Cooperative low-altitude small and medium-sized unmanned aerial vehicle anti-collision system

Abstract

The invention discloses a cooperative low-altitude small and medium-sized unmanned aerial vehicle collision avoidance system, and the system comprises a flight path prediction module which is used for receiving the flight data of a local aircraft and the flight information of other targets in a low-altitude airspace, and predicting the flight paths of the local aircraft and other flight targets; the conflict detection module is used for receiving the flight path prediction information of the local machine and other targets sent by the flight path prediction module, receiving the local map information in the task data storage module at the same time, performing conflict judgment according to the prediction information and the map information, and sending a conflict judgment result to the planning anti-collision processing module; and the planning anti-collision processing module is used for receiving the conflict judgment result, if no conflict exists, task execution is carried out according to an original plan, and if the conflict exists, anti-collision planning processing is carried out and anti-collision planning is carried out. Operation rule support and safety guarantee are provided for low-altitude operation, and meanwhile the low-altitude airspace utilization rate is increased.

Description

Technical Field

[0001] The present application relates to the field of low-altitude UAV flight safety technology, and in particular to a cooperative low-altitude small UAV collision avoidance system. Background Art

[0002] The safety of low-altitude UAV flight is of vital importance to the effective use of low-altitude airspace and socio-economic development, and is an essential element for UAVs to play a role in low-altitude operations. Compared with high-altitude flight, low-altitude airspace is full of various static and dynamic obstacles. The randomness and unpredictability of these factors place extremely high demands on the perception, autonomous path planning, and collision avoidance capabilities of UAVs. Secondly, the limitations of small UAVs themselves also pose a huge safety hazard. Due to their size, weight, and cost, small and medium-sized UAVs are usually equipped with relatively limited sensors and computing power, which makes it difficult for them to accurately perceive the surrounding environment, especially the lack of recognition and response capabilities to fast-moving objects or subtle obstacles.

[0003] The difficulties faced by the flight safety and collision avoidance technology of small UAVs at low altitudes are concentrated in several aspects such as perception, planning and coordination. First of all, accurate perception is the basis of safe flight. In a complex and changeable low-altitude environment, UAVs need to be able to accurately and quickly perceive the surrounding environment, including static and dynamic obstacles. Secondly, UAVs need to plan safe and reliable flight paths in real time based on the perceived environmental information, and be able to effectively avoid obstacles. This requires an efficient path planning algorithm that can find the best path within limited computing resources and time, and can adapt to dynamic changes in the environment. Finally, multi-UAV collaborative obstacle avoidance is an important direction for future development, which requires effective communication and collaboration between UAVs to avoid collisions with each other. Solving the above problems is of great significance for establishing and improving the low-altitude operation system, improving the utilization rate of low-altitude airspace, and improving the efficiency of low-altitude operations. Summary of the invention

[0004] In view of this, the present application provides a cooperative low-altitude small UAV collision avoidance system, which aims to provide operating rules support and safety guarantees for low-altitude operations, while improving the utilization rate of low-altitude airspace.

[0005] The present application discloses a cooperative low-altitude small UAV collision avoidance system, which includes: The local data receiving module is used to receive the local flight data sent by the local flight control data system. The local flight data includes position information, speed information, flight attitude information and local identity information. The target data receiving module is used to receive target flight information broadcasted by other flying targets other than the aircraft in low-altitude airspace; The track prediction module is used to receive the flight data of the aircraft and the flight information of other targets in the low-altitude airspace, and predict the flight tracks of the aircraft and other flight targets.

[0006] The conflict detection module is used to receive the track prediction information of the aircraft and other targets sent by the track prediction module, and at the same time receive the local map information in the mission data storage module, make conflict judgments based on the prediction information and map information, and send the conflict judgment results to the planning and collision avoidance processing module; The planning and collision avoidance processing module is used to receive the conflict judgment results. If there is no conflict, the task is executed according to the original plan. If there is a conflict, the collision avoidance planning processing is performed and the collision avoidance planning is executed.

[0007] Furthermore, the track prediction module includes: A historical track generation unit is used to generate the historical tracks of the aircraft and the target aircraft based on the historical position information and the drone ID information; A model building unit is used to build flight models of the own aircraft and the target aircraft according to the own aircraft's flight plan and the target aircraft's historical track and flight status. The flight models include a uniform motion model, a uniform acceleration motion model and a turning model. The track vector generating unit is used to predict the discrete tracks of the own aircraft and the target aircraft and generate the discrete track vectors of the own aircraft and the target aircraft.

