Dynamic obstacle avoidance method for unmanned aerial vehicles based on repeated game

Abstract

The present invention provides a method for dynamic obstacle avoidance for unmanned aerial vehicles (UAVs) based on iterative gaming, comprising the following steps: establishing a set of game participants by treating UAVs as game participants; calculating a game payoff value based on the states between the game participants, and measuring the quality of the game strategy selected by the UAV by calculating the game payoff value; enabling the game participants to quickly converge to the optimal strategy during the iterative gaming process, updating a mixed strategy based on the decisions of the game participants, and making decisions based on the mixed strategy and the game payoff, so that the payoffs of the game participants reach a Nash equilibrium and the corresponding obstacle avoidance strategy is selected. The method for dynamic obstacle avoidance for UAVs based on iterative gaming enables the UAV system to perform obstacle avoidance maneuvers in real time when facing moving obstacles or moving targets, quickly calculates a motion path with better performance for the UAV, and ensures that the UAV safely operates to its respective target point.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a dynamic obstacle avoidance method for an unmanned aerial vehicle based on repeated games, an electronic device and a storage medium. Background Art

[0002] Unmanned aerial vehicles (UAVs) use sensors to sense their surroundings during flight. To ensure they can autonomously avoid obstacles, appropriate obstacle avoidance methods must be designed. These methods allow the UAV to perform avoidance maneuvers based on information such as obstacle location, type, and its own position, ensuring it can quickly and safely avoid obstacles. Obstacle avoidance is a key indicator of a UAV's autonomous capabilities.

[0003] The existing obstacle avoidance algorithms are not ideal in complex environments and have poor adaptability to the environment, so how to improve the performance of unmanned aerial vehicle obstacle avoidance algorithms is a crucial issue. Summary of the Invention

[0004] In order to solve the problems existing in the prior art, the present invention provides a dynamic obstacle avoidance method for unmanned aerial vehicles based on repeated games, so that the unmanned aerial vehicle system can perform obstacle avoidance maneuvers in real time when facing moving obstacles and moving targets, and quickly calculate a motion path with better performance for the unmanned aerial vehicle, ensuring that the unmanned aerial vehicle can safely operate to its respective target point.

[0005] To achieve the above objectives, the present invention provides a method for unmanned aerial vehicle dynamic obstacle avoidance based on repeated game, comprising the following steps:

[0006] Take the UAV as a participant in the game and establish a set of game participants;

[0007] Calculating a game gain value based on the status between the game participants, and measuring the pros and cons of the game strategy selected by the unmanned aerial vehicle by calculating the game gain value;

[0008] The game participants can quickly converge to the optimal strategy during repeated games, update the mixed strategy according to the decisions of the game participants, make decisions based on the mixed strategy and the game benefits, so that the benefits of the game participants reach Nash equilibrium and choose the corresponding obstacle avoidance strategy.

[0009] Furthermore, the step of establishing a set of game participants using the UAV as a game participant also includes:

[0010] Assume that the UAV encounters H obstacles OB during operation h (h=1,…,H), the current position coordinates of the UAV are (x r ,y r), the coordinates of the closest point of each obstacle to the UAV are (x h ,y h ), the set of game participants is A, and the UAV is placed in the set of game participants A;

[0011] Calculate the distance between the UAV and the obstacle. When Formula 1 is satisfied, the obstacle enters the UAV's safety range and plays a game of obstacle avoidance with the UAV.

[0012]

[0013] Among them, d safe is the safe distance. When the distance between the UAV and the obstacle is less than or equal to d safe When the obstacle OB h Add the set of game participants A and perform obstacle avoidance maneuvers;

[0014] When performing obstacle avoidance, the original obstacles that appear in the working environment within the safety range of the UAV are regarded as game participants.

