Collision probability determination method and device and storage medium

Abstract

The invention discloses a collision probability determination method and apparatus, and a storage medium. The method comprises the steps of obtaining current state data of each unmanned aerial vehicle in a plurality of unmanned aerial vehicles; the current state data comprises a current position and a speed; determining first position information of each unmanned aerial vehicle at different moments according to the current state data of each unmanned aerial vehicle; according to the first position information of each unmanned aerial vehicle at different moments, the collision probability between the unmanned aerial vehicles is calculated; therefore, the future position of the unmanned aerial vehicle can be predicted, and the collision probability between the unmanned aerial vehicles is calculated based on the predicted position.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicles, and in particular to a collision probability determination method, device and storage medium. Background Art

[0002] With the continuous development of science and technology, drone technology has become more mature. Its small size, flexible flight, easy operation and strong adaptability have made drones occupy a place in logistics transportation, agricultural planting, disaster relief and other application scenarios. As drone applications become more and more extensive, the safety issues that have gradually emerged in recent years have also become the focus of public attention. How to prevent drones from colliding in the air needs to be solved urgently. Summary of the invention

[0003] In view of this, the main purpose of the present invention is to provide a method, device and storage medium for determining collision probability.

[0004] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0005] An embodiment of the present invention provides a method for determining a collision probability, the method comprising:

[0006] Acquire current status data of each of the plurality of drones; the current status data includes current position and speed;

[0007] Determine the first position information of each drone at different times according to the current state data of each drone;

[0008] The collision probability between the drones is calculated according to the first position information of each drone at different times.

[0009] In the above solution, determining the first position information of each drone at different times according to the current state data of each drone includes:

[0010] Determine, according to the current state data of the drone, a flight distance of the drone in a first time period starting from the current time;

[0011] Based on the current position of the drone, projecting the flight distance of the drone in the first time period onto a predetermined flight route of the drone;

[0012] Determine first position information of the UAV on the predetermined flight route at each moment in a first time period.

[0013] In the above solution, determining the flight distance of the drone in a first time period starting from the current time according to the current state data of the drone includes:

[0014] According to the current state data of the drone, the acceleration and the corrected state data of the drone are determined by Kalman filtering; the corrected state data includes a corrected position and a corrected speed;

[0015] The flight distance of the UAV in the first time period is determined according to the acceleration, the corrected position and the corrected speed.

[0016] In the above scheme, the method further includes:

[0017] According to the motion model of the UAV, a motion equation corresponding to the UAV is established;

[0018] Determine a state equation of the UAV according to the motion equation; the state variables of the state equation include the position, velocity and acceleration of the UAV;

[0019] Determining a measurement equation of the UAV according to the state equation;

[0020] The Kalman filter is established according to the state equation and the measurement equation.

[0021] In the above scheme, determining the flight distance of the UAV in the first time period starting from the current moment according to the acceleration, the corrected position and the corrected speed includes:

[0022] Determine second position information of the UAV at each of the moments in the first time period according to the acceleration, the corrected position, and the corrected speed;

[0023] The flight distance of the UAV in the first time period is determined according to the second position information of the UAV at each moment.

[0024] In the above solution, the calculation of the collision probability between the drones according to the first position information of each drone at different times includes:

[0025] Determine the distance between the two drones according to the first position information of the two drones at the same time;

[0026] Dividing the smaller value between the distance between the two drones and the minimum safety distance of the drones by the minimum safety distance of the drones to obtain a first quotient;

[0027] The difference obtained by subtracting the first quotient from 1 is determined as the collision probability of the two drones.

[0028] In the above scheme, the method further includes:

[0029] The sum of current positioning accuracy information, the size of the UAV, and a preset safety margin of the UAV is determined as the minimum safety distance of the UAV; wherein the current positioning accuracy information is determined through Kalman filtering based on the current state data of the UAV.

[0030] In the above solution, the method further includes: storing the first position information of each drone at different times; the storing the first position information of each drone at different times includes:

[0031] A buffer zone is established to store the first position information of each drone at different times in the form of a queue.

[0032] In the above scheme, the method further includes:

[0033] The collision probability is smoothed by first-order low-pass filtering to obtain a processed collision probability.

[0034] An embodiment of the present invention provides a collision probability determination device, the device comprising: a processing module, wherein:

[0035] A processing module, used to obtain current status data of each of the multiple drones; the current status data includes current position and speed;

[0036] The processing module is further used to determine the first position information of each drone at different times according to the current state data of each drone;

[0037] The processing module is further used to calculate the collision probability between the drones according to the first position information of each drone at different times.

[0038] In the above solution, the processing module is further used to determine the flight distance of the drone in the first time period starting from the current moment according to the current state data of the drone;

[0039] Based on the current position of the drone, projecting the flight distance of the drone in the first time period onto a predetermined flight route of the drone;

[0040] Determine first position information of the UAV on the predetermined flight route at each moment in a first time period.