[0008] Furthermore, the track vector generating unit is used for: When the flight models of the aircraft and the target aircraft are uniform motion models, for time :

[0009] If there is a height restriction, After reaching the altitude limit, it remains unchanged. Indicates that the local machine and the target machine are The position of the machine at the moment, where the origin of the coordinate system is the machine's current position at the moment, that is, the machine's current position at the moment is ; The coordinate system uses the ENU coordinate system ， Respectively represent the north velocity, east velocity and vertical velocity, M is a positive integer, , , Respectively represent the local machine and the target machine in The northing coordinate, easting coordinate and vertical coordinate of the moment, 、 , are the coordinates at the origin of the coordinate system; When the flight models of the aircraft and the target aircraft are uniform acceleration models, :

[0010]

[0011]

[0012] If during the acceleration process, the speed has reached the upper limit of the drone's performance, the uniform acceleration model is converted to a uniform speed model; 、 、 They are the north acceleration, east acceleration, and vertical acceleration of the own aircraft and the target aircraft respectively; When the flight model of the aircraft and the target aircraft is a turning model, 、 、 For variables, :

[0013]

[0014]

[0015] in, t For time; According to the flight plan and historical track information, the flight models of the aircraft and the target meet the time-sharing combination of uniform motion model, uniform acceleration motion model and turning model; According to the above process, the discrete track vectors of the own aircraft and the target aircraft within M seconds are generated, which can be expressed as:

[0016] in, , , They are respectively the north coordinate, east coordinate and vertical coordinate of the local or target aircraft at M seconds.

[0017] Furthermore, the conflict detection module includes: A track vector acquisition unit is used to input the prediction vector of the aircraft and traverse the prediction vectors of all targets within the receiving range; The collision risk judgment unit is used to calculate the distance and altitude difference between the own aircraft and the target aircraft at the same time according to the discrete track vectors of the own aircraft and the target aircraft, and judge whether there is a collision risk between the own aircraft and the target aircraft according to the distance and altitude difference between the own aircraft and the target aircraft.

[0018] Furthermore, the collision risk judgment unit is used to: Calculate the distance between the local aircraft and the target aircraft respectively according to the following formula and the altitude difference between the local aircraft and the target aircraft :

[0019]

[0020] wherein, N, N is less than M, and N is a positive integer; the vector represents the predicted information of the local aircraft at times 1, 2,..., N, represents the predicted information of the target aircraft at times 1, 2,..., N; is the absolute value function; 、 、 respectively represent the north coordinate, east coordinate and vertical coordinate of the local aircraft at time N; 、 、 respectively represent the north coordinate, east coordinate and vertical coordinate of the target aircraft at time N; If there exists a time 0 < t < N + 1, where t is time and t is a positive integer, satisfying one of the following conditions: and

[0021] and

[0022] then it is determined that there is a risk of conflict between the local aircraft and the target in the airspace, and path planning and anti-collision maneuvers are performed. Otherwise, it is considered that there is no risk of collision, and the flight mission continues to be executed according to the original plan; wherein, represents the horizontal speed of the local aircraft, represents the horizontal speed of the target aircraft, represents the distance between the local aircraft and the target aircraft at time t, is the maximum value function, represents the altitude difference between the local aircraft and the target aircraft.

[0023] Furthermore, the collision risk judgment unit is further configured to: Regarding the collision risk between the local aircraft and the building, compare the discrete flight track vector of the local aircraft with the map information in the mission data storage module. If at any time t, with as the center, the anti-collision protection area A intersects non-emptily with the area B occupied by any building, that is, satisfies , it is considered that there is a risk of collision between the aircraft and the building, and path planning and collision avoidance maneuvers are required; otherwise, there is no risk of collision; The collision protection area is a cylinder built with the aircraft or target as the center; the diameter of the cylinder is determined by the horizontal speed of the aircraft or target and is defined as:

[0024] in, Indicates the horizontal flight speed of the aircraft or target. It indicates the distance generated by the aircraft / drone when flying at the current speed for a preset time T. Generally, T is set to 4 seconds and M is set to 20 meters. The horizontal diameter of the anti-collision protection area is related to the track prediction frequency, conflict detection mechanism, drone position and speed accuracy, and path planning mechanism. Sets the height of the collision protection area cylinder to a fixed value.

[0025] Furthermore, the planning collision avoidance processing module includes: The execution path planning unit is used to adopt a path planning method of finding connected areas, adopt discretization to plan the collision avoidance path, establish multiple optional discrete tracks through discretization of the three dimensions of azimuth, altitude and speed, and determine whether they are connected, and select the connected path as the execution path; Maneuver and recovery unit, used to execute maneuvers and complete post-collision avoidance recovery flight plans.

