that the value of the obstacle specification is directly proportional to the value of the avoidance complexity coefficient, and the value of the obstacle spacing is inversely proportional to the value of the avoidance complexity coefficient;

[0147] Obtain the real-time environmental wind resistance FZ, real-time environmental temperature HT, real-time environmental humidity HW and real-time ground height LH of the drone during flight, and calculate the environmental interference coefficient HJ of the drone during flight according to the formula. The specific formula is as follows:

[0148] HJ=FZ×LH×a1+|HT-BT|×a2+HW×a3;

[0149] Among them, a1, a2 and a3 are weight coefficients with fixed values, the values ​​of a1, a2 and a3 are all greater than zero and a1+a2+a3=1, BT is the standard flight temperature of the drone pre-stored in the processor, and the standard flight temperature is set according to the appropriate working temperature of precision electronic devices;

[0150] Compare the avoidance complexity coefficient of the drone in the corresponding obstacle perception area with the avoidance complexity critical value, and compare the environmental interference coefficient of the drone during flight with the environmental interference critical value;

[0151] If the avoidance complexity coefficient of the drone in the corresponding obstacle perception area is less than or equal to the avoidance complexity critical value, and the environmental interference coefficient of the drone during flight is less than or equal to the environmental interference critical value, no operation will be performed;

[0152] If the avoidance complexity coefficient of the drone in the corresponding obstacle perception area is greater than the avoidance complexity critical value or the environmental interference coefficient of the drone during flight is greater than the environmental interference critical value, a path replanning signal is generated;

[0153] The path analysis module sends a path re-planning signal to the processor, and the processor sends the path re-planning signal to the path correction module;

[0154] If the path correction module receives a path re-planning signal, it will correct and optimize the current flight path based on the AI ​​algorithm. It can be understood that the path correction module acts on the special moment when a complex obstacle group appears in the obstacle perception area of ​​the drone. The avoidance strategy of the drone changes from short-distance displacement through obstacles to long-distance detours to avoid buildings.

[0155] In this application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficient and the proportional coefficient can be determined as long as it does not affect the proportional relationship between the parameter and the result value.

[0156] Embodiment 2:

[0157] See also Figure 7 As shown, based on another concept of the same invention, a method for intelligent control of a UAV flight path based on an AI algorithm is proposed, comprising the following steps:

[0158] Step S101, first upload the initial flight path, and perform a shaft smoothness test and a blade tilt dead zone test on the drone body, and generate a test abnormality signal or a test pass signal;

[0159] Step S102: if a test pass signal is generated, the environmental interference data and the signal interference data of the drone are obtained, and the anti-interference strength of the drone is intelligently adjusted;

[0160] Step S103, simultaneously sensing obstacles during the flight of the UAV, generating a horizontal avoidance signal or a vertical avoidance signal, and obtaining an avoidance response level of the obstacle, and controlling the avoidance behavior of the UAV in combination with the horizontal avoidance signal, the vertical avoidance signal or the avoidance response level;

[0161] Step S104, finally obtaining the path obstacle data of the UAV, optimizing the flight route of the UAV in real time, generating a path re-planning signal, and correcting and optimizing the current flight path of the UAV according to the path re-planning signal.

