A method for determining the search direction of unmanned aerial vehicle based on time series

Abstract

The present invention discloses a method for determining the search heading of an unmanned aerial vehicle (UAV) based on a time series, relating to the field of information perception and recognition technology. The method first acquires ship navigation data in real time, recording sequences comparing different times and positions. Based on a time series analysis model, the method predicts the possible location and heading of a vessel after the vessel's AIS track disappears for a period of time. Then, based on the search encounter point, the method calculates the UAV heading determination method for searching the possible area of ​​the vessel at different speeds. In the event that the current vessel's position information cannot be obtained, a UAV needs to be dispatched to search and rescue the vessel. By calculating the UAV's search and rescue heading after a period of interruption based on a time series method, the method is easy to implement and apply in engineering.

Description

Technical Field

[0001] The present invention relates to information perception and recognition technology, and in particular to a method for determining the search heading of an unmanned aerial vehicle based on time series. Background Art

[0002] In real-world scenarios, ships at sea may encounter unusual circumstances, such as their track disappearing from surveillance cameras or losing contact after receiving a rescue call. This necessitates the deployment of drones for search and rescue operations. To quickly locate the vessel, the drone needs to determine the direction in which to search. Previously, searches typically began based on the location where the vessel disappeared. This resulted in a large search area and uncertain direction, resulting in lengthy searches and difficulty finding the target in the vast ocean. Alternatively, manual decisions about search locations were made, which is subjective and lacks a reliable automated solution or direct computational support. Summary of the Invention

[0003] In light of this, this paper proposes a time-series-based method for determining the search course for drones. This method uses a vessel's current position, predicted after a certain period of track interruption, as a starting point for drone search. At a certain speed, the method calculates the required course to encounter the vessel. Based on a time-series approach, this method is easy to implement and apply in engineering applications.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] A method for determining a UAV search heading based on a time series comprises the following steps:

[0006] Step 1: Obtain ship navigation data in real time and record the comparison sequence of different time and position;

[0007] Step 2: Disappear at the track point After a certain time, the predicted position, course and speed of the vessel are calculated;

[0008] Step 3: Calculate the orientation of the ship's predicted position relative to the drone's initial position ;according to The value range of the ship is used to calculate the side angle of the predicted position of the ship relative to the initial position of the drone. ; Further calculate the heading that the drone should take when searching for a heading , complete the UAV search heading determination based on time series.

[0009] Furthermore, the sequences at different moments in step 1 are as follows:

[0010] ;

[0011] The position sequences at different times are as follows:

[0012] .

[0013] Furthermore, the specific method of step 2 is:

[0014] Disappeared at the track point After a certain time, the last multiple track points and ship headings before disappearance are obtained; using Python and statsmodels library, the ARIMA model based on time series analysis is used to predict the ship's position at time Predicted location at the moment , while combining Point location information and disappearance time Estimate the current heading , speed ; is the corresponding longitude and latitude.

[0015] Furthermore, the specific method of step 3 is:

[0016] Step 301: Calculate the vessel's Predicted location at the moment Relative to the initial position of the drone Azimuth , For the corresponding longitude and latitude:

[0017] (1) The initial position of the UAV and the ship Convert the latitude and longitude coordinates of the predicted position at the moment to radians as well as ;

[0018] (2) Calculate the longitude difference ;

[0019] (3) Based on Python, calculate the corresponding and Quantity;

[0020] ;

[0021] ;

[0022] (4) Calculate the azimuth , the range is ;

[0023] Step 302: record the speed of the drone as , calculate the course the drone should take when searching for a ship :

[0024] (1) :

[0025] ;

[0026] ;

[0027] (2) :

[0028] ;

[0029] ;

[0030] (3) :

[0031] ;

[0032] ;

[0033] (4) :

[0034] ;

[0035] ;

[0036] in, For ships in The heading angle of the predicted position at the moment relative to the initial position of the UAV.

[0037] Due to the adoption of the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0038] 1. This invention provides a time-series-based method for determining the heading of a drone search. After a vessel loses contact at sea, this method takes into account the duration of the loss of contact and the controllable speed of the drone to determine its heading. This method provides a more accurate drone search method and improves search efficiency.

[0039] 2. The present invention also considers that after the ship loses contact, it supports the use of the ocean environment to influence the speed and direction of the ship. , speed The heading and speed of the ocean current are dynamically adjusted, and the heading of the drone is dynamically determined according to steps 202 and 3. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 The figure is an overall flow chart of a method for determining a UAV search heading based on a time series in an embodiment of the present invention.

[0041] Figure 2Schematic diagram of the search heading determination process in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0043] A method for determining the search direction of a UAV based on time series, such as Figure 1 As shown, the following steps are included:

[0044] Step 1: Obtain ship navigation data in real time and obtain real-time target ships Location information at all times , ; Record the comparison sequence of different time and position, such as Figure 2 The first three points of the ship represent the construction of dynamic navigation data, forming different time series as follows:

[0045] ;

[0046] The position sequences at different times are as follows:

[0047] .

[0048] go through After time, the ship's location information cannot be obtained, and the drone is sent to search for the target. The initial position of the drone is .in , The corresponding longitude and latitude respectively.