[0026] Furthermore, the execution path planning unit is used to: When the aircraft predicts a collision risk with other target aircraft or buildings in the airspace, it is necessary to replan the path and design the target coordination mechanism and track recovery method. The process is as follows: Set the bearing threshold interval , height threshold interval and speed threshold interval They respectively represent the azimuth range, altitude range and speed range that the aircraft can select when performing collision avoidance maneuvers; The azimuth threshold interval, altitude threshold interval and speed threshold interval are discretized on average. Suppose there are M azimuth options, N altitude options and Q speed options after discretization. The traversed track paths are strip; Traverse each optional track, the track is used express, Indicates that the selected direction changes to , a track with a height change of h and a speed change of v; The traversal method is in lexicographic order, with azimuth first, altitude second, and speed last. That is, priority is given to traversing tracks where the aircraft's altitude and speed do not change and only maneuvers in azimuth, and priority is given to traversing tracks around the aircraft's current heading. For the traversed tracks , the trajectory of the aircraft is predicted, that is, the aircraft maneuvers within the performance range to reach the maneuvering target value, that is, the heading, altitude and speed reach Then, the track prediction value of the uniform motion process is obtained; and it is determined whether the track prediction value will cause a collision risk; If the judgment result is that there is no collision risk, that is, the track is a collision avoidance safety connection area, the track is used as the collision avoidance path and the next step is executed. If the judgment result is that there is a collision risk, the next optional track is traversed.

[0027] Furthermore, the maneuvering and recovery unit is used to: After the anti-collision path is set, determine the speed of the aircraft Speed ​​of the target machine to be avoided If the relationship between , the aircraft will immediately perform the maneuver according to the collision avoidance path and broadcast the collision avoidance status as performing the collision avoidance maneuver; if , the aircraft will not perform maneuvers temporarily and will immediately perform track prediction, conflict detection and collision avoidance track traversal; If the aircraft does not perform a maneuver, the aircraft will continue to perform track prediction and conflict detection. If the target aircraft still does not broadcast the anti-collision state as an anti-collision maneuver before the conflict occurs and there is still a risk of anti-collision, the aircraft will immediately perform a maneuver and broadcast the anti-collision state. During the execution of the collision avoidance path, the path between the current path point and the planned path point is calculated and predicted, the point in the planned path point closest to the current path point and the point near the closest point in the planned path are found, and the track from the current path point to the closest point in the planned path and its nearby points is planned. If the planned track has no conflict risk and the collision avoidance with the conflicting target has been completed when the planned point is reached, the maneuver is performed to complete the planned track recovery. If there is a conflict, continue to operate according to the traversed track until the recovery conditions are met to restore the planned track.

[0028] Furthermore, it also includes: The aircraft data broadcast module is used to send the aircraft's flight data and anti-collision status to the outside world through broadcasting; The mission data storage module is used to store the local map information required for the UAV to perform flight missions.

[0029] Due to the adoption of the above technical solution, the present application has the following advantages: 1) The anti-collision protection area, conflict detection, anti-collision path planning method and coordination mechanism are set up to improve the flight safety index of small UAVs in low altitude; 2) It can provide basic flight operation method support for UAV flight management; 3) It can improve the efficiency of airspace use and airspace operation; 4) Since discretization is used to reduce computational complexity, all modules can be designed in an integrated hardware manner, reducing costs and power consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0031] Figure 1 A schematic diagram of a cooperative low-altitude small UAV collision avoidance system according to an embodiment of the present application; Figure 2 A schematic diagram of an anti-collision protection area according to an embodiment of the present application; Figure 3 A schematic diagram of the anti-collision execution effect of an embodiment of the present application. DETAILED DESCRIPTION

[0032] The present application is further described in conjunction with the accompanying drawings and embodiments, and the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field should fall within the scope of protection of the embodiments of the present application.

[0033] See also Figure 1 The present application provides an embodiment of a cooperative low-altitude small UAV collision avoidance system, which includes: The local data receiving module is used to receive the local flight data sent by the local flight control data module, including position information, speed information, flight posture information, and local identity information, which is used for subsequent flight conflict prediction and collision avoidance processing. At the same time, the corresponding information can be sent to the local data broadcasting module.

[0034] The target data receiving module is used to receive the target flight information broadcasted by other flying targets except the aircraft in the low-altitude airspace, and send the received target flight data to the track prediction module for track prediction and conflict detection.

[0035] The track prediction module is used to receive the flight data of the aircraft and the flight information of other targets in the low-altitude airspace, and use the track prediction algorithm to predict the flight trajectory of the aircraft and other flying targets.

