Local target point selection method and device for rolling route planning of unmanned aerial vehicle

Abstract

The invention provides a local target point selection method and device for rolling route planning of an unmanned aerial vehicle. The method comprises the following steps: 1, establishing and initializing an environment model; 2, determining a global planning starting point, a global planning final target point and a pre-planned optimal route; 3, determining a local planning starting point; 4, updating the environment threat situation; 5, if yes, an optimal route target point selection strategy is adopted, local target points are selected on the pre-planned optimal route, and if not, a heuristic target point selection strategy is adopted, and the local target points are generated through a heuristic function; 6, if the generated local target points are feasible target points, adopting the generated local target points; otherwise, adopting a protection strategy to generate a new feasible local target point; 7, performing route planning according to the local planning starting point and the generated local target point, and flying for a long step distance according to the route to reach a local planning step length arrival point; and 8, judging whether a final planning target point is reached or not, if so, ending the planning, otherwise, returning to the step 3.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle (UAV) route planning, and in particular relates to a method and a device for selecting local target points for rolling route planning of an unmanned aerial vehicle (UAV). Background Art

[0002] As one of the key technologies in the field of UAVs, UAV path planning mainly includes two parts: offline planning and online planning.

[0003] Offline planning is mainly based on the predetermined tasks and known environmental information, and the global route planning is carried out by comprehensively considering the performance constraints of the UAV. Online planning is carried out in real time during the flight of the UAV based on real-time sensor information, and the route is locally replanned according to the continuously incoming obstacle information, so as to quickly generate a flyable route and make corresponding reasonable obstacle avoidance behaviors.

[0004] To realize the rolling route planning of UAV in complex environment, the selection of local target points is particularly important, which is not only related to the optimality of local route planning, but also has a certain impact on the global optimality.

[0005] However, existing UAV online route planning technologies often have the problem of difficulty in balancing global optimal indicators and local optimal indicators. Summary of the invention

[0006] The present invention provides a method and device for selecting local target points for rolling route planning of unmanned aerial vehicles, which can solve the problem that it is difficult to take into account both global optimal indicators and local optimal indicators in existing unmanned aerial vehicle online route planning technologies; while ensuring real-time online local planning, global optimization indicators are considered, and the optimality of the entire route is not lost while flexibly avoiding obstacles.

[0007] In a first aspect, the present application provides a method for selecting local target points for rolling route planning of an unmanned aerial vehicle, the method comprising:

[0008] Step 1: Establish and initialize an environment model, wherein the parameter information of the environment model includes: threat area, obstacle area and no-fly zone;

[0009] Step 2: Determine the global planning starting point, the global planning final target point and the pre-planned optimal route;

[0010] Step 3: Determine the local planning starting point based on the global planning starting point;

[0011] Step 4: Update the environmental threat situation based on the environmental model, the local planning starting point, and the situation measurement values ​​within the single-step planning window period;

[0012] Step 5: Determine whether there is a pre-planned optimal route. If so, use the optimal route target point selection strategy to select a local target point on the pre-planned optimal route. If not, use the heuristic target point selection strategy to generate a local target point by a heuristic function.

[0013] Step 6: According to the preset judgment principle, determine whether the generated local target point is a feasible target point. If the generated local target point is a feasible target point, the generated local target point is adopted; if the generated local target point is not a feasible target point, a protection strategy is adopted to generate a new feasible local target point;

[0014] Step 7: Perform route planning based on the local planning starting point and the generated local target point, and use the point reached by flying only one step distance along the route as the local planning step arrival point;

[0015] Step 8: Determine whether the planned final target point has been reached. If so, the planning ends; otherwise, return to step 3.

[0016] Furthermore, step 5 includes:

[0017] 501, establish a local rectangular coordinate system with the local planning starting point as the center of the circle, the vector from the starting point to the final target point of the planning as the x-axis, and the direction perpendicular to the vector as the y-axis;

[0018] 502, in the local rectangular coordinate system, A total of k1 points are randomly generated within the range, and the randomly generated k1 points are arranged in descending order according to their corresponding x-coordinates;

[0019] 503. According to the arrangement order of step 502, local target points are searched in sequence according to the heuristic function f(P)=g(P)+h(P).