[0015] Furthermore,

[0016] The strategy set chosen by the game participants consists of four strategy sets;

[0017] The first strategy set is for the game participants to change their own speed while keeping other states unchanged;

[0018] The second strategy set is for the game participants to change their own running direction while keeping other states unchanged;

[0019] The third strategy set is that the running speed and direction of the game participants remain unchanged;

[0020] The fourth strategy set is a stop strategy, where the game participants keep their current positions unchanged.

[0021] Furthermore, the step of calculating the game benefit value based on the status between the game participants and measuring the pros and cons of the game strategy selected by the unmanned aerial vehicle by calculating the game benefit value also includes:

[0022] Assuming there are a total of a game participants, the profit calculation formula is:

[0023]

[0024] Among them, R is the coefficient for judging whether the strategy can ensure the safety of the UAV. When the selected strategy can ensure safety, R = 1, and when it cannot ensure safety, R = 0; d safe is the safe distance range of the UAV; aved is A iThe average distance between other game participants; mind is A i The minimum distance between other game participants; A i A is the strategy adopted by game participant i; -i is the strategy adopted by all game players except player i; X(A i ) is the consumption value of different strategies.

[0025] Furthermore, the step of enabling the game participants to quickly converge to the optimal strategy during the game process, updating the mixed strategy according to the decisions of the game participants, making decisions based on the mixed strategy and the game benefits so that the benefits of the game participants reach a Nash equilibrium and selecting the corresponding obstacle avoidance strategy also includes:

[0026] Assume that the number of repeated games is T, the number of current games is t, the number of optional strategies for game participants is n, and the initialization of game participant A is i Mixed strategy, t = 1 and game player A i The initial probability is:

[0027]

[0028] in, n is the number of strategies available to game participants; are the probabilities of the game participants choosing the 1st, 2nd, …, nth strategies at the initial moment.

[0029] Furthermore,

[0030] Combine the mixed strategy with the decision of each iterative game player, so that game player A i The probability of changes with the number of iterations, and the calculation formula is:

[0031]

[0032] Among them, the parameters Function I(τ i )(i=1,2,…,n) is the decision record function, which records the number of times the current strategy is selected and the number of times the current strategy is not selected in the previous t-1 iterations. The calculation formula is:

[0033]

[0034] Among them, PR i For game participant A i The game priority of ; β is a constant greater than 0, which is used to record the number of times the game participant chooses the current or other strategies; when the game participant is a static obstacle, PR is 0.

[0035] Furthermore,

[0036] The initial priority PR of the game participants is 0, and the priority is changed according to the priority calculation formula, which is:

[0037]

[0038] Among them, L i For game participant A i The distance to the target point, d pr To determine the priority distance, when the priority of the game participant is 1, the third strategy set is selected.

[0039] Furthermore, the information of each of the game participants includes the participant's current location (i=1,2,…,a), running speed The angle θ between the running direction and the X-axis, priority PR.

[0040] To achieve the above-mentioned purpose, the present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a program running on the processor, and when the processor runs the program, it executes the steps of the above-mentioned unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game.

[0041] To achieve the above objectives, the present invention also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed, execute the steps of the above-mentioned unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game.

[0042] The unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game of the present invention has the following beneficial effects:

[0043] An iterative game approach is adopted to select the obstacle avoidance strategy. Each game participant selects the behavior he should take based on the mixed strategy, so that each player can quickly converge to the optimal strategy. It is safe and reliable, can adapt to a variety of complex environments, and can effectively solve the collaborative dynamic obstacle avoidance problem of the unmanned aerial vehicle system, thereby avoiding all obstacles efficiently and safely.