[0036] The conflict detection module receives the track prediction information of the aircraft and other targets sent by the track prediction module, and at the same time receives the local map information in the mission data storage module, makes conflict judgments based on the prediction information and map information, and sends the conflict judgment results to the planning and collision avoidance processing module for collision avoidance planning and processing.

[0037] The planning and collision avoidance processing module receives the flight prediction information of the aircraft and the target. If there is no conflict, the mission is executed according to the original plan. If there is a conflict, the collision avoidance planning processing is performed and the collision avoidance plan is executed.

[0038] Optionally, this embodiment may also include a local data broadcast module to send the local flight data and anti-collision status of the aircraft to the outside through broadcasting; the mission data storage module is used to store the local map information required for the UAV to perform the flight mission. The information is stored before the flight mission is executed, and the information is local, so the amount of stored data can be reduced.

[0039] According to the relevant national documents on small and medium-sized UAVs, UAVs need to send broadcast operation identification messages during operation, including the UAV's ID information, position vector information, operation description information, system information, etc.

[0040] The drone ID information can be used as a unique identifier for the drone to distinguish different drone devices. The position vector information is periodic information, including but not limited to latitude and longitude, altitude, speed, track angle, horizontal accuracy, vertical accuracy, speed accuracy, etc. The above information is the input for track prediction.

[0041] Operation description information is periodic information used by drone operators and operators to express their flight objectives or carry any other supplementary information. System information is used to describe the location and altitude of the drone control station, aircraft groups, and additional system information. Track prediction is in seconds, predicting the discrete track information of the aircraft and the target within 20 seconds that can be received. That is, a 20-dimensional vector is generated for the aircraft and the target to represent the predicted position information of the aircraft and the target within the next 0-20 seconds. The prediction result is used as the input of conflict detection.

[0042] The input of track prediction is the drone ID information and position vector information. The drone ID information is used for track association, and the drone position vector information is used for track generation and prediction. The track prediction module is used to: a) Generate the historical flight tracks of the aircraft and the target aircraft based on the historical location information and the drone ID information; b) Establish the flight models of the own aircraft and the target aircraft according to the own aircraft's flight plan and the target aircraft's historical track and flight status. The models include uniform motion model, uniform acceleration motion model and turning model; c) Predict the discrete track of the aircraft and the target aircraft. For the uniform speed model, the prediction method is: for

[0043]

[0044] If there is a height restriction, After reaching the height limit, it remains unchanged. Indicates that the local / target machine is The position of the machine at the moment, where the origin of the coordinate system is the machine's current position at the moment, that is, the machine's current position at the moment is , here it is not written as (0,0,0) because the above formula not only represents the motion model of the aircraft, but also the motion model of the target; the coordinate system uses the ENU coordinate system, , , They represent north velocity, east velocity and vertical velocity respectively; , , Respectively represent the local machine and the target machine in The northing coordinate, easting coordinate and vertical coordinate of the moment, 、 , are the coordinates at the origin of the coordinate system; In the case of uniform acceleration model, for

[0045]

[0046]

[0047]

[0048] If during the acceleration process, the speed has reached the upper limit of the drone's performance, it will be converted to a uniform speed model; 、 、 They are the north acceleration, east acceleration, and vertical acceleration of the own aircraft and the target aircraft respectively; In the turning model, 、 、 For variables,

[0049]

[0050]

[0051]

[0052] in, t For time; According to the flight plan and historical track information, the aircraft and target motion models satisfy the time-sharing combination of these three models; d) According to the above process, the discrete track vector of the aircraft / target within 20 seconds is generated, which is expressed as

[0053] in, , , They are respectively the north coordinate, east coordinate and vertical coordinate of the local or target aircraft at 20 seconds.

[0054] e) The track prediction frequency shall not be less than 1 Hz.

[0055] The above track prediction method uses a discretization method to predict with 1 second as the minimum unit, and the prediction period is limited to 1Hz (updated once every 1s). The error caused by the discretization of track prediction and the discretization of prediction period corresponds to the diameter of the established anti-collision protection area, ensuring the safety of UAV flight.