[0020] Furthermore, the heuristic target point selection strategy is: after each rolling window information update, the heuristic function f(P)=g(P)+h(P) is used to select the local target point, where P is the point on the rolling window, g(P) is the cost of the drone flying from the current position to the target position, and h(P) is the cost of flying from P to the end point.

[0021] Furthermore, in step 6, a protection strategy is adopted to generate a new feasible local target point, and the steps are as follows:

[0022] 601, if the local target point generated according to the optimal route target point selection strategy is located in the threat situation detected by the rolling window, then the heuristic target point selection strategy is used to generate the local target point;

[0023] 602, if the k1 points generated by the heuristic target point selection strategy are all infeasible points, then Generate k2 points randomly within the range and arrange them in descending order according to the corresponding x coordinates, and search for local target points in turn according to the heuristic function f(P) = g(P) + h(P);

[0024] 603. If the k2 points generated in step 602 are also infeasible points, the radius of the rolling window is reduced by 10%, and the process returns to step 601 until a feasible local target point is found.

[0025] Furthermore, the judgment principle in step 6 is: the generated local target point is in the feasible domain.

[0026] Furthermore, the judgment condition for reaching the final target point of the plan in step 8 is: during planning, the final target point of the global plan appears in a circle with the starting point of the next round of local planning as the center and the rolling window distance as the radius.

[0027] Furthermore, the starting point of the first planning in step 3 is the global planning starting point (x1, y1) of the aircraft, and the starting point of each subsequent round of planning is the step arrival point of the previous round of planning.

[0028] In a second aspect, the present application provides a local target point selection device for UAV rolling route planning, and the device is used to implement the above-mentioned local target point selection method for UAV rolling route planning.

[0029] Advantages of the invention: The invention proposes a method and device for selecting local target points for rolling route planning of unmanned aerial vehicles. The method and device refer to the principle of A* algorithm for local environmental information measured by the unmanned aerial vehicle in real time, make full use of the pre-planned optimal route, take into account the rapidity and optimality indicators, and design corresponding protection strategies to ensure the feasibility of generating local target points during rolling online planning of unmanned aerial vehicles. The technical effects are as follows:

[0030] (1) The method is simple in principle, easy to implement, and requires little computation, and can better meet the requirements of rapidity and real-time performance of online route planning under airborne conditions;

[0031] (2) This method can make full use of the local environmental information detected in real time, effectively combine the optimization and feedback mechanisms, and achieve the reasonable integration of global environmental information and local information detected in real time;

[0032] (3) This method not only ensures real-time online local planning, but also takes into account the global optimization index. It can flexibly avoid obstacles while always approaching the final target point of the plan, ensuring the correctness of the planning direction while maintaining the optimality of the entire route.

[0033] (4) This method is highly flexible. It can flexibly select the target point generation strategy and flexibly set the parameters according to the focus of the problem to be solved.

[0034] (5) This method is highly versatile and is applicable to both fixed-wing and rotary-wing UAVs. For different models of UAVs, one only needs to change the parameters related to the UAV performance in the algorithm to use it directly. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0036] Figure 1 Execute a flow chart for the local target point selection method;

[0037] Figure 2 Execute flow chart for protection strategy;

[0038] Figure 3 Schematic diagram of the strategy for selecting the optimal route target point;

[0039] Figure 4 Schematic diagram of the heuristic target point selection strategy. DETAILED DESCRIPTION

[0040] The invention belongs to the technical field of unmanned aerial vehicle route planning, and specifically relates to a local target point selection method for rolling route planning of unmanned aerial vehicle. Firstly, the starting point of rolling online planning is determined to update the environmental threat situation within a single-step planning window period; secondly, it is judged whether there is a pre-planned optimal route, if yes, an optimal route target point selection strategy is adopted, if no, a heuristic target point selection strategy is adopted; then, it is judged whether the generated local target point is a feasible target point, if yes, the generated local target point is adopted, if not, a protection strategy is adopted to generate a new feasible local target point; then, route planning is performed with the starting point and the generated local target point, and a step length is flown according to the route; finally, it is judged whether the planned final target point is reached, if yes, the planning is ended, otherwise, the above-mentioned local route online planning steps are executed rollingly.

