An adaptive control method and system for virtual enemies

Abstract

This invention discloses an adaptive control method and system for virtual enemies. It collects historical action data of players, classifies players based on this data to determine player types. If the player type is a ranged attack type, a first response strategy corresponding to the ranged attack type is used to control the virtual enemy. If the player type is a defensive type, a second response strategy corresponding to the defensive type is used to control the virtual enemy. This dynamically adjusts the response strategy for the virtual enemy based on the player's actions, ensuring the virtual enemy remains challenging for the player. This encourages players to continuously adjust their actions, increasing the game's replayability and effectively enhancing its fun and challenge.

Description

Technical Field

[0001] This invention relates to the field of virtual enemy control, and more particularly to an adaptive control method and system for virtual enemies. Background Technology

[0002] In the field of game design, there exists an enemy mechanic used to simulate realistic enemy behavior, providing players with opportunities for practice and challenge. However, traditional, fixed-pattern virtual enemies are easily familiar to players, who can easily master countermeasures, thus reducing the game's difficulty and appeal. Therefore, how to make virtual enemies more realistic and flexible in their interaction with players to enhance the game's fun and challenge has become one of the issues that game developers need to consider. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide an adaptive control method and system for virtual enemies, which can effectively enhance the fun and challenge of the game.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: An adaptive control method for virtual enemies, comprising the following steps: Collect players' historical action data; The players are categorized based on their historical operational behavior data to determine their player types; If the type is a remote attack type, then the first response strategy corresponding to the remote attack type is used to control the virtual enemy; If the type is a defensive type, then a second response strategy corresponding to the defensive type is used to control the virtual enemy.

[0005] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows: An adaptive control system for a virtual enemy includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: Collect players' historical action data; The players are categorized based on their historical operational behavior data to determine their player types; If the type is a remote attack type, then the first response strategy corresponding to the remote attack type is used to control the virtual enemy; If the type is a defensive type, then a second response strategy corresponding to the defensive type is used to control the virtual enemy.

[0006] The beneficial effects of this invention are as follows: By collecting players' historical operation behavior data and classifying players based on this data, the player's type is determined. If the type is a ranged attack type, a first response strategy corresponding to the ranged attack type is used to control the virtual enemy. If the type is a defensive type, a second response strategy corresponding to the defensive type is used to control the virtual enemy. This dynamically adjusts the response strategy to the virtual enemy based on the player's operation behavior, ensuring that the virtual enemy is always challenging for the player. This prompts players to continuously adjust their operation methods, increasing the replayability of the game and effectively enhancing its fun and challenge. Attached Figure Description

[0007] Figure 1 This is a flowchart illustrating the steps of an adaptive control method for a virtual enemy according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of an adaptive control system for a virtual enemy according to an embodiment of the present invention. Detailed Implementation

[0008] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0009] Please refer to Figure 1 An adaptive control method for virtual enemies, comprising the following steps: Collect players' historical action data; The players are categorized based on their historical operational behavior data to determine their player types; If the type is a remote attack type, then the first response strategy corresponding to the remote attack type is used to control the virtual enemy; If the type is a defensive type, then a second response strategy corresponding to the defensive type is used to control the virtual enemy.

[0010] As can be seen from the above description, the beneficial effects of the present invention are as follows: Players' historical operational behavior data is collected, and players are classified based on this data to obtain player types. If the type is a ranged attack type, a first response strategy corresponding to the ranged attack type is used to control the virtual enemy; if the type is a defensive type, a second response strategy corresponding to the defensive type is used to control the virtual enemy. This dynamically adjusts the response strategy to the virtual enemy based on the player's operational behavior, ensuring that the virtual enemy remains challenging for the player, prompting the player to continuously adjust their operational methods, increasing the game's replayability, and thus effectively enhancing the game's fun and challenge.

[0011] Furthermore, the historical operation behavior data includes key press information, mouse click frequency, and mouse click location; The player classification based on the historical operation behavior data, resulting in player types, includes: By analyzing the key press information, mouse click frequency, and mouse click location, the player's operating habits and skill usage frequency can be obtained. The player's behavioral feature vector is determined based on the operating habits and the frequency of skill usage. The player is classified using a classification algorithm based on the behavioral feature vector to obtain the player's type.

[0012] As described above, historical operation behavior data includes key press information, mouse click frequency, and mouse click location. By analyzing this operation behavior data, the player's behavior feature vector can be determined. Based on the behavior feature vector, a classification algorithm can be used to classify the player, which can more accurately and effectively determine the player type. This allows for the adoption of different strategies to control virtual enemies for different types of players.

