High-speed unmanned aerial vehicle taboo fallback sparse A-Star flight path planning method

Abstract

The invention discloses a high-speed unmanned aerial vehicle taboo fallback sparse A-Star flight path planning method, which comprises the following steps of: determining task parameters of a high-speed unmanned aerial vehicle according to a superior flight task, initializing a flight path starting node, performing high-node expansion under constraint based on a current node, judging the legality of the expanded node, performing node heuristic cost calculation, and performing high-speed unmanned aerial vehicle taboo fallback sparse A-Star flight path planning. The method comprises the following steps: selecting an OPEN table and a TABU table, setting legal nodes in the OPEN table, setting illegal nodes in the TABU table, if the legal nodes exist, selecting minimum-cost nodes from expansion nodes of a current node in the OPEN table to a CLOSE table, otherwise, carrying out rollback processing on the current node, finally carrying out termination processing on a final node in the CLOSE table, and selecting a preliminary planning track formed by serial sequence in the CLOSE table. And storing the selected nodes in a PATH table, and determining a final planned track. The method can solve the problems of more turns and voyage waste during track generation, is low in operation memory requirement, is rapid in generation, and is easy for engineering realization.

Description

Technical Field

[0001] The invention belongs to the field of unmanned aerial vehicle (UAV) trajectory planning, and in particular relates to a high-speed UAV taboo-backoff sparse A-Star trajectory planning method. Background Art

[0002] According to the superior mission requirements, the trajectory planning system needs to autonomously explore the flight path to reach the mission end point in the three-dimensional mission space. For high-speed UAVs, the flight path must not only meet its performance constraints and terrain environment constraints, but also avoid potential threat areas. It is also necessary to consider the impact of actual flight altitude and position errors on the planned trajectory to reduce the probability of impromptu adjustments to the planned trajectory during the actual flight.

[0003] In the process of large-scale UAV trajectory planning, the trajectory planning system needs to deal with a large number of inconsistent constraints, which places high demands on the memory required for system operation, and also means an overly lengthy running time, which is inconsistent with the low cost and fast response required by today's high-speed UAV application scenarios. In addition, in order to make the resulting trajectory meet the mission reachability requirements, the introduction of expert experience is also crucial to the system. Therefore, blindly using traditional trajectory planning algorithms, such as dynamic programming, Dubins and other deterministic algorithms, and genetic algorithms, ant colony algorithms and other random algorithms, without considering the actual engineering scenarios and mission requirements of the trajectory planning system, will not be able to carry out effective engineering implementation. Summary of the invention

[0004] In view of the above problems, the object of the present invention is to provide a high-speed UAV taboo-backoff sparse A-Star trajectory planning method.

[0005] The specific technical solution for achieving the purpose of the present invention is:

[0006] A high-speed UAV taboo-backoff sparse A-Star trajectory planning method comprises the following steps:

[0007] Step 1: Determine the starting point P of the high-speed UAV mission according to the upper-level flight mission s , end point P e and the task setting speed V m ;

[0008] Step 2: Set up the node sets required in the planning process, including OPEN table, CLOSE table, TABU table, PATH table, and initialize the track starting node N (i) |i=1;

[0009] Step 3: Based on the current node N (i) , given a step length L and a maximum turning angle ψ max Under the constraints of the climbing rate C(z), the height settlement and sparse node expansion are performed to obtain N(i+1) ;

[0010] Step 4: Determine the expansion node N based on environmental constraints (i+1) The legitimacy of the node is determined by the node heuristic cost J (i+1)(j)(k) Calculate and place the legal nodes in the OPEN table and the illegal nodes in the TABU table. If there is a legal node, go to step 5, otherwise go to step 6 and set the back-off step number B. s =1;

[0011] Step 5: For the extended node of the current node in the OPEN table, select the node with the minimum cost to the CLOSE table. If the optimal node and the end point meet the distance condition, go to step 7; otherwise, let i=i+1 and go to step 3 for iterative expansion.

[0012] Step 6: If i=1, then exit the expansion process and return the planning failure information; otherwise, (i) Perform rollback processing, delete the current node from the CLOSE table and move it to the TABU table, and select For the current node, the optimal selection is made for its expansion node. If there is no The expansion node re-enters step 6 and sets the number of steps back to B s =B s +1, if the optimal node and the end point meet the distance condition, go to step 7, otherwise let i=i+1 and go to step 3 for iterative expansion;

[0013] Step 7: Perform termination processing on the last node in the CLOSE table, that is, modify the node position information to the termination point;

[0014] Step 8: Select the preliminary planned trajectory formed by the sequence in the CLOSE table, store the selected nodes in the PATH table, and determine the final planned trajectory.