[0041] The present invention provides a method and device for selecting local target points for rolling route planning of unmanned aerial vehicles. The method and device use the heuristic function principle of the A* algorithm for reference, based on the local environmental information measured by the unmanned aerial vehicle in real time, make full use of the pre-planned optimal route, take into account the rapidity and optimality indicators, and design corresponding protection strategies to ensure the feasibility of generating local target points during the rolling online planning of the unmanned aerial vehicle, so as to achieve rapid adjustment of the dynamic environment.

[0042] The technical solution of the present invention is a method for selecting local target points for rolling route planning of unmanned aerial vehicles. In the whole rolling update process, the heuristic algorithm is combined with the pre-planned optimal route. According to the local perceived situation information, the heuristic strategy is used to update the search for local target points or select the pre-planned optimal route points, and the flight process is continuously advanced, which effectively improves the applicability and optimality of the local window route search. From the initial stage to the final stage of the planning, the unmanned aerial vehicle continuously approaches the final target point of the planning, and the directionality of the planning is also strengthened. When the planning window scrolls to the end, the local target point is also the final target point of the planning.

[0043] like Figure 1 As shown, the present application provides a method for selecting local target points for UAV rolling route planning, the method comprising:

[0044] Step 1: Establish and initialize an environment model, wherein the parameter information of the environment model includes: threat area, obstacle area and no-fly zone;

[0045] Step 2: Determine the global planning starting point, the global planning final target point and the pre-planned optimal route;

[0046] Step 3: Determine the local planning starting point based on the global planning starting point;

[0047] Among them, the starting point of the first planning in step 3 is the global planning starting point of the aircraft (x1, y1), and the starting point of each subsequent round of planning is the step length arrival point of the previous round of planning.

[0048] Step 4: Update the environmental threat situation based on the environmental model, the local planning starting point, and the situation measurement values ​​within the single-step planning window period;

[0049] Step 5: Determine whether there is a pre-planned optimal route. If so, use the optimal route target point selection strategy to select a local target point on the pre-planned optimal route. If not, use the heuristic target point selection strategy to generate a local target point by a heuristic function.

[0050] Optionally, the premise for using the optimal route target point selection strategy in step 5 is that there is a pre-planned optimal route. At this time, the local target point selection strategy is: taking the local planning starting point as the center of the circle and the rolling window distance as the radius, the intersection of the circle and the pre-planned optimal route.

[0051] Furthermore, the specific solution method of the local target point selection strategy can be expressed as:

[0052] 501, establish a local rectangular coordinate system with the local planning starting point as the center of the circle, the vector from the starting point to the final target point of the planning as the x-axis, and the direction perpendicular to the vector as the y-axis;

[0053] 502, in the local rectangular coordinate system, A total of k1 points are randomly generated within the range, and the randomly generated k1 points are arranged in descending order according to their corresponding x-coordinates;

[0054] 503. According to the arrangement order of step 502, local target points are searched in sequence according to the heuristic function f(P)=g(P)+h(P).

[0055] Optionally, the heuristic target point selection strategy in step 5 is: after each rolling window information update, the heuristic function f(P)=g(P)+h(P) is used to select the local target point, where P is a point on the rolling window, g(P) is the cost of the drone flying from the current position to the target position, and h(P) is the cost of flying from P to the end point.

[0056] It should be noted that g(P) can be estimated based on the position of P and the environmental information in the current window.