[0044] Other features and advantages of the present invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0046] Figure 1 is a flow chart of a method for avoiding dynamic obstacles of an unmanned aerial vehicle based on repeated games according to the present invention;

[0047] Figure 2 A schematic diagram of an obstacle avoidance process for an unmanned aerial vehicle based on iterative game theory according to an embodiment of the present invention;

[0048] Figure 3 A schematic diagram of a first group of obstacle avoidance game processes performed by three unmanned aerial vehicles according to an embodiment of the present invention;

[0049] Figure 4 Schematic diagram of a second obstacle avoidance game process performed by three unmanned aerial vehicles according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0051] Example 1

[0052] Game theory, also known as strategy theory or game theory, originated from games like chess and bridge and is often used for strategic analysis in economics and computer science. John von Neumann was the first to prove the principles of game theory and further developed its foundations and theoretical framework. Game theory primarily studies how, under certain circumstances, individuals choose strategies based on the current situation and following specific rules to maximize the interests of both parties and achieve equilibrium.

[0053] Figure 1 The following is a flow chart of the unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game according to the present invention. Figure 1 , the unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game of the present invention is described in detail.

[0054] In step 101, the UAVs are regarded as participants in the game and a player set is established.

[0055] Preferably, it is assumed that the UAV encounters H obstacles OB during operation. h (h=1,…,H), the current position coordinates of the UAV are (x r ,y r ), the coordinates of the closest point of each obstacle to the UAV are (x h ,y h ). Let the set of game participants be A, and put the unmanned aerial vehicle into the set of game participants A.

[0056] Preferably, the distance between the UAV and the obstacle is calculated. When Formula 1 is satisfied, the obstacle is considered to have entered the UAV's safety range and needs to engage in a game of obstacle avoidance with the UAV. During the obstacle avoidance process, if the original obstacle in the working environment appears within the UAV's safety range, it is also considered a participant in the game.

[0057]

[0058] Among them, d safe The safe distance is when the distance between the UAV and the obstacle is less than or equal to d safe When the obstacle OB h Add the set A of game participants to perform obstacle avoidance maneuvers.

[0059] Preferably, the strategy set that can be selected by the game participants is S, and the player's S includes four strategy sets, as shown in Formula 2.

[0060] S={CV,CD,NC,ST} Formula 2

[0061] Preferably, the first strategy set CV strategy represents the speed at which the game participants change themselves while keeping other states unchanged.

[0062] In this embodiment, when the UAV selects the CV strategy, the current player's speed is reduced by υ, that is, the speed is adjusted to v = v - υ. The size of the parameter υ is related to the environment in which the UAV is located, υ = ρ × mind. dA i,k For game participant A i The distance between the participants and the other participants is calculated as shown in Formula 3, where mind is dA i,k The minimum value in, min=min(dA i,k ).

[0063] dA i，k =||A i -A j ||(j=1,...i-1,i+1,...ak=1,2,...,a-1) Formula 3

[0064] Preferably, the second strategy set CD strategy represents the game participants changing their own running direction while keeping other states unchanged.

[0065] In this embodiment, when the UAV selects the CD strategy, L k-i is the line between obstacle k and player i, L k-i The direction is from k to i; θ k For L k-i Angle with the X axis; θ ave is the average value of all angles, as shown in Formula 4; Lave is θ ave The corresponding straight line; θ an is the direction of the UAV's movement L run The angle with the X axis; θ is L ave With L run The angle between the two is less than 180°. Calculate the corrected angle using Formula 5.

[0066]

[0067]

[0068] Preferably, the third strategy set NC strategy represents that the running speed and running direction of the game participants do not change at all, and the game players continue to run at the original speed and direction.

[0069] Preferably, the fourth strategy set, ST, represents a stop strategy, in which the player maintains their current position. When the obstacle is static, the ST strategy is selected by default, requiring the UAV to adjust its operating state to avoid the obstacle. When the obstacle is dynamic, the dynamic obstacle is treated as a UAV, requiring the two UAVs to engage in a game of obstacle avoidance.

[0070] In step 102, the quality of the game strategy selected by the UAV is measured by the game profit value, and the profit value is calculated by the profit function.

[0071] Preferably, assuming there are a total of a players, the profit calculation formula is shown in Formula 6.