[0041] In the above scheme, the processing module is further used to determine the acceleration and corrected state data of the drone through Kalman filtering according to the current state data of the drone; the corrected state data includes a corrected position and a corrected speed;

[0042] The flight distance of the UAV in the first time period is determined according to the acceleration, the corrected position and the corrected speed.

[0043] In the above solution, the processing module is further used to establish a motion equation corresponding to the drone according to the motion model of the drone;

[0044] Determine a state equation of the UAV according to the motion equation; the state variables of the state equation include the position, velocity and acceleration of the UAV;

[0045] Determining a measurement equation of the UAV according to the state equation;

[0046] The Kalman filter is established according to the state equation and the measurement equation.

[0047] In the above solution, the processing module is further used to determine the second position information of the drone at each time in the first time period according to the acceleration, the corrected position and the corrected speed;

[0048] The flight distance of the UAV in the first time period is determined according to the second position information of the UAV at each moment.

[0049] In the above solution, the processing module is further used to determine the distance between the two drones according to the first position information of the two drones at the same time;

[0050] Dividing the smaller value between the distance between the two drones and the minimum safety distance of the drones by the minimum safety distance of the drones to obtain a first quotient;

[0051] The difference obtained by subtracting the first quotient from 1 is determined as the collision probability of the two drones.

[0052] In the above scheme, the processing module is also used to determine the sum of the current positioning accuracy information, the size of the UAV, and the preset safety margin of the UAV as the minimum safety distance of the UAV; wherein the current positioning accuracy information is determined by Kalman filtering based on the current state data of the UAV.

[0053] In the above solution, the device also includes a storage module; the storage module is used to establish a buffer zone to store the first position information of each drone at different times in the form of a queue.

[0054] In the above solution, the processing module is further used to smooth the collision probability through a first-order low-pass filter to obtain a processed collision probability.

[0055] An embodiment of the present invention provides a collision probability determination device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any one of the above collision probability determination methods are implemented.

[0056] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above methods for determining collision probability are implemented.

[0057] A method, device and storage medium for determining collision probability are provided in an embodiment of the present invention. The method comprises: obtaining current status data of each drone among multiple drones; the current status data comprises current position and speed; determining first position information of each drone at different times according to the current status data of each drone; calculating collision probability between the drones according to the first position information of each drone at different times; in this way, prediction of future positions of the drones can be achieved, and collision probability between the drones can be calculated based on the predicted positions. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A schematic diagram of a flow chart of a method for determining collision probability provided by an embodiment of the present invention;

[0059] Figure 2 A schematic diagram of a UAV flight route provided by an embodiment of the present invention;

[0060] Figure 3 A schematic diagram of a buffer zone provided by an embodiment of the present invention;

[0061] Figure 4 A schematic diagram of the system architecture of a collision probability determination system provided by an embodiment of the present invention;

[0062] Figure 5 A schematic diagram of a flow chart of a method for determining collision probability provided by an embodiment of the present invention;

[0063] Figure 6 A schematic diagram of the structure of a collision probability determination device provided by an embodiment of the present invention;

[0064] Figure 7 A schematic structural diagram of another device for determining collision probability provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0065] The UAV collision judgment method can use the current UAV position information to calculate the collision probability. When the distance between the two is less than a certain threshold, it is considered a collision. The UAV collision judgment method does not predict the future UAV position information, but directly uses the current UAV position information to calculate the collision probability between each UAV. It has low functionality and cannot achieve the problem of early warning before the UAV collision.

[0066] The present invention will be further described in detail below in conjunction with the embodiments.

[0067] Figure 1 A flowchart of a method for determining a collision probability provided by an embodiment of the present invention; Figure 1 As shown, the method includes:

[0068] Step 101: obtaining current status data of each of the plurality of drones; the current status data includes a current position and a speed;

[0069] In some possible implementations, obtaining the current status data of each of the plurality of drones may include: obtaining the current status data of each of the plurality of drones within a predetermined airspace. The predetermined airspace may be pre-defined.

[0070] In some possible implementations, the status data may also include the acceleration of the drone.

[0071] In some possible implementations, the drone may be monitored by sensors to collect the current status data of the drone; or a collection module may be provided inside the drone to collect the current status data output by the drone's own positioning system.

[0072] In some possible implementations, the position of the drone may include: drone coordinate position and altitude.

[0073] Specifically, a system can be composed of a drone and a ground server, the drone includes a collection module and a data encryption module, and the ground server includes a data decryption module; the drone collects the position and speed information output by the drone's own positioning system through the collection module, and encrypts the collected data through the data encryption module. The drone sends the encrypted collected data to the ground server through network communication, and the ground server obtains the encrypted collected data and decrypts the data through the data decryption module.

[0074] Step 102: determining the first position information of each drone at different times according to the current state data of each drone;

[0075] Here, different moments refer to different moments in the future, and the first position information refers to the position information of the drone at a certain moment in the future.