[0057] It should be noted that since the information outside the boundary cannot be obtained, h(P) is estimated using the Euclidean distance from P to the end point, and finally the window boundary point P with the minimum cost is taken as the local target point P sub , when the obtained P sub If there is more than one, you can choose any one, namely:

[0058] minf(P)=g(P)+h(P)

[0059] In order to simplify the calculation, the method reduces the optimization requirements and focuses on finding a feasible algorithm. Therefore, the value of g(P) is only determined by whether P belongs to the feasible domain, that is:

[0060]

[0061] The definition of the feasible domain is: the generated local target point is not located in the threat area, obstacle area and no-fly zone.

[0062] It should be noted that the heuristic target point selection strategy is a compromise between the requirements of global optimization and the constraints of local limited information.

[0063] Step 6: According to the preset judgment principle, determine whether the generated local target point is a feasible target point. If the generated local target point is a feasible target point, the generated local target point is adopted; if the generated local target point is not a feasible target point, a protection strategy is adopted to generate a new feasible local target point;

[0064] Optionally, the determination principle is: the generated local target point is in a feasible domain.

[0065] Furthermore, a protection strategy is adopted to generate new feasible local target points. The steps are as follows:

[0066] 601, if the local target point generated according to the optimal route target point selection strategy is located in the threat situation detected by the rolling window, then the heuristic target point selection strategy is used to generate the local target point;

[0067] 602, if the k1 points generated by the heuristic target point selection strategy are all infeasible points, then Generate k2 points randomly within the range and arrange them in descending order according to the corresponding x coordinates, and search for local target points in turn according to the heuristic function f(P) = g(P) + h(P);

[0068] 603. If the k2 points generated in step 602 are also infeasible points, the radius of the rolling window is reduced by 10%, and the process returns to step 601 until a feasible local target point is found.

[0069] Step 7: Perform route planning based on the local planning starting point and the generated local target point, and use the point reached by flying only one step distance along the route as the local planning step arrival point;

[0070] Among them, the distance between the local planning step arrival point (x3, y3) and the local planning starting point (x1, y1) is the step distance.

[0071] Step 8: Determine whether the planned final target point has been reached. If so, the planning ends; otherwise, return to step 3.

[0072] The judgment condition for reaching the final target point of the plan is: in a certain planning, the final target point of the global plan appears in a circle with the starting point of the next round of local planning as the center and the rolling window distance as the radius.

[0073] Example 1

[0074] This embodiment provides a method for selecting local target points for UAV rolling route planning, combining Figure 1 , Figure 2 , Figure 3 , Figure 4 As shown, the following steps are included:

[0075] Step 1: Establish and initialize an environment model, wherein the parameter information of the environment model includes: threat area, obstacle area and no-fly zone.

[0076] name shape Center X coordinate (m) Center Y coordinate (m) Circumscribed circle radius (m) No-fly Prism 1133.3 0.1 179.2 obstacle Prism 3732.9 -81.7 213.7 threaten Ellipsoid 2215.8 261.9 198.2

[0077] Step 2: Determine the global planning starting point, the global planning final target point and the pre-planned optimal route;

[0078] Specifically, the planned starting point is (200m, 0m), the planned final target point is (5000m, 0m), and the initial pre-planned optimal route is the line connecting the planned starting point and the planned final target point.

[0079] Step 3: Determine the local planning starting point based on the global planning starting point;

[0080] Specifically, the starting point of the first planning is the global planning starting point of the aircraft (200m, 0m), and the starting point of each subsequent round of planning is the step length arrival point of the previous round of planning.

[0081] Step 4: Update the environmental threat situation based on the environmental model, the local planning starting point, and the situation measurement values ​​within the single-step planning window period;

[0082] Specifically, set the scroll window size to D w =500m.

[0083] Step 5: Determine whether there is a pre-planned optimal route. If so, use the optimal route target point selection strategy to select a local target point on the pre-planned optimal route. If not, use the heuristic target point selection strategy to generate a local target point by a heuristic function.

[0084] Specifically, there is an initial pre-planned optimal route, which is the line connecting the planning starting point and the planning final target point. Therefore, the optimal route target point selection strategy is adopted to select local target points on the pre-planned optimal route.

[0085] Furthermore, with the planning starting point (200m, 0m) as the center, and the rolling window distance D w =500m is the radius, and the coordinates of the intersection of this circle and the pre-planned optimal route are (700m, 0m).