[0072]

[0073] Among them, R is the coefficient for judging whether the strategy can ensure the safety of the UAV. When the selected strategy can ensure safety, R = 1, and when it cannot ensure safety, R = 0; d safe is the obstacle avoidance distance, i.e. the safety distance; aved is A i The average distance between other game participants; mind is A i The minimum distance between other game participants; A i A is the strategy adopted by the game participant (i.e., UAV) i, -i is the strategy adopted by the game participants other than participant i; X(A) is the consumption value of different strategies.

[0074] In step 103, in order to enable all game participants to reach Nash equilibrium, it is necessary to use an iterative game algorithm so that the game participants can play multiple games and eventually arrive at the optimal strategy.

[0075] Preferably, let the number of repeated games be T, the number of current games be t, and the number of optional strategies for the game participants be n. Initialize the game participant A i Mixed strategy. t=1 and game player A i The initial probability is as shown in Formula 7:

[0076]

[0077] in, n is the number of strategies available to the players; are the probabilities of the game participants choosing the 1st, 2nd, …, nth strategies at the initial moment.

[0078] Preferably, the mixed strategy is combined with the decision of each iterative player so that the player A i The probability of changes with the number of iterations, and the calculation formula is as follows:

[0079]

[0080] Among them, the parameters Function I(τ i )(i=1,2,…,n) is the decision record function, which mainly records the number of times the current strategy is selected and the number of times the current strategy is not selected in the previous t-1 iterations. The specific calculation is shown in Formula 9:

[0081]

[0082] Among them, PRi represents the game participant A i The game priority of ; β is a constant greater than 0, which is used to record the number of times the game participants choose the current or other strategies, and is generally taken as β = 0.5; when the game participants are static obstacles, PR is always 0.

[0083] Preferably, the initial priority of the game participants is PR=0, and then the priority is changed with reference to formula 10, d pr The distance for determining priority. When the player's priority is 1, the NC strategy is selected by default.

[0084]

[0085] Among them, L i For game participant A i The distance to the target point.

[0086] The information of each game participant includes: the participant's current position (i=1,2,…,a), running speed The angle θ between the running direction and the X axis, priority PR. The UAV autonomously senses the surrounding environment and the sensing distance dsafe Meanwhile, it is the safety range of the unmanned aerial vehicle.

[0087] In this embodiment, when the unmanned aerial vehicle encounters other unmanned aerial vehicles or obstacles during operation, it needs to conduct a game. After the game is completed, the unmanned aerial vehicle will retrieve the original path and continue to operate. The specific steps are as follows:

[0088] Step 1: The unmanned aerial vehicle runs along the path and determines whether it has reached the target point. If it has reached the target point, go to Step 8; otherwise, go to Step 2.

[0089] Step 2: The unmanned aerial vehicle senses the surrounding environment. If an obstacle is sensed, conduct a game for obstacle avoidance and go to Step 3; otherwise, go to Step 1.

[0090] Step 3: Let t = 1, initialize P(1) according to Formula 7, set the number of repeated games T, and the optional strategies S for the game (including n strategies);

[0091] Step 4: Establish the set of game participants A according to Formula 1, and there are a players in the set;

[0092] Step 5: Calculate u according to Formula 6, select the strategy corresponding to the maximum benefit based on P and u, record the currently selected strategy, and update I(τ1) simultaneously;

[0093] Step 6: Let t = t + 1. If t < T, update P(t) according to Formula 8 and return to Step 4; if t ≥ T, output the strategy selected in the last iteration;

[0094] Step 7: Complete the current game, determine whether the obstacle avoidance is successful. If successful, the unmanned aerial vehicle approaches the original path and goes to Step 1; otherwise, go to Step 3.

[0095] Step 8: The unmanned aerial vehicle safely reaches the target point, and the planning ends.

[0096] Embodiment 2

[0097] For reference Figure 2 The environment of is taken as an example for illustration, where the starting point is represented by "*" and the target point is represented by "★". This experiment involves three unmanned aerial vehicles and a static obstacle originally in the environment, and the information is shown in Table 1 below.