[0076] In some possible implementations, the first location information may be predicted based on current state data through a prediction model or the like.

[0077] In some possible implementations, the first location information may be calculated based on current state data using a prediction algorithm or the like.

[0078] In some possible implementations, the method of determining the first position information of the drone at different times based on the current state data of the drone may include a trend extrapolation prediction method, a regression prediction method, a Kalman filter prediction model, a combined prediction model, a back propagation (BP) neural network prediction model, and the like.

[0079] Step 103: Calculate the collision probability between the drones according to the first position information of each drone at different times.

[0080] Specifically, the distance between every two drones can be obtained based on the predicted first position information of each drone at the same moment in the future, a threshold is preset, and the distance between the two drones is compared with the preset threshold to obtain the collision probability between the drones.

[0081] In this way, the future position of the drone can be predicted, and the probability of collision between drones can be calculated based on the predicted position.

[0082] In some embodiments, determining the first position information of each drone at different times according to the current state data of each drone includes:

[0083] Determine, according to the current state data of the drone, a flight distance of the drone in a first time period starting from the current time;

[0084] Based on the current position of the drone, projecting the flight distance of the drone in the first time period onto a predetermined flight route of the drone;

[0085] Determine first position information of the UAV on the predetermined flight route at each moment in a first time period.

[0086] Here, the first time period is a preset time range.

[0087] In some possible implementations, the flight distance at each moment in the first time period is calculated, and the flight distance at each moment is projected onto a predetermined flight route to obtain first position information of the UAV on the predetermined flight route at each moment in the first time period.

[0088] Here, the scheduled flight route may be a flight route of one segment, or a flight route consisting of multiple flight routes.

[0089] In some possible implementations, when a flight route is composed of multiple flight routes, the flight distance of the drone in the first time period may be greater than the distance from the current position to the end point of the first flight route, or may be less than the distance from the current position to the end point of the first flight route. When the flight distance of the drone in the first time period is less than the distance from the current position to the end point of the first flight route, the flight distance is directly projected onto the first flight route; when the flight distance of the drone in the first time period is greater than the distance from the current position to the end point of the first flight route, the flight distance is subtracted from the distance from the current position to the end point of the first flight route, and the distance is compared with the distance of the second flight route, and the above process is repeated until the above steps are completed or the drone reaches the end point of the flight route.

[0090] Specifically, Figure 2 As shown, assuming that the flight route consists of AB and BC segments, point A is the starting point of the flight route, point C is the end point of the flight route, point B is the end point of the first segment of the route, and point P is the current position of the UAV. According to the current position of the UAV, the position of the UAV at the first moment is predicted, and the distance L between the position at the first moment and the current position is calculated. The distance L is projected onto the predetermined flight route to obtain the first position information of the UAV; when PB is greater than L, that is, the flight distance is less than one segment of the flight route, PB is interpolated according to L to obtain the first position information of the UAV; when PB is less than L, that is, the flight distance is greater than one segment of the flight route, the value of L-PB, that is, the distance of the lower segment of the broken line, is calculated, and L and P are updated, L=L-PB, P=B, and then PC is interpolated according to L to obtain the first position information of the UAV.

[0091] In this way, by integrating the predicted UAV position information with the scheduled flight route information, the information of the UAV flying on the route is fully utilized, which can greatly improve the accuracy of UAV position prediction.

[0092] In some embodiments, the method further comprises:

[0093] According to the motion model of the UAV, a motion equation corresponding to the UAV is established;

[0094] Determine a state equation of the UAV according to the motion equation; the state variables of the state equation include the position, velocity and acceleration of the UAV;

[0095] Determining a measurement equation of the UAV according to the state equation;

[0096] The Kalman filter is established according to the state equation and the measurement equation.

[0097] Here, the state equation is used to infer the predicted value at the next moment based on the predicted value at the previous moment; the measurement equation is used to obtain the measurement value collected by the sensor; the state variables in the state equation can be position, velocity and acceleration in only one direction, position, velocity and acceleration in two directions, or position, velocity and acceleration in three directions.

[0098] In some possible implementations, the motion model describes the changing relationship between the motion state and time. The motion model of the drone may be a constant velocity model, a constant acceleration model, a Singer model, and the like.

[0099] Specifically, it can be assumed that the acceleration obeys the classic Singer model, and the acceleration a(t) can be expressed by expression (1):

[0100]

[0101] Among them, w(t) expresses Gaussian white noise, and its mean is zero and its variance is α is the inverse of the time constant of the maneuvering target.

[0102] Establish the motion equation of the drone's position-velocity-acceleration. The drone's position in the inertial coordinate system is x. I ,y I , z I , the speed of the drone in the inertial coordinate system is v xI , v yI , v zI , the projection of the UAV acceleration in the inertial coordinate system is a xI , a yI , a zI According to the motion model, the motion equations corresponding to the axes in each direction are obtained, which can be expressed by expressions (2)-(4):

[0103]

[0104]

[0105]

[0106] Among them, λ x , y , z That is, the inverse of the maneuvering target time constant corresponding to each axis of the UAV, w xI 、w yI 、w zI That is, it expresses the system noise of the axes in each direction.