[0086] Step 6: According to the preset judgment principle, determine whether the generated local target point is a feasible target point. If it is feasible, the generated local target point is adopted. If it is not feasible, a protection strategy is adopted to generate a new feasible local target point.

[0087] Specifically, the coordinates of the local target point are (700m, 0m), which is in the feasible domain and is therefore a feasible target point.

[0088] Step 7: Perform route planning based on the local planning starting point and the generated local target point, and use the point reached by flying only one step distance along the route as the local planning step arrival point;

[0089] Specifically, the step distance is D s =200m, the local planning step length arrival point coordinates are (400m, 0m).

[0090] Step 8: Determine whether the planned final target point has been reached. If so, the planning ends; otherwise, return to step 3.

[0091] Specifically, the final target point of the plan is (5000m, 0m), the starting point of the next round of local planning is (400m, 0m), and the rolling window distance is D w =500m, the termination condition has not been reached, and it is necessary to return to step 3 and execute the loop until the condition is met.

[0092] Example 2

[0093] This embodiment provides a method for selecting local target points for UAV rolling route planning, combining Figure 1 , Figure 2 , Figure 3 , Figure 4 As shown, the following steps are included:

[0094] Step 1: Establish and initialize an environment model, wherein the parameter information of the environment model includes: threat area, obstacle area and no-fly zone.

[0095] name shape Center X coordinate (m) Center Y coordinate (m) Circumscribed circle radius (m) No-fly Prism 1350 0 100 obstacle Prism 1882.4 104.3 176.6 threaten Ellipsoid 3625.8 18.7 162.5 threaten Ellipsoid 4678.2 61.9 176.5 threaten Ellipsoid 4389.7 53.6 214.4

[0096] Step 2: Determine the global planning starting point, the global planning final target point and the pre-planned optimal route;

[0097] Specifically, the planned starting point position is (350m, 0m), the planned final target point position is (5550m, 0m), and the initial pre-planned optimal route is the line connecting the planned starting point and the planned final target point.

[0098] Step 3: Determine the local planning starting point based on the global planning starting point;

[0099] Specifically, the starting point of the first planning is the global planning starting point of the aircraft (350m, 0m), and the starting point of each subsequent round of planning is the step length arrival point of the previous round of planning.

[0100] Step 4: Update the environmental threat situation based on the environmental model, the local planning starting point, and the situation measurement values ​​within the single-step planning window period;

[0101] Specifically, set the scroll window size to D w =1000m.

[0102] Step 5: Determine whether there is a pre-planned optimal route. If so, use the optimal route target point selection strategy to select a local target point on the pre-planned optimal route. If not, use the heuristic target point selection strategy to generate a local target point by a heuristic function.

[0103] Specifically, there is an initial pre-planned optimal route, which is the line connecting the planning starting point and the planning final target point. Therefore, the optimal route target point selection strategy is adopted to select local target points on the pre-planned optimal route.

[0104] Furthermore, with the planning starting point (350m, 0m) as the center, the rolling window distance D w =1000m is the radius, and the coordinates of the intersection of this circle and the pre-planned optimal route are (1350m, 0m).

[0105] Step 6: According to the preset judgment principle, determine whether the generated local target point is a feasible target point. If it is feasible, the generated local target point is adopted. If it is not feasible, a protection strategy is adopted to generate a new feasible local target point.

[0106] Specifically, the coordinates of the local target point are (1350m, 0m). This point is in the environmental situation, so it is not in the feasible domain and is an infeasible target point. A protection strategy needs to be adopted to generate a new feasible local target point.