[0056] The conflict detection module is specifically used for: A1) Input the local prediction vector ; B1) Traverse the prediction vectors of all targets within the receiving range

[0057] C1) According to the following 15 ways to calculate the distance between the local machine and the target machine and the height difference between the host and the target machine :

[0058]

[0059] Among them, the vector Indicates that the machine is 15-hour forecast information, Represents the prediction information of the target machine at time 1, 2, ...15; is the absolute value function; 、 、 They represent the northing coordinate, easting coordinate and vertical coordinate of the aircraft at time N respectively; 、 、 respectively represent the north coordinate, east coordinate and vertical coordinate of the target aircraft at time 15; D1) If there exists a time 0 < t < 16, where t is time and t is a positive integer, satisfying one of the following conditions and

[0060] and

[0061] then it is determined that there is a risk of conflict between the aircraft and the target in the airspace, and path planning and anti-collision maneuvers need to be performed. Otherwise, it is considered that there is no collision risk, and the flight mission continues to be executed according to the original plan; among them, represents the horizontal speed of the aircraft, represents the horizontal speed of the target aircraft, represents the distance between the aircraft and the target aircraft at time t, is the maximum value function, represents the height difference between the aircraft and the target aircraft.

[0062] E1) For the risk of collision with buildings, input the predicted vector of the aircraft

[0063] and compare it with the map information in the mission data storage module. If at a certain time t, the anti-collision protection area A centered on intersects non-emptily with the area B occupied by a certain building, that is, satisfies

[0064] then it is considered that there is a risk of collision between the aircraft and the building and path planning and anti-collision maneuvers are required. Otherwise, there is no risk.

[0065] The frequency of conflict prediction is consistent with the frequency of track prediction and is not less than 1 Hz. The reason for the above conflict prediction time of 15 s and track prediction time of 20 s is that in most cases, early conflict prediction is an important sign of excellent system performance. Therefore, the conflict prediction can be consistent with the track prediction and increased to 20 s or longer. Here, 15 s is the performance lower limit.

[0066] The planned anti-collision processing module is used for: After track prediction and conflict prediction, if there is no risk of collision between the aircraft and other targets and buildings in the airspace, the aircraft continues to execute according to the flight plan; if there is a risk of collision between the aircraft and other targets or buildings in the airspace, then the anti-collision path needs to be re-planned, and after anti-collision, the flight plan is restored.

[0067] This application adopts a path planning method to find connected areas, and uses discretization to plan collision avoidance paths to reduce the amount of calculation. By discretizing the three dimensions of azimuth, altitude, and speed, multiple optional discrete tracks are established to determine whether they are connected, and the connected path is selected as the execution path. After executing the maneuver and completing the collision avoidance, the flight plan needs to be quickly restored.

[0068] The planning collision avoidance processing module is specifically used for: When the aircraft predicts a collision risk with other flying targets or buildings in the airspace, it is necessary to replan the path, design a coordination mechanism between targets and a track recovery method, and execute the following process: A2) Setting the position threshold range , height threshold interval and speed threshold interval They respectively represent the azimuth range, altitude range and speed range that the aircraft can select when performing collision avoidance maneuvers; B2) Average discretization is performed on the position threshold interval, height threshold interval and speed threshold interval. Suppose there are M position options, N height options and Q speed options after discretization. Then the track paths to be traversed are strip; C2) traverse each optional track, and the track is used express, Indicates that the selected direction changes to , a track with a height change of h and a speed change of v; D2) The traversal method is in lexicographic order, with azimuth first, altitude second, and speed last. That is, priority is given to traversing tracks where the aircraft's altitude and speed do not change and only maneuvers in azimuth, and priority is given to traversing tracks near the aircraft's current heading; E2) For the traversed tracks , predict the flight path of the aircraft, that is, the aircraft maneuvers within the performance range to reach the maneuvering target value (i.e. the heading, altitude and speed reach ) and then the track prediction value within 20 seconds of the uniform motion process

[0069] F2) Determine the track prediction value Whether there is a risk of collision, that is, whether there is So that:

[0070]

[0071] Or if at a certain time t, The collision protection area A centered on is intersected with the area B occupied by a certain building and is not empty, that is,

[0072] G2) If the result of the previous step is that there is no collision risk, that is, the path is a collision-avoiding safety connected area (with no dangerous area blocking in the middle), then the path is used as the collision-avoiding path and the next step is executed. If the result of the previous step is that there is a collision risk, then step C2) is executed to continue traversing; H2) After the anti-collision path is set, determine the speed of the aircraft Speed ​​of the target machine to be avoided If the relationship between , the aircraft will immediately perform the maneuver according to the collision avoidance path and broadcast the collision avoidance status as performing the collision avoidance maneuver; if , the aircraft will not perform maneuvers temporarily and will immediately perform track prediction, conflict detection and collision avoidance track traversal; I2) If the aircraft does not perform a maneuver, the aircraft continues to perform track prediction and conflict detection. If the target aircraft still does not broadcast the anti-collision state as an anti-collision maneuver 8 seconds before the conflict time and there is still a risk of anti-collision, the aircraft immediately performs a maneuver and broadcasts the anti-collision state. J2) During the execution of the collision avoidance path, the path between the current path point and the planned path point is calculated and predicted in real time, the point closest to the current point and the points within 5s (discrete in seconds) of the planned path point are found, and the track between the current point and the above points is planned. If the planned track has no conflict risk and the collision avoidance with the conflicting target is completed when the planned point is reached, the maneuver is performed to complete the planned track recovery. If there is a conflict, continue to operate according to the traversed track until the recovery conditions are met to recover the planned track.