[0107] For the system, the state variables are defined, which can be expressed by expression (5):

[0108] x=[x I y I z I v xI v yI v zI a xI a yI a zI ] T (5)

[0109] The state equation can be expressed by expression (6):

[0110]

[0111] Among them, F tx is a matrix consisting of the rate of change of position on each axis over time, the rate of change of velocity over time, and the rate of change of acceleration over time, which can be expressed by expression (7):

[0112]

[0113] w is the system noise on each axis, which can be expressed by expression (8):

[0114] w=[000000 w xI w yI w zI ] (8)

[0115] Discretize the state equation with a certain period △t to obtain Φ, where Φ is the state transfer matrix and can be expressed by expression (9):

[0116] Φ=exp(F tx △t) (9)

[0117] Then the discretized state equation can be expressed by expression (10):

[0118] x(k+1)=Φx(k)+w(k) (10)

[0119] According to the position and speed of the UAV included in the measurement information, the measurement equation is obtained, which can be expressed by expression (11):

[0120] y(k)=Hx(k)+v(k) (11)

[0121] Where v(k) is the measurement noise, H is the transformation matrix from state variables to observations, which represents the linear relationship connecting the state and observations, and can be expressed by expression (12):

[0122]

[0123] In some embodiments, determining the flight distance of the drone in a first time period starting from the current time according to the current state data of the drone includes:

[0124] According to the current state data of the drone, the acceleration and the corrected state data of the drone are determined by Kalman filtering; the corrected state data includes a corrected position and a corrected speed;

[0125] The flight distance of the UAV in the first time period is determined according to the acceleration, the corrected position and the corrected speed.

[0126] Here, Kalman filtering is used to combine state data and measurement data to reduce the impact of noise in order to obtain an estimate that is closer to the actual data; the corrected position and speed are the current position and speed obtained through Kalman filtering that are closer to the actual data, which are used for subsequent more accurate calculation of the flight distance of the drone.

[0127] Specifically, the Kalman filter process can be expressed by expressions (13)-(17):

[0128] (1) Predicting the state at time k+1 based on the state at time k can be expressed by expression (13):

[0129]

[0130] in, represents the estimated value of the state variable at time k+1 predicted using the prediction at time k, Φ k+1,k represents the state transition matrix from time k to time k+1, represents the predicted estimated value at time k;

[0131] (2) Calculate the optimal state estimate at time k+1 It can be expressed by expression (14):

[0132]

[0133] Among them, Z k+1 represents the observed value at time k+1, represents the optimal state estimate at time k+1, K k+1 represents the Kalman filter gain;

[0134] (3) Calculate K k+1 , can be expressed by expression (15):

[0135]

[0136] Among them, R k+1 represents the covariance of the observation equation noise at time k+1, P k+1 / k It means estimating the covariance of the system prediction error at time k+1 based on the covariance of the system prediction error at time k;

[0137] (4) Calculate P k+1 / k , can be expressed by expression (16):

[0138]

[0139] Among them, Q k represents the covariance of the state equation noise, P k represents the covariance of the system prediction error at time k;

[0140] (5) Calculate the covariance of the system prediction error at time k+1, which can be expressed by expression (17):

[0141] P k+1 =(IK k+1 H k+1 ) k+1 / k (17)

[0142] Among them, P k+1 Represents the covariance of the system prediction error at time k+1.

[0143] When the position and velocity information of the drone is updated, a Kalman filter is constructed based on the state equation and measurement equation, and the estimated state is obtained using the Kalman filter process. That is, we get the acceleration, the corrected position and the corrected velocity.

[0144] In this way, the Kalman filter itself can eliminate the negative impact of position and velocity wild values, and can also eliminate the influence of noise. Therefore, compared with the acceleration obtained by differentiating the velocity, the acceleration estimated by the Kalman filter can obtain higher accuracy.

[0145] In some embodiments, determining the flight distance of the drone in a first time period starting from the current moment according to the acceleration, the corrected position and the corrected speed includes:

[0146] Determine second position information of the UAV at each of the moments in the first time period according to the acceleration, the corrected position, and the corrected speed;

[0147] The flight distance of the UAV in the first time period is determined according to the second position information of the UAV at each moment.

[0148] Here, the second position information is the position information at a certain moment in the future predicted based on the corrected current state data of the drone.

[0149] In some possible implementations, a time interval can be preset, and the first time period includes n time intervals; based on the acceleration, the corrected position and the corrected speed, the position of the drone after the first time interval is predicted, and the position of the drone after the first time interval is used as the current position, and the position of the drone after the second time interval is predicted, and this cycle is repeated to obtain the position of the drone at the first moment, and the flight distance of the drone is obtained based on the position of the drone at the first moment and the position of the drone at the current moment.