[0015] Compared with the prior art, the present invention has the following beneficial effects:

[0016] The present invention aims at the problem of high-speed UAV trajectory planning in three-dimensional space. Based on the heuristic sparse A* algorithm idea and the height settlement mechanism, the taboo area in the three-dimensional space is mapped to the constraint space in real time by designing a taboo mechanism. In combination with the performance constraints of the high-speed UAV and the environmental constraints including the threat area and the terrain elevation, the sparse nodes are expanded in a single step. When there are no legal nodes, a fallback mechanism is introduced to meet the trajectory accessibility requirements. At the same time, a trajectory selection mechanism is designed to solve the problems of more turns and wasted range in the generated trajectory. The present invention has small computing memory requirements, fast generation, and is easy to implement in engineering.

[0017] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flowchart of a high-speed UAV taboo-backoff sparse A-Star trajectory planning method provided by an embodiment of the present invention.

[0019] Figure 2 It is a two-dimensional schematic diagram of node sparse expansion in an embodiment of the present invention.

[0020] Figure 3 It is a three-dimensional schematic diagram of node sparse expansion in an embodiment of the present invention.

[0021] Figure 4 Schematic diagram of terrain constraints of extended nodes in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] Example

[0023] Combination Figure 1 , a high-speed UAV taboo backoff sparse A-Star trajectory planning method, comprising the following steps:

[0024] Step 1: Determine the starting point P of the high-speed UAV mission according to the upper-level flight mission s , end point P e and the task setting speed V m :

[0025] Determine the high-speed UAV mission starting point P according to the flight mission s (x s ,y s ,z s ), end point P e (x e ,y e ,z e ) and the task setting speed V m , where x s 、x e are the forward positions of the starting point and the end point respectively; y s ,y e are the horizontal positions of the starting point and the end point respectively; z s 、z e They are the vertical heights of the starting point and the ending point respectively.

[0026] Step 2: Set up the node sets required in the planning process, including OPEN table, CLOSE table, TABU table, PATH table, and initialize the track starting node N (i) |i=1;

[0027] The OPEN table is used to store legal nodes, the TABU table is used to store illegal nodes, the CLOSE table is used to store initially planned track nodes, and the PATH table is used to store finally formed track planning nodes.

[0028] Step 3: Based on the current node N (i) , given a step length L and a maximum turning angle ψ max Under the constraints of the climbing rate C(z), the height settlement and sparse node expansion are performed to obtain N (i+1) :

[0029] Combination Figure 2 , based on the current node N (i) First, in the two-dimensional task space, given the step size L and the maximum turning angle ψ max Under the constraint conditions, the expansion is performed in the fan-shaped area based on the sparse A-Star algorithm, where the expansion angle is 2ψ max , the expansion radius is L, and the expansion area is evenly divided into N L Equal parts, then the number of optional nodes to be expanded is N L +1;

[0030] Afterwards, combined Figure 3 , in the three-dimensional task space, under the given climbing rate C(z) constraint, obtain the vertical expansion height range of the node Δz=L / V m C(z), given height interval H s , then the total number of node vertical expansion hierarchical levels is N z =1+ceil(Δz / H s ), where ceil(·) is the upward rounding function, the current number of expandable nodes is N = (N L +1)N z ;

[0031] Finally, we get the current node N (i) All the expansion nodes are:

[0032] N (i+1) (id pre(i+1)(j)(k) ,x (i+1)(j)(k) ,y (i+1)(j)(k) ,z (i+1)(j)(k) ,ψ (i+1)(j)(k) ,J (i+1)(j)(k) ,f (i+1)(j)(k) )

[0033] Based on the geometric relationship, the node related elements can be obtained in the following way:

[0034]

[0035] where i>1, j∈[1,N L+1],k∈[1,N z ], id pre(i+1)(j)(k) 、x (i+1)(j)(k) ,y (i+1)(j)(k) , (i+1)(j)(k) 、z (i+1)(j)(k) They represent the forward node identification, forward position, lateral position, heading, and longitudinal height of the jth node in the two-dimensional expansion direction and the kth node in the longitudinal expansion direction in the i+1th node sequence respectively; L min Indicates the minimum given step size;