[0049] Assume that the ship's time sequence is as follows:

[0050] ;

[0051] The position sequences at different times are as follows:

[0052] ;

[0053] Step 2: Disappear at the track point Get the last multiple track point positions after time and the ship's heading. Calculated using the time series ARIMA model The possible position of the ship after time and heading ;

[0054] a) Using Python and the statsmodels library, predict the next moment based on the ARIMA model of time series analysis Possible target location ,Right now Figure 2The dashed arrows connecting the ships and the predicted positions are shown, and the current heading is estimated together with the position information of the previous point. , speed ;

[0055] Assume that the vessel is discovered 30 minutes after it disappears and at 1:40 p.m., and disposal begins.

[0056] At this time, the longitude and latitude of the ship's position calculated using the time series is approximately (109.10, 17.20);

[0057] The heading was approximately -0.4636 degrees west of north, and the speed was approximately 18 knots.

[0058] b) If Figure 2 The line between the initial position of the UAV and the predicted position of the ship is shown in the figure. Time distance , calculated using the Vincenty formula based on the latitude and longitude information of the two points .

[0059] Assume that the initial position of the drone is is (105.0, 19.0), then the distance between the drone and the target ship is Kilometers. The calculation process uses the Vincenty formula in Python. From geopy.distanceimport geodesic

[0060] .

[0061] .

[0062] .

[0063] Step 3: Use Vincentey’s formula to calculate the predicted position of the drone and the ship Distance, calculate the position of the ship relative to the drone ;according to The value range of the ship is used to calculate the ship's predicted position relative to the UAV's side angle. ; Assume that the speed of the drone is ,according to The value range of 、 、 , calculate the heading the drone should take ,Right now Figure 2 The UAV is traveling along the dashed arrow.

[0064] a) Calculate the target Azimuth relative to the drone at any moment :

[0065] (1) Convert the longitude and latitude coordinates of the predicted positions of the drone and ship into radians as well as ;

[0066] (2) Calculate the longitude difference ;

[0067] (3) Based on Python, calculate the corresponding and Quantity;

[0068] ;

[0069] ;

[0070] (4) Calculate the azimuth , the range is .

[0071] Based on the above example, we can calculate approximately radian.

[0072] b) If Figure 2 As shown, it is assumed that the speed of the UAV is , calculate the heading the drone should take Perform target search, where is the sideways angle of the target vessel relative to the UAV.

[0073] (1) :

[0074] ;

[0075] ;

[0076] At this time, the example is based on the above calculation formula, and the calculation results are approximately .

[0077] Assuming the drone's speed is 350 km / h,

[0078] Then we can calculate approximately radian.

[0079] (2) :

[0080] ;

[0081] ;

[0082] (3) :

[0083] ;

[0084] ;

[0085] (4) :

[0086] ;

[0087] .

[0088] In summary, this invention predicts the current location of a ship based on the time and duration of the ship's track disappearance. Based on this starting point, the drone and the ship's encounter point is calculated within a certain search range, and the drone's heading is finally determined. Through continuous accumulation and improvement, various random factors and natural environmental factors are integrated to reduce the search range and time.

[0089] Those skilled in the art will appreciate that the embodiments described are intended to help readers understand the principles of the present invention and should be understood that the scope of protection of the present invention is not limited to the embodiments described. It will be apparent to those skilled in the art that various modifications and variations are possible in the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of the present invention.

Claims

1. A method for determining the search direction of a UAV based on time series, characterized in that: The following steps are involved: Step 1: Obtain ship navigation data in real time and record the comparison sequence of different times and positions. Specifically: The different time sequences are as follows: ； The position sequences at different times are as follows: ； Step 2: Disappear at the track point After a certain time, the predicted position, course and speed of the ship are calculated. Specifically: Disappeared at the track point After a certain time, the last multiple track points and ship headings before disappearance are obtained; using Python and statsmodels library, the ARIMA model based on time series analysis is used to predict the ship's position at time Predicted location at the moment , while combining Point location information and disappearance time Estimate the current heading , speed ; is the corresponding longitude and latitude; Step 3: Calculate the orientation of the ship's predicted position relative to the drone's initial position ;according to The value range of the ship is used to calculate the side angle of the predicted position of the ship relative to the initial position of the drone. ; Further calculate the heading that the drone should take when searching for a heading , complete the UAV search heading determination based on time series.

2. The method for determining the search direction of a UAV based on a time series according to claim 1, characterized in that: The specific method of step 3 is: Step 301: Calculate the vessel's Predicted location at the moment Relative to the initial position of the drone Azimuth , For the corresponding longitude and latitude: (1) The initial position of the UAV and the ship Convert the latitude and longitude coordinates of the predicted position at the moment to radians as well as ; (2) Calculate the longitude difference ; (3) Based on Python, calculate the corresponding and Quantity; ； ； (4) Calculate the azimuth , the range is ; Step 302: record the speed of the drone as , calculate the course the drone should take when searching for a ship : （1） ： ； ； （2） ： ； ； （3） ： ； ； （4） ： ； ； in, For ships in The heading angle of the predicted position at the moment relative to the initial position of the UAV.

Images