[0150] Exemplarily, a Kalman filter cycle can be preset as a time interval, and the first time period includes n time intervals; the position of the drone at the first moment in the future can be expressed by expression (18):

[0151]

[0152] in, For the future 0 The position of the UAV at the moment, △t is the period of Kalman filtering, and Φ is the state transfer matrix in the state equation;

[0153] After T 0 After time, the predicted flight distance of the UAV can be expressed by expression (19):

[0154]

[0155] in, is the current position of the drone.

[0156] In some embodiments, calculating the collision probability between the drones according to the first position information of each drone at different times includes:

[0157] Determine the distance between the two drones according to the first position information of the two drones at the same time;

[0158] Dividing the smaller value between the distance between the two drones and the minimum safety distance of the drones by the minimum safety distance of the drones to obtain a first quotient;

[0159] The difference obtained by subtracting the first quotient from 1 is determined as the collision probability of the two drones.

[0160] Here, the collision probability of two drones can be determined by the following expression (20):

[0161]

[0162] Among them, d ij Represents the distance between two drones, d min Indicates the minimum safe distance of the drone.

[0163] Here, the collision probability is the collision probability between two drones. A target drone can be preset, and the collision probability between the target drone and other drones can be calculated; d min There are different values ​​depending on the target drone.

[0164] Specifically, P in the expression ij , UAV i can be the target UAV, j can be other UAVs, at this time, d min is the minimum safe distance of UAV i; UAV j is the target UAV, and UAV i is another UAV. At this time, d min is the minimum safe distance of UAV j.

[0165] In some embodiments, the method further comprises:

[0166] The sum of current positioning accuracy information, the size of the UAV, and a preset safety margin of the UAV is determined as the minimum safety distance of the UAV; wherein the current positioning accuracy information is determined through Kalman filtering based on the current state data of the UAV.

[0167] Here, the minimum safety distance of the drone can be determined according to the following expression (21):

[0168] d min =h acc +l+s (21)

[0169] Among them, h acc Indicates the current positioning accuracy information, h acc It is determined by Kalman filtering based on the current state data of the UAV; l represents the size of the UAV, and s represents the preset safety margin of the UAV.

[0170] Here, d min h in acc , l and s will have different values ​​depending on the drone. At the same time, h acc It will change according to the current status data of the drone.

[0171] Specifically, h acc can be expressed as P in expression (17) k+1 The value of .

[0172] In this way, by combining the current position accuracy information of the UAV output by the Kalman filter to calculate the collision probability, the collision probability can be adaptively adjusted.

[0173] In some embodiments, the method further includes: storing the first position information of each of the drones at different times; the storing the first position information of each of the drones at different times includes:

[0174] A buffer zone is established to store the first position information of each drone at different times in the form of a queue.

[0175] In some possible implementations, after the buffer is established, the first location information within a preset time range is stored in the buffer. ij It means that among the distances between the positions of the two UAVs at each moment within the preset time range, a minimum value is selected as the distance between the two UAVs.

[0176] Here, the queue is in the form of a deletion operation at one end of the queue and an insertion operation at the other end of the queue. The end where the insertion operation is performed is called the tail of the queue, and the end where the deletion operation is performed is called the head of the queue, following the first-in-first-out principle.

[0177] Specifically, assuming that each moment is separated by a time interval dt, the length of the first time period is T0, and the location information is stored once every time interval dt in the buffer, and the total length is T0 / dt, such as Figure 3 As shown, after dt, the leftmost UAV position information is deleted, and the latest predicted UAV position information is inserted on the right.

[0178] In this way, by storing the future position information of each drone in the form of a queue in the buffer, the collision probability can be directly calculated by applying the buffer data of each drone. The calculation method is simple and the reuse rate is high.

[0179] In some embodiments, the method further comprises:

[0180] The collision probability is smoothed by first-order low-pass filtering to obtain a processed collision probability.

[0181] Here, the first-order low-pass filter uses the current sampling value and the last filtering output value to weight and obtain the effective filtering value.

[0182] Specifically, the first-order low-pass filtering formula can be expressed by expression (22):

[0183]

[0184] in, represents the collision probability between UAV i and UAV j at the kth moment, Pij represents the collision probability between UAV i and UAV j at the current moment, and a is the first-order low-pass filter coefficient.

[0185] In this way, the collision probability can be smoothed by a first-order low-pass filter to prevent the collision probability from being affected by a single outlier.