[0036] Step 4: Combine Figure 4 , based on environmental constraints, determine the expansion node N (i+1) The legitimacy of the node is determined by the node heuristic cost J (i+1)(j)(k) Calculate and place the legal nodes in the OPEN table and the illegal nodes in the TABU table. If there is a legal node, go to step 5, otherwise go to step 6 and set the back-off step number B. s =1:

[0037] Step 4-1: Determine environmental constraints, including terrain constraints, threat constraints, and taboo constraints:

[0038] The environmental constraints include:

[0039] (1) Terrain constraints

[0040] The trajectory planning of the UAV needs to meet the terrain constraints to avoid the UAV from hitting the ground. In order to ensure flight safety, the maximum side deviation L safe is the lateral safety distance, and at a given safety height z safe Under the constraints, the flight segment FL(N (i) ,N (i+1) ), and then perform lateral translation to obtain plane FS(P L(i) ,P R(i) ,P L(i+1) ,P R(i+1) ), point P in the plane fs (x fs ,y fs ,z fs ) must satisfy the following constraints

[0041] z fs ≥z fsm (x fs ,y fs )+z safe

[0042] Among them, z fsm (x fs ,y fs ) is point P fs The terrain height; P L(i) , PR(i) , P L(i+1) and P R(i+1) N (i) and N (i+1) Points that translate to the left and right;

[0043] (2) Threat Constraint

[0044] The threat model faced by the UAV during flight is established. The known threat source is P d(ix) (x d(ix) ,y d(ix) ,z d(ix) ,r d(ix )), ix∈[1,x], x is the number of threats, and r d(ix) For its threat radius, to prevent drones from entering the threat area, the extended nodes must meet the following constraints:

[0045]

[0046] Among them, CosLaw (P A ,P B ,P C ) is the cosine theorem function, which is used to find the line segment FL(P B , P C ) corresponds to the angle value; Dis(·) represents the straight-line distance between two points;

[0047] (3) Taboo constraints

[0048] In the real-time trajectory planning process, the area where the illegal extension node is located is marked as a taboo area and stored in the TABU table. The extension node must meet the following constraints:

[0049]

[0050] Among them, N tb (m) is a node in the TABU table, z tb (m) is the corresponding height of the node, m∈[1,length(TABU)], and length(·) is the function of the number of nodes in the table.

[0051] Step 4-2: Determine the legal nodes and illegal nodes in the expanded nodes based on the environmental constraints, and determine their node heuristic costs respectively:

[0052] If the extended node meets the terrain constraint, threat constraint and taboo constraint, the extended node is considered to be a legal node. If it does not meet any of the terrain constraint, threat constraint and taboo constraint, the extended node is considered to be an illegal node.

[0053] At this time, set the node heuristic cost J of the illegal node (i+1)(j)(k) =Jmax , and put the illegal nodes in the TABU table;

[0054] The node heuristic cost of a legal node is:

[0055] J (i+1)(j)(k) =f(n)

[0056] f(n)=g(n)+h(n)

[0057]

[0058] Among them, g(n) is the actual cost from the starting point to the current node n; h(n) represents the estimated cost from the current node to the target node, T(n) represents the threat cost, and L T , L d They represent the threat distance and the danger distance respectively. That is, if the distance between the extended node and all obstacles in the task area is greater than the threat distance, the cost is 0; if the extended node is within the danger distance of an obstacle, the cost is infinite, indicating that the extended node is illegal; k 1 , k 2 , k 3 is the cost weight;

[0059] After calculating the legal extension nodes in sequence according to the above design, we get J (i+1)(j)(k) =f(n), put the legal nodes into the OPEN table.

[0060] Step 4-3: Determine whether there is a legal node. If there is a legal node, store the legal node in the OPEN table, put the illegal node in the TABU table, and go to step 5. If there is no legal node, go to step 6 and set the back-off step number B. s =1.