[0162] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. The intelligent control system of UAV flight path based on AI algorithm is characterized by: It includes an airframe self-check module, an environment monitoring module, an anti-disturbance control module, an obstacle perception module, an avoidance execution module, a path analysis module, a path correction module, a control terminal and a processor; the control terminal is used to upload the initial flight path; the airframe self-check module is used to perform a shaft smoothness test and a blade tilt dead zone test on the airframe of the UAV, and generate a test abnormality signal or a test pass signal and send it to the control terminal through the processor; the control terminal is connected to a display, and the display is used to display the test abnormality signal or the test pass signal; When performing flight work according to the initial flight path, the environment monitoring module is used to obtain environmental interference data and signal interference data of the UAV during flight, and the environmental interference data is sent to the path analysis module and the signal interference data is sent to the anti-interference control module via the processor; The anti-interference control module is used to intelligently control the anti-interference strength of the UAV during flight; the obstacle perception module is used to perceive obstacles during the flight of the UAV, generate horizontal avoidance signals or lifting avoidance signals, and obtain the obstacle avoidance response level, and the horizontal avoidance signal or lifting avoidance signal and the obstacle avoidance response level are sent to the avoidance execution module via the processor; the avoidance execution module is used to control the avoidance behavior of the UAV; The obstacle sensing module is also used to obtain the path obstacle data of the UAV during flight, and send the path obstacle data of the UAV to the path analysis module through the processor; the path analysis module is used to optimize the flight route of the UAV in real time, generate a path re-planning signal and send it to the path correction module through the processor; the path correction module corrects and optimizes the current flight path of the UAV; Environmental interference data includes the real-time environmental wind resistance, real-time environmental temperature, real-time environmental humidity and the real-time altitude of the drone during flight; Signal interference data includes the center distance, center interference intensity and interference coverage radius of each potential interference source in the drone's corresponding signal detection area; The path obstacle data includes the avoidance buffer distance, lateral avoidance distance, longitudinal avoidance distance between the UAV and the obstacle, and the real-time flight speed of the UAV; The path obstacle data includes the number of obstacles in the obstacle sensing area corresponding to the drone, as well as the obstacle specifications, obstacle distance and obstacle spacing of each obstacle; The test steps for the shaft smoothness test include: After the rotor (1) is started, the shaft amplitude and shaft temperature of the rotor (1) on the drone are monitored; The shaft vibration amplitude of the rotor wing (1) on the drone is compared with the test vibration amplitude critical value, and at the same time, the shaft temperature of the rotor wing (1) on the drone is compared with the test temperature critical value; If the value of the shaft body amplitude is greater than the test amplitude or the value of the shaft body temperature is greater than the test temperature critical value, a test abnormality signal is generated; If the value of the shaft body amplitude is less than or equal to the test amplitude and the value of the shaft body temperature is less than or equal to the test temperature critical value, the blade tilt dead zone test is performed; The test steps of the blade tilt dead zone test specifically include: A plane rectangular coordinate system is established with the center of the spiral wing (1) as the coordinate origin, the positive direction of the longitudinal axis is taken as the initial position, and the inclination angle of the rotation axis (2) is set to a maximum value for inclination; With the rotating shaft (2) as the center of the circle, the rotating shaft (2) is rotated clockwise for one circle, and jamming during the rotation process is monitored; During the rotation of the rotating shaft (2), if a jam occurs, a test abnormality signal is generated; if no jam occurs, the inclination angle of the rotating shaft (2) is adjusted and the test process is repeated; When the value of the tilt angle of the rotating shaft (2) returns to zero, if no jamming occurs, a test pass signal is generated; The control process of the anti-disturbance control module is as follows: Obtain the center distance and interference coverage radius of each potential interference source in the UAV's corresponding signal detection area; Compare the value of the center distance with the value of the interference coverage radius; If the value of the center distance is greater than the value of the interference coverage radius, the corresponding potential interference source is screened out; If the value of the center distance is less than or equal to the value of the interference coverage radius, the corresponding potential interference source is marked as an actual interference source and proceeds to the next step; Obtain the center distance, interference coverage radius and center interference intensity of the actual interference source in the UAV corresponding signal detection area; Calculate the interference intensity of each actual interference source in the UAV’s corresponding signal detection area; The interference strength of each actual interference source is compared to obtain the maximum interference strength in the UAV's corresponding signal detection area; Compare the maximum interference intensity in the signal detection area corresponding to the drone with the interference intensity threshold; If the value of the interference intensity is less than or equal to the first interference intensity threshold, the anti-interference level of the UAV is adjusted to the first anti-interference level; If the value of the interference intensity is greater than the first interference intensity threshold and less than or equal to the second interference intensity threshold, the anti-interference level of the UAV is adjusted to the second anti-interference level; If the interference intensity value is greater than the third interference intensity threshold, the anti-interference level of the UAV is adjusted to the third anti-interference level; Among them, the values ​​of the first interference intensity threshold and the second interference intensity threshold are both greater than zero, the first interference intensity threshold is less than the second interference intensity threshold, the anti-interference performance of the first anti-interference level is lower than the anti-interference performance of the first anti-interference level, and the anti-interference performance of the second anti-interference level is lower than the anti-interference performance of the third anti-interference level.