[0186] The following provides multiple specific examples in combination with any of the above embodiments:

[0187] This embodiment designs a system for predicting the mutual collision of multiple drones in the air, including: a drone and a ground server, a collection module is provided in the drone, the collection module is connected to a data encryption module, a decryption module is provided in the ground server, the decryption module is connected to a collision probability estimation module, and the collection module is communicatively connected to the collision probability estimation module. This embodiment also proposes an estimation method for predicting the mutual collision probability of multiple drones in the air, and the calculation method is as follows: the acceleration of the drone is estimated by establishing a Kalman filter equation, and then the displacement of the drone is predicted, and then the predicted displacement is integrated with the route information to predict the position of the drone at a future time, and then the collision prediction is performed based on the position of the drone at a future time. When calculating the collision probability, this embodiment takes into account the predicted positioning accuracy of the Kalman filter output, so that the collision prediction is more accurate and the safety of the drone is guaranteed. At the same time, a method for correcting the collision probability by a first-order low-pass filter is also provided to eliminate the influence of wild values.

[0188] 1) System

[0189] like Figure 4 As shown, each UAV is provided with a collection module to collect the position and speed information output by the UAV's own positioning system; the collection module is connected to a data encryption module to encrypt the data; each UAV communicates data with the ground server through the network; the ground server is provided with a decryption module, and the decryption module is connected to a collision probability estimation module, which estimates the collision probability between each UAV, and outputs a collision warning signal if it is higher than a certain threshold.

[0190] 2) Method flow

[0191] Step 1: The ground server obtains the data of each drone, including the current position and speed of each drone, as well as the flight route of each drone.

[0192] Step 2: Predict the location information of each drone in the future time period T0. For each drone, follow the steps 2.1-2.4 below to predict the location information at the future time.

[0193] 2.1 Establishment of Kalman filter equation

[0194] The motion model selection is very helpful for the estimation of motion information. It describes the relationship between the motion state and time. Here we assume that the acceleration obeys the classic Singer model. Therefore, its acceleration a(t) can be expressed by expression (1).

[0195] Here, the motion equation of the position-velocity-acceleration of the drone is established. The position of the drone in the inertial coordinate system is x I ,y I , z I , the speed of the drone is v xI , v yI , v zI , the projection of the UAV acceleration in the inertial coordinate system is a xI 、a yI 、a zI , we can establish the following three sets of motion equations, which can be expressed by expressions (2)-(4).

[0196] For the system, the state variables are defined and can be expressed by expression (5).

[0197] Then the state equation can be obtained, which can be expressed by expressions (6)-(8).

[0198] After discretizing the above equation with a certain period △t, we get Φ, which can be expressed by expression (9).

[0199] Then the discretized state equation can be expressed by expression (10).

[0200] Its measurement information contains position and velocity, so its measurement equation can be expressed by expressions (11) and (12).

[0201] 2.2 Accelerometer estimation

[0202] First, the Kalman filter process can be expressed by expressions (13)-(17).

[0203] Then, according to the system equation and measurement equation in Section 2.1, a Kalman filter is constructed, and then the acceleration information of the UAV is estimated in real time using the Kalman filter. That is, when the position and velocity information of the UAV is updated, the estimated state is obtained using the above Kalman filter process.

[0204] The acceleration information estimated by Kalman filtering is more accurate than the acceleration obtained by differentiating the velocity, because the Kalman filter itself can eliminate the negative impact of position and velocity wild values, and can also eliminate the influence of noise.

[0205] 2.3 Displacement prediction

[0206] According to step 2, the corrected position, velocity and acceleration information at the current moment are obtained, so the future T 0 The position at the time can be calculated by expression (18), so after T 0 After time, the predicted travel distance of the UAV is which can be calculated by expression (19).

[0207] 2.4 Integrated route information

[0208] Figure 2 The middle route is composed of two segments AB and BC. P is the current position of the drone. The future T is obtained from step 2.3. 0 The flight distance of the drone is L, and the drone flies along the route, so the drone position should be on the route, so the distance L is projected to obtain the predicted final position of the drone. The projection method is as follows:

[0209] Step 1): Calculate the PB distance. If PB is greater than L, interpolate PB according to L and end;

[0210] Step 2): If PB is less than L, calculate the distance of the next broken line and update L and P, L = L-PB, P = B, and then repeat step 1) until the above steps are completed or the end of the drone route is reached.

[0211] By integrating it with the route information, the prior information of the UAV flying on the route is fully utilized, which can greatly improve the accuracy of UAV position prediction.

[0212] Step 3: Adaptively adjust the drone collision probability.

[0213] Each drone establishes a corresponding buffer, which stores the future T 0 The predicted location information of the time period is in the form of a queue, following the first-in-first-out principle. The buffer has a time interval of dt and a total length of T. 0 / dt.

[0214] According to the predicted positioning accuracy information h output by the Kalman filter in step 2.1 acc , UAV size l and safety margin s, the minimum safety distance between UAVs can be expressed by expression (21). Then, the relative distances between buffer zones corresponding to different UAVs are calculated respectively, and the minimum distance d between the buffer zones is calculated. ij =min(d_ij), then the collision probability between different UAVs can be calculated by expression (20).

[0215] from Figure 3It can be seen that the drone i queue stores the drone position information. After dt, the leftmost drone position information is deleted, and the earliest predicted position information is inserted on the right, which follows the first-in-first-out principle of the queue.