[0061] Step 5: For the extended node of the current node in the OPEN table, select the node with the minimum cost to the CLOSE table. If the optimal node and the end point meet the distance condition, go to step 7. Otherwise, let i=i+1 and go to step 3 for iterative expansion:

[0062] Identify the forward node in the OPEN table pre = i's expansion node, select the node with the minimum cost as the optimal node:

[0063] minJ{N Open |id pre =i}

[0064] Let node expand flag f (i+1) =1, and put the optimal node in the CLOSE table;

[0065] If the optimal node and the terminal point Pe If the distance condition is met, go to step 7, otherwise let i=i+1 and go to step 3 for iterative expansion:

[0066]

[0067] Where L c 、z c They are the horizontal and vertical approximate distance thresholds set respectively.

[0068] Step 6: For the current node N (i) Perform rollback processing, delete the current node from the CLOSE table and move it to the TABU table, and select For the current node, the optimal selection is made for its expansion node. If there is no The expansion node re-enters step 6 and sets the number of steps back to B s =B s +1, if the optimal node and the end point meet the distance condition, go to step 7, otherwise let i=i+1 and go to step 3 for iterative expansion:

[0069] If i = 1, then the expansion process is exited and the planning failure message is returned. Otherwise, if there is no legal expansion node, in order to meet the track reachability requirement, the current node N (i) Perform rollback processing, delete the current node from the CLOSE table and move it to the TABU table, and select For the current node, select the optimal extension node:

[0070] minJ{N Open |id pre =iB s}

[0071] If there is no The expansion node re-enters step 6 and sets the number of steps back to B s =B s +1, if the optimal node and the end point meet the distance condition:

[0072]

[0073] Then go to step 7, otherwise let i=i+1 and go to step 3 for iterative expansion.

[0074] Step 7: Perform termination processing on the last node in the CLOSE table, that is, modify the node position information to the termination point;

[0075] Step 8: Select the preliminary planned track formed by the sequence in the CLOSE table, store the selected nodes in the PATH table, and determine the final planned track:

[0076] The nodes in the CLOSE table are serially ordered to form a preliminary planned trajectory CL{N∈CLOSE}. However, in order to save computing resources and computing time, this trajectory is not calculated and optimally selected for all nodes in the OPEN table every time like the conventional sparse A-star algorithm. Instead, it only calculates and optimally selects the single-step expansion nodes, and only introduces a backtracking mechanism when there is no legal expansion node to avoid the occurrence of no solution. Therefore, the preliminary planned trajectory will have more bends and is not optimal;

[0077] Therefore, this step designs a selection mechanism. For any two nodes in the preliminary planned trajectory CL{N∈CLOSE}, if the two nodes meet the environmental constraints and the performance constraints of the high-speed UAV, the intermediate nodes between the two nodes are screened out to reduce the trajectory curvature and reduce the flight risk.

[0078] The final planned trajectory is formed and the planned trajectory nodes are stored in the PATH table.

[0079] The present invention aims at the problem of high-speed UAV trajectory planning in three-dimensional space. Based on the heuristic sparse A* algorithm idea and the height settlement mechanism, the taboo area in the three-dimensional space is mapped to the constraint space in real time by designing a taboo mechanism. In combination with the performance constraints of the high-speed UAV and the environmental constraints including the threat area and the terrain elevation, the sparse nodes are expanded in a single step. When there are no legal nodes, a fallback mechanism is introduced to meet the trajectory accessibility requirements. At the same time, a trajectory selection mechanism is designed to solve the problems of more turns and wasted range in the generated trajectory. The present invention has small computing memory requirements, fast generation, and is easy to implement in engineering.