[0107] Furthermore, when the target point generated by the optimal route target point selection strategy is an infeasible point, a heuristic target point selection strategy is required to select a local target point. Specifically, after each rolling window information update, a heuristic function is used to select a local target point, which is expressed as;

[0108] f(P)=g(P)+h(P)

[0109] Among them, P is a point on the rolling window, g(P) is the cost of the drone flying from the current position to the target position, and its value can be estimated based on the position of P and the environmental information in the current window, and h(P) is the cost of flying from P to the end point. Since the information outside the boundary cannot be obtained, h(P) is estimated using the Euclidean distance from P to the end point, and finally the window boundary point P with the minimum cost is taken as the local target point P sub , when the obtained P sub If there is more than one, you can choose any one, namely:

[0110] minf(P)=g(P)+h(P)

[0111] Furthermore, the specific solution method of the local target point selection strategy can be expressed as:

[0112] Step 601, establish a local rectangular coordinate system with the planning starting point (350m, 0m) as the center, the vector from the starting point to the planning final target point (5550m, 0m) as the x-axis, and the direction perpendicular to the vector as the y-axis;

[0113] Step 602, in the local coordinate system, A total of k1 = 4 points are randomly generated within the range, and the 4 randomly generated points are arranged in descending order according to the corresponding x-coordinates as (1350m, 0m), (1334.8m, 178.6m), (1289.7m, 342m), (1057.1m, 707.6m);

[0114] Step 603, according to the arrangement order of step 602, search for local target points in sequence according to the designed heuristic function.

[0115] Furthermore, through the solution, it can be obtained that (1057.1m, 707.6m) is in the feasible domain and is a feasible target point, so this point is used as the local target point.

[0116] Step 7: Perform route planning based on the local planning starting point and the generated local target point, and use the point reached by flying only one step distance along the route as the local planning step arrival point;

[0117] Specifically, the local planning starting point coordinates are (350m, 0m), the planning local target point coordinates are (1057.1m, 707.6m), and the execution step distance is D s =500m, the local planning step length arrival point coordinates are (703.55m, 353.8m).

[0118] Step 8: Determine whether the planned final target point has been reached. If so, the planning ends; otherwise, return to step 3.

[0119] Specifically, the judgment condition for reaching the final target point of the plan is: in a certain planning, the final target point of the plan appears in a circle with the starting point of the following local planning as the center and the rolling window distance as the radius.

[0120] Specifically, the final target point of the plan is (5550m, 0m), the starting point of the next round of local planning is (703.55m, 353.8m), and the rolling window distance is D w = 1000m, the termination condition has not been reached, and it is necessary to return to step 3 and execute the loop until the condition is met.

[0121] In this embodiment, combined Figure 2 The rolling optimization execution process specifically includes the following steps:

[0122] Step 1: If the local target point generated by the optimal route target point selection strategy conflicts with the detected threat environment, a heuristic target point selection strategy is used to generate the local target point;

[0123] Step 2: If the k1=4 points generated by the heuristic target point selection strategy are all infeasible points, then Randomly generate k2 = 4 points within the range and arrange them in descending order according to the corresponding x coordinates, and search for local target points in turn according to the designed heuristic function;

[0124] Step 3: If the k2=4 points generated in step 2 are also infeasible points, reduce the radius of the rolling window by 10%, that is, D w =900m, return to step 1 until a feasible local target point is found.

[0125] The present invention proposes a method for selecting local target points for rolling route planning of unmanned aerial vehicles. The method refers to the local environmental information measured by the unmanned aerial vehicle in real time, draws on the principle of A* algorithm, makes full use of the pre-planned optimal route, takes into account the rapidity and optimality indicators, and designs corresponding protection strategies to ensure the feasibility of generating local target points during the rolling online planning of the unmanned aerial vehicle. The method specifically includes the following technical effects:

[0126] (1) The principle of this method is simple, easy to implement, and requires little computation, which can better meet the rapidity requirements of online route planning;

[0127] (2) This method can make full use of the local environmental information detected in real time, effectively combine the optimization and feedback mechanisms, and achieve the reasonable integration of global environmental information and local information detected in real time;

[0128] (3) This method not only ensures real-time online local planning, but also takes into account the global optimization index. It can flexibly avoid obstacles while always approaching the final target point of the plan, ensuring the correctness of the planning direction while maintaining the optimality of the entire route.

[0129] (4) This method is highly flexible. It can flexibly select the target point generation strategy and flexibly set the parameters according to the focus of the problem to be solved.