[0216] Step 4: Filter the collision probability.

[0217] Considering the wild values ​​in the drone estimation data and the collected data, the above probability is corrected after each buffer update, which can be calculated by expression (22).

[0218] Above represents the collision probability between drone i and drone j at the kth moment. a is the first-order low-pass filter coefficient. The collision probability is smoothed by the first-order low-pass filter to avoid being affected by a single outlier.

[0219] Step 5: Warning when the collision probability exceeds the threshold

[0220] When the collision probability obtained above exceeds a certain threshold, it is considered that a collision has occurred and a collision warning signal is issued.

[0221] Figure 5 FIG. 1 is a flow chart of a method for determining a collision probability provided by an embodiment of the present invention. Figure 5 As shown, the process includes the following steps:

[0222] Step 501: The ground server obtains data of each drone;

[0223] Step 502: predict the location information of each drone in the future T0 time period;

[0224] Step 503: adaptively adjust the collision probability of the drones;

[0225] Step 504: filter the collision probability;

[0226] Step 505, determine whether it exceeds the preset threshold; if it exceeds the threshold, execute step 506; if it does not exceed the threshold, execute step 501;

[0227] Step 506: Issue a warning signal.

[0228] Figure 6 A schematic diagram of a collision probability determination device provided by an embodiment of the present invention; Figure 6 As shown, the device comprises: a processing module, wherein:

[0229] The processing module is used to obtain current status data of each of the multiple drones; the current status data includes current position and speed;

[0230] The processing module is further used to determine the first position information of each drone at different times according to the current state data of each drone;

[0231] The processing module is further used to calculate the collision probability between the drones according to the first position information of each drone at different times.

[0232] Specifically, the processing module is further used to determine the flight distance of the drone in a first time period starting from the current moment according to the current state data of the drone;

[0233] Based on the current position of the drone, projecting the flight distance of the drone in the first time period onto a predetermined flight route of the drone;

[0234] Determine first position information of the UAV on the predetermined flight route at each moment in a first time period.

[0235] Specifically, the processing module is further used to determine the acceleration and corrected state data of the drone through Kalman filtering according to the current state data of the drone; the corrected state data includes a corrected position and a corrected speed;

[0236] The flight distance of the UAV in the first time period is determined according to the acceleration, the corrected position and the corrected speed.

[0237] Specifically, the processing module is further used to establish a motion equation corresponding to the drone according to the motion model of the drone;

[0238] Determine a state equation of the UAV according to the motion equation; the state variables of the state equation include the position, velocity and acceleration of the UAV;

[0239] Determining a measurement equation of the UAV according to the state equation;

[0240] The Kalman filter is established according to the state equation and the measurement equation.

[0241] Specifically, the processing module is further used to determine the second position information of the drone at each moment in the first time period according to the acceleration, the corrected position and the corrected speed;

[0242] The flight distance of the UAV in the first time period is determined according to the second position information of the UAV at each moment.

[0243] Specifically, the processing module is further used to determine the distance between the two drones according to the first position information of the two drones at the same time;

[0244] Dividing the smaller value between the distance between the two drones and the minimum safety distance of the drones by the minimum safety distance of the drones to obtain a first quotient;

[0245] The difference obtained by subtracting the first quotient from 1 is determined as the collision probability of the two drones.

[0246] Specifically, the processing module is also used to determine the sum of the current positioning accuracy information, the size of the UAV, and the preset safety margin of the UAV as the minimum safety distance of the UAV; wherein the current positioning accuracy information is determined by Kalman filtering based on the current state data of the UAV.

[0247] Specifically, the device also includes a storage module; the storage module is used to establish a buffer zone to store the first position information of each drone at different times in a queue form.

[0248] Specifically, the processing module is further used to perform smoothing processing on the collision probability through a first-order low-pass filter to obtain a processed collision probability.

[0249] It should be noted that: the collision probability determination device provided in the above embodiment only uses the division of the above program modules as an example when implementing the corresponding collision probability determination method. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the processing device is divided into different program modules to complete all or part of the processing described above. In addition, the device provided in the above embodiment and the embodiment of the corresponding method belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0250] Figure 7 A schematic diagram of the structure of another device for determining collision probability provided by an embodiment of the present invention; Figure 7 As shown, the device 70 includes: a processor 701 and a memory 702 for storing a computer program that can be run on the processor; when the processor 701 is used to run the computer program, it executes: obtaining the current state data of each drone among multiple drones; the current state data includes the current position and speed; according to the current state data of each drone, determining the first position information of each drone at different times; according to the first position information of each drone at different times, calculating the collision probability between the drones. Wherein, when the processor runs the computer program, the corresponding processes implemented by the processing device in each method of the embodiment of the present invention are implemented, and for the sake of brevity, they are not repeated here.