[0080] The above embodiments show and describe the basic principles and main features of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A high-speed UAV taboo-backoff sparse A-Star trajectory planning method, characterized in that: The following steps are involved: Step 1: Determine the starting point P of the high-speed UAV mission according to the upper-level flight mission s , end point P e and the task setting speed V m ; Step 2: Set up the node sets required in the planning process, including OPEN table, CLOSE table, TABU table, PATH table, and initialize the track starting node N (i) |i=1; Step 3: Based on the current node N (i) , given a step length L and a maximum turning angle ψ max Under the constraints of the climbing rate C(z), the height settlement and sparse node expansion are performed to obtain N (i+1) ; Step 4: Determine the expansion node N based on environmental constraints (i+1) The legitimacy of the node is determined by the node heuristic cost J (i+1)(j)(k) Calculate and place the legal nodes in the OPEN table and the illegal nodes in the TABU table. If there is a legal node, go to step 5, otherwise go to step 6 and set the back-off step number B. s =1; Step 5: For the extended node of the current node in the OPEN table, select the node with the minimum cost to the CLOSE table. If the optimal node and the end point meet the distance condition, go to step 7; otherwise, let i=i+1 and go to step 3 for iterative expansion. Step 6: If i=1, then exit the expansion process and return the planning failure information; otherwise, (i) Perform rollback processing, delete the current node from the CLOSE table and move it to the TABU table, and select N (i-Bs) For the current node, the optimal selection is made for its expansion node. If there is no N in the OPEN table (i-Bs) The expansion node re-enters step 6 and sets the number of steps back to B s =B s +1, if the optimal node and the end point meet the distance condition, go to step 7, otherwise let i=i+1 and go to step 3 for iterative expansion; Step 7: Perform termination processing on the last node in the CLOSE table, that is, modify the node position information to the termination point; Step 8: Select the preliminary planned trajectory formed by the sequence in the CLOSE table, store the selected nodes in the PATH table, and determine the final planned trajectory.

2. The high-speed UAV taboo-backoff sparse A-Star trajectory planning method according to claim 1 is characterized in that: Determine the starting point P of the high-speed UAV mission in step 1 s , end point P e and the task setting speed V m , specifically: Determine the high-speed UAV mission starting point P according to the flight mission s (x s ,y s ,z s ), end point P e (x e ,y e ,z e ) and the task setting speed V m , where x s 、x e are the forward positions of the starting point and the end point respectively; y s ,y e are the horizontal positions of the starting point and the end point respectively; z s 、z e They are the vertical heights of the starting point and the ending point respectively.

3. The high-speed UAV taboo-backoff sparse A-Star trajectory planning method according to claim 2 is characterized in that: The step 3 performs height settling and sparse node expansion to obtain N (i+1) , specifically: Based on the current node N (i) First, in the two-dimensional task space, given the step size L and the maximum turning angle ψ max Under the constraint conditions, the expansion is performed in the fan-shaped area based on the sparse A-Star algorithm, where the expansion angle is 2ψ max , the expansion radius is L, and the expansion area is evenly divided into N L Equal parts, then the number of optional nodes to be expanded is N L +1; Afterwards, in the three-dimensional task space, under the given climbing rate C(z) constraint, the height range of the node longitudinal expansion Δz = L / V is obtained. m C(z), given height interval H s , then the total number of node vertical expansion hierarchical levels is N z =1+ceil(Δz / H s ), where ceil(·) is the upward rounding function, the current number of expandable nodes is N = (N L +1)N z ; Finally, we get the current node N (i) All the expansion nodes are: N (i+1) (id pre(i+1)(j)(k) ,x (i+1)(j)(k) ,the (i+1)(j)(k) ,z (i+1)(j)(k) ,ψ (i+1)(j)(k) ,J (i+1)(j)(k) ,f (i+1)(j)(k) ) where i>1, j∈[1,N L +1],k∈[1,N z ], id pre(i+1)(j)(k) 、x (i+1)(j)(k) ,y (i+1)(j)(k) , (i+1)(j)(k) 、z (i+1)(j)(k) They represent the forward node identification, forward position, lateral position, heading, and longitudinal height of the jth node in the two-dimensional expansion direction and the kth node in the longitudinal expansion direction in the i+1th node sequence respectively; L min Indicates the minimum given step size.

4. The high-speed UAV taboo-backoff sparse A-Star trajectory planning method according to claim 1 is characterized in that: The determination of the expansion node N in step 4 (i+1) The legality of Step 4-1, determine environmental constraints, including terrain constraints, threat constraints and taboo constraints; Step 4-2: determine the legal nodes and illegal nodes in the expanded nodes based on the environmental constraints, and determine their node heuristic costs respectively; Step 4-3: Determine whether there is a legal node. If there is a legal node, store the legal node in the OPEN table, put the illegal node in the TABU table, and go to step 5. If there is no legal node, go to step 6 and set the back-off step number B. s =1.