[0073] The above path planning process is based on the discretized vector threshold judgment method, which searches for connected tracks without collision risk to perform maneuvers. It has the advantages of small calculation amount and easy implementation, and is suitable for engineering applications. Track prediction, conflict prediction and path planning are all discretized in seconds, and the total error estimate is strictly less than , thus forming a system with the anti-collision protection area, the execution effect is as follows Figure 3 As shown, v Indicates speed, Indicates the direction chosen by the aircraft to perform the collision avoidance maneuver. Indicates the lower limit of the azimuth that can be selected for performing collision avoidance maneuvers. Indicates the upper limit of the possible orientations for performing collision avoidance maneuvers. , Determined by the machine's maneuverability.

[0074] In summary, the cooperative low-altitude small UAV collision avoidance system of the present application performs the following steps: A3) Setting the UAV mission and loading the local map data into the mission data storage module according to the UAV mission; B3) The local data receiving module receives the local flight data, and the target data receiving module receives the broadcast data of other targets in the airspace; C3) The receiving module sends the received data to the track prediction module, and the track prediction module and the conflict detection module perform track prediction and conflict detection; D3) If the conflict detection module outputs that there is no collision risk, the flight plan will continue to be executed. If the conflict detection module detects a collision risk, the collision avoidance processing module will be planned to execute; E3) The machine broadcasts its own location information, ID information, and anti-collision status information; F3) Repeat steps B3) to E3) during the flight until the mission is completed and the flight ends.

[0075] The flight collision avoidance protection area in the above-mentioned embodiments of the present application is a constraint condition for flight conflict detection and collision avoidance planning. It aims to define a three-dimensional space area. If the aircraft enters the flight collision avoidance protection area of ​​other targets during flight or other targets enter the flight collision avoidance protection area of ​​the aircraft, it is judged as a flight conflict. The flight collision avoidance protection area is not a static concept, but a dynamically adjusted and multi-dimensionally defined space area. Its goal is to minimize the risk of collision during the flight of the drone. It is the theoretical basis and basis for track prediction and conflict judgment. The present application gives the following definition of the flight collision avoidance protection area based on geographic space, characteristics of the aircraft and target aircraft, and risk assessment: The flight collision avoidance protection area of ​​the aircraft / target refers to a dynamically changing area, which establishes a spatial cylinder with the aircraft / target as the center, such as Figure 2 shown.

[0076] The diameter of the collision protection area cylinder is determined by the horizontal velocity of the aircraft / target and is defined as:

[0077] in, Indicates the horizontal flight speed of the aircraft / target. Indicates the distance generated by the aircraft / drone flying at the current speed for 4 seconds. The 20-meter threshold is adjustable. The low-altitude operation department or manufacturer can adjust the relevant threshold according to their own situation; the horizontal diameter of the anti-collision protection area is selected as The reason is related to the track prediction frequency, conflict detection mechanism, UAV position and speed accuracy, and path planning mechanism. Therefore, the definition of the anti-collision protection area is dynamically changing and is related to both the target characteristics and the geographic space; and from the definition, the faster the target speed, the larger the anti-collision protection area, so a risk assessment is performed. The height of the anti-collision protection area cylinder is set to a fixed value, and the threshold is temporarily set to 20 meters. The low-altitude operation department or manufacturer can set the threshold based on factors such as the characteristics of the aircraft / target, airspace characteristics, and mission type.

[0078] This application uses the flight collision avoidance protection area as the conceptual basis and uses a cooperative approach to build a low-altitude safe flight system architecture and method that integrates perception, planning and coordination; The present application is based on a method for trajectory prediction, conflict prediction and connected area path planning designed using discretized azimuth, altitude and speed in an anti-collision protection area. Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present application can still be modified or replaced by equivalents, and any modification or equivalent replacement that does not depart from the spirit and scope of the present application should be included in the scope of protection of the claims of the present application.