[0251] In practical applications, the device 70 may further include: at least one network interface 703. The various components in the collision probability determination device 70 are coupled together via a bus system 704. It is understood that the bus system 704 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 704 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, Figure 7 In the figure, various buses are marked as bus system 704. There may be at least one processor 701. The network interface 703 is used for wired or wireless communication between the collision probability determination device 70 and other devices.

[0252] The memory 702 in the embodiment of the present invention is used to store various types of data to support the operation of the collision probability determination device 70 .

[0253] The method disclosed in the above embodiment of the present invention can be applied to the processor 701, or implemented by the processor 701. The processor 701 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 701 or the instruction in the form of software. The above processor 701 may be a general processor, a digital signal processor (DSP, DiGital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 701 can implement or execute the methods, steps and logic block diagrams disclosed in the embodiment of the present invention. The general processor may be a microprocessor or any conventional processor, etc. In combination with the steps of the method disclosed in the embodiment of the present invention, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in the memory 702. The processor 701 reads the information in the memory 702 and completes the steps of the above method in combination with its hardware.

[0254] In an exemplary embodiment, the collision probability determination device 70 can be implemented by one or more application-specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD), field-programmable gate array (FPGA), general-purpose processor, controller, microcontroller (MCU), microprocessor, or other electronic components to execute the aforementioned method.

[0255] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon; when the computer program is executed by a processor, the computer program executes: obtaining current status data of each of a plurality of drones; the current status data including a current position and a speed; determining first position information of each of the drones at different times based on the current status data of each of the drones; and calculating the collision probability between the drones based on the first position information of each of the drones at different times.

[0256] Among them, when the computer program is executed by the processor, the corresponding processes implemented by the processing device in each method of the embodiment of the present invention are implemented, and for the sake of brevity, they are not repeated here.

[0257] In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0258] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0259] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0260] A person skilled in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, etc. Various media that can store program codes.

[0261] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0262] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for determining collision probability, characterized in that: The method comprises: Acquire current status data of each of the plurality of drones; the current status data includes current position and speed; Determine the first position information of each drone at different times according to the current state data of each drone; The collision probability between the drones is calculated according to the first position information of each drone at different times.

2. The method according to claim 1, characterized in that Determining the first position information of each drone at different times according to the current state data of each drone includes: Determine, according to the current state data of the drone, a flight distance of the drone in a first time period starting from the current time; Based on the current position of the drone, projecting the flight distance of the drone in the first time period onto a predetermined flight route of the drone; Determine first position information of the UAV on the predetermined flight route at each moment in a first time period.

3. The method according to claim 2, characterized in that The step of determining the flight distance of the drone in a first time period starting from the current time according to the current state data of the drone comprises: According to the current state data of the drone, the acceleration and the corrected state data of the drone are determined by Kalman filtering; the corrected state data includes a corrected position and a corrected speed; The flight distance of the UAV in the first time period is determined according to the acceleration, the corrected position and the corrected speed.

4. The method according to claim 3, characterized in that The method further comprises: According to the motion model of the UAV, a motion equation corresponding to the UAV is established; Determine a state equation of the UAV according to the motion equation; the state variables of the state equation include the position, velocity and acceleration of the UAV; Determining a measurement equation of the UAV according to the state equation; The Kalman filter is established according to the state equation and the measurement equation.

5. The method according to claim 3, characterized in that: Determining the flight distance of the UAV in a first time period starting from the current moment according to the acceleration, the corrected position and the corrected speed includes: Determine second position information of the UAV at each of the moments in the first time period according to the acceleration, the corrected position, and the corrected speed; The flight distance of the UAV in the first time period is determined according to the second position information of the UAV at each moment.

6. The method according to claim 1, characterized in that The calculating the collision probability between the drones according to the first position information of each drone at different times includes: Determine the distance between the two drones according to the first position information of the two drones at the same time; Dividing the smaller value between the distance between the two drones and the minimum safety distance of the drones by the minimum safety distance of the drones to obtain a first quotient; The difference obtained by subtracting the first quotient from 1 is determined as the collision probability of the two drones.

7. The method according to claim 6, characterized in that The method further comprises: The sum of current positioning accuracy information, the size of the UAV, and a preset safety margin of the UAV is determined as the minimum safety distance of the UAV; wherein the current positioning accuracy information is determined through Kalman filtering based on the current state data of the UAV.

8. The method according to any one of claims 1 to 7, characterized in that: The method further includes: storing the first position information of each drone at different times; the storing the first position information of each drone at different times includes: A buffer zone is established to store the first position information of each drone at different times in the form of a queue.

9. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: The collision probability is smoothed by first-order low-pass filtering to obtain a processed collision probability.

10. A device for determining collision probability, characterized in that: The device comprises: A processing module, used to obtain current status data of each of the multiple drones; the current status data includes current position and speed; The processing module is further used to determine the first position information of each drone at different times according to the current state data of each drone; The processing module is further used to calculate the collision probability between the drones according to the first position information of each drone at different times.

11. A collision probability determination device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 10 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images