5. The high-speed UAV taboo-backoff sparse A-Star trajectory planning method according to claim 4 is characterized in that: The environmental constraints include: (1) Terrain constraints In order to ensure flight safety, the maximum lateral deviation L safe is the lateral safety distance, and at a given safety height z safe Under the constraints, the flight segment FL(N (i) ,N (i+1) ), and then perform lateral translation to obtain plane FS(P L(i) ,P R(i) ,P L(i+1) ,P R(i+1) ), point P in the plane fs (x fs ,y fs ,z fs ) must satisfy the following constraints z fs ≥z fsm (x fs ,y fs )+z safe Among them, z fsm (x fs ,y fs ) is point P fs The terrain height; P L(i) , P R(i) , P L(i+1) and P R(i+1) N (i) and N (i+1) Points that translate to the left and right; (2) Threat Constraint The known threat source is P d(ix) (x d(ix) ,y d(ix) ,z d(ix) ,r d(ix) ), ix∈[1,x], x is the number of threats, and r d ( ix ) is its threat radius. To prevent drones from entering the threat area, the extended nodes must meet the following constraints: Among them, CosLaw (P A ,P B ,P C ) is the cosine theorem function, which is used to find the line segment FL(P B , P C ) corresponds to the angle value; Dis(·) represents the straight-line distance between two points; (3) Taboo constraints In the real-time trajectory planning process, the area where the illegal extension node is located is marked as a taboo area and stored in the TABU table. The extension node must meet the following constraints: Among them, N tb ( m ) is a node in the TABU table, z tb ( m ) is the corresponding height of the node, m∈[1,length(TABU)], and length(·) is the function of the number of nodes in the table.

6. The high-speed UAV taboo-backoff sparse A-Star trajectory planning method according to claim 4 is characterized in that: In step 4-2, the legal nodes and illegal nodes in the extended nodes are determined based on the environmental constraints, and their node heuristic costs are determined respectively, specifically: If the extended node meets the terrain constraint, threat constraint and taboo constraint, the extended node is considered to be a legal node. If it does not meet any of the terrain constraint, threat constraint and taboo constraint, the extended node is considered to be an illegal node. At this time, set the node heuristic cost J of the illegal node (i+1)(j)(k) =J max ; The node heuristic cost of a legal node is: J (i+1)(j)(k) =f(n) f(n)=g(n)+h(n) Among them, g(n) is the actual cost from the starting point to the current node n; h(n) represents the estimated cost from the current node to the target node, T(n) represents the threat cost, and L T , L d They represent threat distance and danger distance respectively, that is, if the distance from the extended node to all obstacles in the task area is greater than the threat distance, the cost is 0; if the extended node is within the danger distance of an obstacle, the cost is infinite, indicating that the extended node is illegal; k1, k2, and k3 are cost weights.

7. The high-speed UAV taboo-backoff sparse A-Star trajectory planning method according to claim 1 is characterized in that: The minimum cost node in step 5 is selected and the distance from the end point is determined as follows: Identify the forward node in the OPEN table pre = i's expansion node, select the node with the minimum cost as the optimal node: myJ{N Open |id pre =i} Let node expand flag f (i+1) =1, and put the optimal node in the CLOSE table; If the optimal node and the terminal point P e If the distance condition is met, go to step 7, otherwise let i=i+1 and go to step 3 for iterative expansion: Where L c 、z c They are the horizontal and vertical approximate distance thresholds set respectively.

8. The high-speed UAV taboo-backoff sparse A-Star trajectory planning method according to claim 1 is characterized in that: The current node N in step 6 (i) Perform rollback processing, specifically: If i = 1, then the expansion process is exited and the planning failure message is returned. Otherwise, if there is no legal expansion node, in order to meet the track reachability requirement, the current node N (i) Perform rollback processing, delete the current node from the CLOSE table and move it to the TABU table, and select For the current node, select the optimal extension node: myJ{N Open |id pre =iB s } If there is no The expansion node re-enters step 6 and sets the number of steps back to B s =B s +1, if the optimal node and the end point meet the distance condition: Then go to step 7, otherwise let i=i+1 and go to step 3 for iterative expansion.

9. The high-speed UAV taboo-backoff sparse A-Star trajectory planning method according to claim 1 is characterized in that: Determining the final planned track in step 8 is specifically as follows: Sequence the nodes in the CLOSE table to form a preliminary planned trajectory CL{N∈CLOSE}; For any two nodes in the preliminary planned trajectory CL{N∈CLOSE}, if the two nodes meet the environmental constraints and the performance constraints of the high-speed UAV, the intermediate nodes between the two nodes are screened out; The final planned trajectory is formed and the planned trajectory nodes are stored in the PATH table.

Images