[0098] Table 1 Unmanned Aerial Vehicle Task Information

[0099]

[0100] This experiment generated two sets of game obstacle avoidance processes. The first set includes three obstacle avoidance scenarios. As Figure 3As shown in Figure (c), UAV-1 and UAV-2 are traveling along a pre-designed path. When they meet, the first game decision-making process occurs. This game involves two players, and the specific obstacle avoidance information is shown in Table 2. UAV-1 and UAV-2 engage in three rounds of game play, ultimately successfully avoiding the obstacle. The two players return to their original paths.

[0101] Table 2 Obstacle avoidance strategy information

[0102]

[0103] The second group includes three obstacle avoidances, where the obstacle avoidance environment is as follows Figure 3 shown.

[0104] (1) Figure 3 As shown in (b), UAV-2 and UAV-3 meet and both sense an obstacle. This game involves three players: UAV-2, UAV-3, and obstacle-1. Obstacle-1 defaults to a strategy of staying still, while UAV-2 and UAV-3 choose a strategy of changing direction. UAV-2 and UAV-3 then return to their original path.

[0105] (2) Figure 3 (c) They meet again. This time, there are three players in the game. Obstacle-1 chooses the default strategy of staying still, UAV-2 chooses to stay still, and UAV-3 chooses the strategy of changing its direction of movement. UAV-3 returns to its original path.

[0106] (3) UAV-2 started the third game with Obstacle-1 because it chose to stay still in the last game. There were two players in this game. Obstacle-1 chose the strategy of staying still by default, while UAV-2 chose the strategy of changing the direction of movement. UAV-2 successfully avoided the obstacle and moved towards the target point. This group of games was completed.

[0107] Table 3 Obstacle avoidance strategy information

[0108]

[0109] The simulation results above demonstrate that, in any environment, the dynamic obstacle avoidance algorithm based on iterative game theory enables the UAV to quickly and effectively avoid other game participants. Therefore, this obstacle avoidance algorithm can effectively resolve the collision problem of the UAV system and ensure that the UAV safely reaches its target.

[0110] The present invention proposes a method for avoiding dynamic obstacles for unmanned aerial vehicles based on repeated games. When an unmanned aerial vehicle, as a participant in the game, faces a dynamic obstacle and selects a game strategy, there may be multiple equilibrium states between the game participants. The present invention addresses the problem of real-time dynamic obstacle avoidance for unmanned aerial vehicles and proposes a method for avoiding dynamic obstacles for unmanned aerial vehicles based on repeated games. Its purpose is to provide an effective method for collaborative obstacle avoidance for multiple unmanned aerial vehicles. First, a library of dynamic obstacle avoidance strategies for unmanned aerial vehicles is established, and then the game payoff is calculated based on the states between the game participants. In order to enable the game participants to quickly converge to the optimal strategy during the repeated game process, the mixed strategy is updated according to the decisions of the game participants, and decisions are made based on the mixed strategy and the game payoff, so that the payoffs of the players reach a Nash equilibrium and the corresponding obstacle avoidance strategy is selected. This method enables unmanned aerial vehicles to avoid obstacles quickly and safely when facing various obstacles.

[0111] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a program running on the processor, and when the processor runs the program, the steps of the above-mentioned unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game are executed.

[0112] The present invention also provides a computer-readable storage medium having computer instructions stored thereon. When the computer instructions are executed, the steps of the above-mentioned unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game are executed. The above-mentioned unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game is described in the introduction of the previous part and will not be repeated here.