Claims

1. A cooperative low-altitude small UAV collision avoidance system, characterized in that: Including: The local data receiving module is used to receive the local flight data sent by the local flight control data system. The local flight data includes position information, speed information, flight attitude information, and local identity information; The target data receiving module is used to receive the target flight information broadcast by other flight targets in the low-altitude airspace except the local aircraft; The flight path prediction module is used to receive the local flight data and the flight information of other targets in the low-altitude airspace, and predict the flight paths of the local aircraft and other flight targets; The conflict detection module is used to receive the predicted information of the local aircraft and other target flight paths sent by the flight path prediction module, and at the same time receive the local map information in the task data storage module, perform conflict judgment according to the predicted information and the map information, and send the conflict judgment result to the planning anti-collision processing module; The planning anti-collision processing module is used to receive the conflict judgment result. If there is no conflict, the task will be executed according to the original plan. If there is a conflict, anti-collision planning processing will be carried out and the anti-collision plan will be executed.

2. The cooperative low-altitude small UAV collision avoidance system according to claim 1 is characterized in that: The said flight path prediction module includes: The historical flight path generation unit is used to generate the historical flight paths of the local aircraft and the target aircraft according to the historical position information and the UAV ID information; The model establishment unit is used to establish the flight models of the local aircraft and the target aircraft according to the local flight plan, the historical flight path of the target aircraft, and the flight state. The flight models include uniform motion model, uniformly accelerated motion model, and turning model; The flight path vector generation unit is used to perform discrete flight path prediction on the local aircraft and the target aircraft, and generate the discrete flight path vectors of the local aircraft and the target aircraft.

3. The cooperative low-altitude small UAV collision avoidance system according to claim 2 is characterized in that: The said flight path vector generation unit is used for: When the flight models of the aircraft and the target aircraft are uniform motion models, for time : If there is a height restriction, After reaching the altitude limit, it remains unchanged. Indicates that the local machine and the target machine are The position of the machine at the moment, where the origin of the coordinate system is the machine's current position at the moment, that is, the machine's current position at the moment is ; The coordinate system uses the ENU coordinate system ， Respectively represent the north velocity, east velocity and vertical velocity, M is a positive integer, , , Respectively represent the local machine and the target machine in The northing coordinate, easting coordinate and vertical coordinate of the moment, 、 , are the coordinates at the origin of the coordinate system; When the flight models of the aircraft and the target aircraft are uniform acceleration models, : If during the acceleration process, the speed has reached the upper limit of the drone's performance, the uniform acceleration model is converted to a uniform speed model; 、 、 They are the north acceleration, east acceleration, and vertical acceleration of the own aircraft and the target aircraft respectively; When the flight model of the aircraft and the target aircraft is a turning model, 、 、 For variables, : in, t For time; According to the flight plan and historical flight path information, the flight models of the local aircraft and the target satisfy the time-sharing combination of the uniform motion model, the uniformly accelerated motion model, and the turning model; Generate the discrete flight path vectors of the local aircraft and the target aircraft within M seconds according to the above process, expressed as: in, , , They are respectively the north coordinate, east coordinate and vertical coordinate of the local or target aircraft at M seconds.

4. The cooperative low-altitude small UAV collision avoidance system according to claim 1 is characterized in that: The said conflict detection module includes: The flight path vector acquisition unit is used to input the predicted vector of the local aircraft and traverse the predicted vectors of all targets within the receiving range; The collision risk judgment unit is used to calculate the distance and height difference between the local aircraft and the target aircraft at the same time according to the discrete flight path vectors of the local aircraft and the target aircraft, and judge whether there is a collision risk between the local aircraft and the target aircraft according to the distance and height difference between the local aircraft and the target aircraft.

5. The cooperative low-altitude small UAV collision avoidance system according to claim 4 is characterized in that: The said collision risk judgment unit is used for: According to the following formula, calculate the distance between the local machine and the target machine respectively. and the height difference between the host and the target machine : in, N, N is less than M, N is a positive integer; vector Indicates the prediction information of the machine at time 1, 2, ...N, Represents the prediction information of the target machine at time 1, 2, ...N; is the absolute value function; 、 、 They represent the northing coordinate, easting coordinate and vertical coordinate of the aircraft at time N respectively; 、 、 Respectively represent the north coordinate, east coordinate and vertical coordinate of the target aircraft at time N; If there exists a moment 0 < t < N + 1, where t is time and t is a positive integer, satisfying one of the following conditions: and and If the aircraft is deemed to have a risk of collision with a target in the airspace, path planning and collision avoidance maneuvers will be performed; otherwise, it will be deemed to have no risk of collision and the flight mission will continue to be performed as originally planned. Indicates the horizontal speed of the aircraft. Indicates the horizontal speed of the target aircraft, Indicates the distance between the local machine and the target machine at time t, is the maximum value function, Indicates the altitude difference between the host and the target machine.