2. The AI ​​algorithm-based UAV flight path intelligent control system according to claim 1 is characterized in that: The working process of the obstacle perception module is as follows: The obstacle sensing area corresponding to the UAV is constructed based on the limit sensing distance of the sensing element. When an obstacle enters the obstacle sensing area, the avoidance buffer distance, lateral avoidance distance, longitudinal avoidance distance between the UAV and the obstacle and the real-time flight speed of the UAV are obtained. Calculate the real-time approach rate of the obstacle, and compare the real-time approach rate of the obstacle with the approach rate threshold, and determine whether the avoidance response level of the corresponding obstacle is the first avoidance response level, the second avoidance response level, or the third avoidance response level; wherein the avoidance immediacy of the first avoidance response level is lower than the avoidance immediacy of the second avoidance response level, and the avoidance immediacy of the second avoidance response level is lower than the avoidance immediacy of the third avoidance response level; Then, the translational avoidance displacement and the lifting avoidance displacement of the obstacle are calculated, and the numerical values ​​of the translational avoidance displacement and the lifting avoidance displacement of the obstacle are compared to generate a horizontal avoidance signal or a lifting avoidance signal.

3. The UAV flight path intelligent control system based on AI algorithm according to claim 2 is characterized in that: The working process of the avoidance execution module is as follows: If a horizontal avoidance signal is received, the drone is controlled to avoid obstacles laterally; If the lift avoidance signal is received, the drone is controlled to avoid obstacles longitudinally; When the obstacle avoidance response level is the first avoidance response level, there is no need to change the flight speed of the drone during the avoidance process; When the obstacle avoidance response level is the second avoidance response level, the flight speed of the drone needs to be reduced during the avoidance process; When the obstacle avoidance response level is the third avoidance response level, the drone needs to be controlled to perform emergency braking and avoid the obstacle in time during the avoidance process.

4. The UAV flight path intelligent control system based on AI algorithm according to claim 1 is characterized in that: The analysis and optimization process of the path analysis module is as follows: Obtain the number of obstacles in the obstacle perception area corresponding to the drone, as well as the obstacle specifications and obstacle distances of each obstacle, and calculate the avoidance complexity coefficient in the obstacle perception area corresponding to the drone; Then, the real-time environmental wind resistance, real-time environmental temperature, real-time environmental humidity and real-time altitude of the UAV during flight are obtained to calculate the environmental interference coefficient of the UAV during flight; Compare the avoidance complexity coefficient of the drone in the corresponding obstacle perception area with the avoidance complexity critical value, and compare the environmental interference coefficient of the drone during flight with the environmental interference critical value; If the avoidance complexity coefficient of the drone in the corresponding obstacle perception area is less than or equal to the avoidance complexity critical value, and the environmental interference coefficient of the drone during flight is less than or equal to the environmental interference critical value, no operation will be performed; If the avoidance complexity coefficient of the drone in the corresponding obstacle perception area is greater than the avoidance complexity critical value or the environmental interference coefficient of the drone during flight is greater than the environmental interference critical value, a path replanning signal is generated.

5. The AI ​​algorithm-based UAV flight path intelligent control system according to claim 4 is characterized in that: The correction measures of the path correction module specifically include: If a path re-planning signal is received, the drone's avoidance strategy is adjusted to a long-distance detour to avoid the obstacle group; If no path re-planning signal is received, the drone's avoidance strategy is adjusted to short-distance displacement to pass through obstacles.

6. The intelligent control method of UAV flight path based on AI algorithm is characterized by: According to any one of claims 1 to 5, the method for the intelligent control system of the flight path of a drone based on an AI algorithm specifically comprises: Step S101, uploading the initial flight path, and performing a shaft smoothness test and a blade tilt dead zone test on the drone body, generating a test abnormality signal or a test pass signal; Step S102: if a test pass signal is generated, the environmental interference data and the signal interference data of the drone are obtained, and the anti-interference strength of the drone is intelligently adjusted; Step S103, simultaneously sensing obstacles during the flight of the UAV, generating a horizontal avoidance signal or a vertical avoidance signal, and obtaining an avoidance response level of the obstacle, and controlling the avoidance behavior of the UAV in combination with the horizontal avoidance signal, the vertical avoidance signal or the avoidance response level; Step S104, finally obtaining the path obstacle data of the UAV, optimizing the flight route of the UAV in real time, generating a path re-planning signal, and correcting and optimizing the current flight path of the UAV according to the path re-planning signal.

Images