[0113] Those skilled in the art will understand that the foregoing descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will be able to modify the technical solutions described in the foregoing embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A dynamic obstacle avoidance method for an unmanned aerial vehicle based on repeated game, comprising the following steps: Take the UAV as a participant in the game and establish a set of game participants; Calculating a game gain value based on the status between the game participants, and measuring the pros and cons of the game strategy selected by the unmanned aerial vehicle by calculating the game gain value; The game participants are made to quickly converge to the optimal strategy during repeated game play. The mixed strategy is updated according to the decisions of the game participants. The decisions are made based on the mixed strategy and the game payoffs, so that the payoffs of the game participants reach the Nash equilibrium and the corresponding obstacle avoidance strategy is selected, including: Assume the number of repeated games is T , the current number of games is t , the game participants have n optional strategies, and the initialization game participant A i Mixed strategy, t=1 and game player A i The initial probability is: in, , n The number of strategies that can be chosen by game participants; The first, second, ..., and third players are selected at the initial moment for the game participants. n The probability of a strategy.

2. The unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game according to claim 1, characterized in that: The step of using the UAV as a participant in the game and establishing a set of game participants also includes: Assume that the UAV encounters H obstacles OB during operation h , h=1,…,H, the current position coordinates of the UAV are , the coordinates of the closest point of each obstacle to the UAV are , the set of game participants is A, put the unmanned aerial vehicle into the set of game participants A; Calculate the distance between the UAV and the obstacle. When Formula 1 is satisfied, the obstacle enters the UAV's safety range and plays a game of obstacle avoidance with the UAV. Among them, d safe is the safe distance. When the distance between the UAV and the obstacle is less than or equal to d safe When the obstacle OB h Add the set of game participants A and perform obstacle avoidance maneuvers; When performing obstacle avoidance, the original obstacles that appear in the working environment within the safety range of the UAV are regarded as game participants.

3. The unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game according to claim 2, characterized in that: Also includes, The strategy set chosen by the game participants consists of four strategy sets; The first strategy set is for the game participants to change their own speed while keeping other states unchanged; The second strategy set is for the game participants to change their own running direction while keeping other states unchanged; The third strategy set is that the running speed and direction of the game participants remain unchanged; The fourth strategy set is a stop strategy, where the game participants keep their current positions unchanged.

4. The unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game according to claim 1, characterized in that: The step of calculating the game benefit value according to the status between the game participants and measuring the pros and cons of the game strategy selected by the unmanned aerial vehicle by calculating the game benefit value also includes: Assume a total a The profit calculation formula for game participants is: in, R To determine whether the strategy can ensure the safety of the UAV, when the selected strategy can ensure safety R =1, when safety cannot be guaranteed R =0; d safe is the safe distance range of the UAV; ave d A i The average distance between other game participants; min d A i Minimum distance to other game participants; A i For game participants i the strategies adopted; A -i For all but the game participants i The strategies adopted by the game participants other than X(A i ) is the consumption value of different strategies.

5. The unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game according to claim 1, characterized in that: Also includes, Combine the mixed strategy with the decision of each iterative game player, so that game player A i The probability of changes with the number of iterations, and the calculation formula is: Among them, the parameters ;function , i=1,2,…,n, is the decision record function, including recording the number of times the current strategy is selected and the number of times the current strategy is not selected in the previous t-1 iterations. The calculation formula is: Among them, PR i For game participant A i Game priority; β It is a constant greater than 0, used to record the number of times the game participant chooses the current or other strategy; when the game participant is a static obstacle, PR is 0.

6. The unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game according to claim 5, characterized in that: Also includes, The initial priority PR of the game participants is 0, and the priority is changed according to the priority calculation formula, which is: Among them, L i For game participant A i The distance to the target point, d pr To determine the priority distance, when the priority of the game participant is 1, the third strategy set is selected.

7. The unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game according to claim 6, characterized in that: The information of each of the game participants includes the participant's current location , i=1,2,…,a, running speed , the angle θ between the running direction and the X-axis, priority PR.

8. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a program running on the processor, and when the processor runs the program, the steps of the unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game as described in any one of claims 1 to 7 are executed.

9. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed, the steps of the unmanned aerial vehicle dynamic obstacle avoidance method based on repeated game described in any one of claims 1 to 7 are executed.

Images