[0130] (5) This method is highly versatile and is applicable to both fixed-wing and rotary-wing UAVs. For different models of UAVs, one only needs to change the parameters related to the UAV performance in the algorithm to use it directly.

[0131] Although the embodiments disclosed in the present invention are as above, the contents are only embodiments adopted to facilitate understanding of the present invention and are not intended to limit the present invention. Any technician in the field to which the present invention belongs can make any modifications and changes in the form and details of implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A method for selecting local target points for UAV rolling route planning, characterized in that: The method comprises: Step 1: Establish and initialize an environment model, wherein the parameter information of the environment model includes: threat area, obstacle area and no-fly zone; Step 2: Determine the global planning starting point, the global planning final target point and the pre-planned optimal route; Step 3: Determine the local planning starting point based on the global planning starting point; Step 4: Update the environmental threat situation based on the environmental model, the local planning starting point, and the situation measurement values ​​within the single-step planning window period; Step 5: Determine whether there is a pre-planned optimal route. If so, use the optimal route target point selection strategy to select a local target point on the pre-planned optimal route. If not, use the heuristic target point selection strategy to generate a local target point by a heuristic function. Step 6: According to the preset judgment principle, determine whether the generated local target point is a feasible target point. If the generated local target point is a feasible target point, the generated local target point is adopted; if the generated local target point is not a feasible target point, a protection strategy is adopted to generate a new feasible local target point; Step 7: Perform route planning based on the local planning starting point and the generated local target point, and use the point reached by flying only one step distance along the route as the local planning step arrival point; Step 8: Determine whether the planned final target point has been reached. If so, the planning ends; otherwise, return to step 3.

2. The local target point selection method according to claim 1, characterized in that: Step 5 includes: 501, establish a local rectangular coordinate system with the local planning starting point as the center of the circle, the vector from the starting point to the final target point of the planning as the x-axis, and the direction perpendicular to the vector as the y-axis; 502, in the local rectangular coordinate system, A total of k1 points are randomly generated within the range, and the randomly generated k1 points are arranged in descending order according to their corresponding x-coordinates; 503. According to the arrangement order of step 502, local target points are searched in sequence according to the heuristic function f(P)=g(P)+h(P).

3. The local target point selection method according to claim 2, characterized in that: The heuristic target point selection strategy is: after each rolling window information update, the heuristic function f(P)=g(P)+h(P) is used to select the local target point, where P is the point on the rolling window, g(P) is the cost of the drone flying from the current position to the target position, and h(P) is the cost of flying from P to the end point.

4. The local target point selection method according to claim 1, characterized in that: In step 6, a protection strategy is adopted to generate a new feasible local target point. The steps are: 601, if the local target point generated according to the optimal route target point selection strategy is located in the threat situation detected by the rolling window, then the heuristic target point selection strategy is used to generate the local target point; 602, if the k1 points generated by the heuristic target point selection strategy are all infeasible points, then Generate k2 points randomly within the range and arrange them in descending order according to the corresponding x coordinates, and search for local target points in turn according to the heuristic function f(P) = g(P) + h(P); 603. If the k2 points generated in step 602 are also infeasible points, the radius of the rolling window is reduced by 10%, and the process returns to step 601 until a feasible local target point is found.

5. The local target point selection method according to claim 1, characterized in that: The judgment principle in step 6 is: the generated local target point is in the feasible domain.

6. The local target point selection method according to claim 1, characterized in that: The judgment condition for reaching the final target point of the plan in step 8 is: during planning, the final target point of the global plan appears in a circle with the starting point of the next round of local planning as the center and the rolling window distance as the radius.

7. The local target point selection method according to claim 1, characterized in that: The starting point of the first planning in step 3 is the global planning starting point of the aircraft (x1, y1), and the starting point of each subsequent round of planning is the step length arrival point of the previous round of planning.

8. A local target point selection device for UAV rolling route planning, characterized in that: The device is used to implement the local target point selection method for UAV rolling route planning described in claim 1.

Images