6. The cooperative low-altitude small UAV collision avoidance system according to claim 4 is characterized in that: The said collision risk judgment unit is also used for: To prevent the collision risk between the aircraft and the building, the discrete track vector of the aircraft is Compare with the map information in the task data storage module. If at any time t, The collision protection area A centered on is not empty when it intersects with the area B occupied by any building, that is, , it is considered that there is a risk of collision between the aircraft and the building, and path planning and collision avoidance maneuvers are required; otherwise, there is no risk of collision; The anti-collision protection area is a cylinder established with the local aircraft or the target as the center; the diameter of the cylinder is determined by the horizontal speed of the local aircraft or the target, defined as: in, Indicates the horizontal flight speed of the aircraft or target. Indicates that the aircraft / drone is moving at the current speed The distance generated by the preset flight time T; the horizontal diameter of the collision protection area is related to the track prediction frequency, conflict detection mechanism, drone position and speed accuracy, and path planning mechanism; Set the height of the cylinder in the anti-collision protection area as a fixed value.

7. The cooperative low-altitude small UAV collision avoidance system according to claim 1 is characterized in that: The said planning anti-collision processing module includes: The execution path planning unit is used to adopt the path planning method of finding the connected area, discretize the anti-collision path, establish multiple optional discrete flight paths by discretizing the three dimensions of azimuth, height, and speed, and judge whether they are connected, and select the connected path as the execution path; The maneuvering and recovery unit is used to perform maneuvers and complete the recovery of the flight plan after anti-collision.

8. The cooperative low-altitude small UAV collision avoidance system according to claim 7 is characterized in that: The said execution path planning unit is used for: When the aircraft predicts a collision risk with other target aircraft or buildings in the airspace, it is necessary to replan the path and design the target coordination mechanism and track recovery method. The process is as follows: Set the bearing threshold interval , height threshold interval and speed threshold interval They respectively represent the azimuth range, altitude range and speed range that the aircraft can select when performing collision avoidance maneuvers; The azimuth threshold interval, altitude threshold interval and speed threshold interval are discretized on average. Suppose there are M azimuth options, N altitude options and Q speed options after discretization. The traversed track paths are strip; Traverse each optional track, the track is used express, Indicates that the selected direction changes to , a track with a height change of h and a speed change of v; The traversal method is in lexicographic order, with azimuth first, altitude second, and speed last. That is, priority is given to traversing tracks where the aircraft's altitude and speed do not change and only maneuvers in azimuth, and priority is given to traversing tracks around the aircraft's current heading. For the traversed tracks , the trajectory of the aircraft is predicted, that is, the aircraft maneuvers within the performance range to reach the maneuvering target value, that is, the heading, altitude and speed reach Then, the track prediction value of the uniform motion process is obtained; and it is determined whether the track prediction value will cause a collision risk; If the judgment result is that there is no collision risk, that is, the track is a collision avoidance safety connection area, the track is used as the collision avoidance path and the next step is executed. If the judgment result is that there is a collision risk, the next optional track is traversed.

9. The cooperative low-altitude small UAV collision avoidance system according to claim 7 is characterized in that: The maneuvering and recovery unit is used to: After the anti-collision path is set, determine the speed of the aircraft The speed of the target machine to be avoided If the relationship between , the aircraft will immediately perform the maneuver according to the collision avoidance path and broadcast the collision avoidance status as performing the collision avoidance maneuver; if , the aircraft will not perform maneuvers temporarily and will immediately perform track prediction, conflict detection and collision avoidance track traversal; If the aircraft does not perform a maneuver, the aircraft will continue to perform track prediction and conflict detection. If the target aircraft still does not broadcast the anti-collision state as an anti-collision maneuver before the conflict occurs and there is still a risk of anti-collision, the aircraft will immediately perform a maneuver and broadcast the anti-collision state. During the execution of the collision avoidance path, the path between the current path point and the planned path point is calculated and predicted, the point in the planned path point closest to the current path point and the point near the closest point in the planned path are found, and the track from the current path point to the closest point in the planned path and its nearby points is planned. If the planned track has no conflict risk and the collision avoidance with the conflicting target has been completed when the planned point is reached, the maneuver is performed to complete the planned track recovery. If there is a conflict, continue to operate according to the traversed track until the recovery conditions are met to restore the planned track.

10. The cooperative low-altitude small UAV collision avoidance system according to claim 1 is characterized in that: Also includes: The aircraft data broadcast module is used to send the aircraft's flight data and anti-collision status to the outside world through broadcasting; The mission data storage module is used to store the local map information required for the UAV to perform flight missions.

Images