[0013] Furthermore, controlling the virtual enemy using a first response strategy corresponding to the type of remote attack includes: Obtain the player's location and predicted attack range, as well as information on other obstacles and cover; Based on the player's location and predicted attack range, the other obstacle information, and the cover information, a path planning algorithm is used to generate the optimal path for the virtual enemy to reach the nearest cover. Control the virtual enemy according to the optimal path; Acquire terrain information, distance information between the player and the virtual enemy, and the player's field of vision; The optimal timing and route for close-range assault are calculated based on the terrain information, distance information, and field of view. Control the virtual enemy to launch a surprise attack on the player according to the optimal close-range attack timing and optimal close-range attack path.

[0014] As described above, for players using ranged attacks, a path planning algorithm is used to generate the optimal path for virtual enemies to reach the nearest cover based on the player's position, predicted attack range, other obstacle information, and cover information. Furthermore, the optimal timing and path for close-range attacks are calculated based on terrain information, distance information, and field of vision. This allows virtual enemies to launch surprise attacks on players while effectively avoiding their attacks, thereby increasing the realism and challenge of the game.

[0015] Furthermore, controlling the virtual enemy using a second response strategy corresponding to the defensive type includes: The attack interval of the virtual enemy is shortened to obtain the shortened attack interval. Determine the most powerful weapon and skill combination pattern; The virtual enemy is controlled to randomly attack the player from different locations in order to obtain the player's defensive reaction information and resource allocation information, and the player's defensive vulnerabilities are determined based on the defensive reaction information and resource allocation information. Based on the aforementioned defensive vulnerabilities, the virtual enemy is controlled to attack the player using the most powerful weapon and skill combination mode according to the shortened attack interval.

[0016] As described above, when facing players who excel at defense, the game utilizes defensive vulnerabilities and controls virtual enemies to attack players using the most powerful weapons and skill combinations at shortened attack intervals. This increases the frequency and intensity of attacks on these players, making the game more challenging and enhancing its immersion and fun.

[0017] Furthermore, the method of controlling the virtual enemy using a first response strategy corresponding to the type of remote attack also includes: The player's attack methods and locations are monitored, and monitoring results are obtained. Based on the monitoring results, determine whether the player's attack method has changed to melee attack. If so, control the virtual enemy to perform melee defense. Based on the monitoring results, determine whether the player's position is gradually approaching the virtual enemy. If so, control the virtual enemy to counterattack.

[0018] As described above, while controlling virtual enemies to deal with players who use ranged attacks, the game also monitors the players' attack methods and positions. When the players switch to melee attacks, the game controls the virtual enemies to perform melee defense. As the players gradually approach the virtual enemies, the game controls the virtual enemies to counterattack. This allows the virtual enemies to interact with the players' actions in real time, improving the flexibility of virtual enemy control and enhancing the realism of the game.

[0019] Please refer to Figure 2 An adaptive control system for a virtual enemy includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: Collect players' historical action data; The players are categorized based on their historical operational behavior data to determine their player types; If the type is a remote attack type, then the first response strategy corresponding to the remote attack type is used to control the virtual enemy; If the type is a defensive type, then a second response strategy corresponding to the defensive type is used to control the virtual enemy.

[0020] As can be seen from the above description, the beneficial effects of the present invention are as follows: Players' historical operational behavior data is collected, and players are classified based on this data to obtain player types. If the type is a ranged attack type, a first response strategy corresponding to the ranged attack type is used to control the virtual enemy; if the type is a defensive type, a second response strategy corresponding to the defensive type is used to control the virtual enemy. This dynamically adjusts the response strategy to the virtual enemy based on the player's operational behavior, ensuring that the virtual enemy remains challenging for the player, prompting the player to continuously adjust their operational methods, increasing the game's replayability, and thus effectively enhancing the game's fun and challenge.

[0021] Furthermore, the historical operation behavior data includes key press information, mouse click frequency, and mouse click location; The player classification based on the historical operation behavior data, resulting in player types, includes: By analyzing the key press information, mouse click frequency, and mouse click location, the player's operating habits and skill usage frequency can be obtained. The player's behavioral feature vector is determined based on the operating habits and the frequency of skill usage. The player is classified using a classification algorithm based on the behavioral feature vector to obtain the player's type.

[0022] As described above, historical operation behavior data includes key press information, mouse click frequency, and mouse click location. By analyzing this operation behavior data, the player's behavior feature vector can be determined. Based on the behavior feature vector, a classification algorithm can be used to classify the player, which can more accurately and effectively determine the player type. This allows for the adoption of different strategies to control virtual enemies for different types of players.

[0023] Furthermore, controlling the virtual enemy using a first response strategy corresponding to the type of remote attack includes: Obtain the player's location and predicted attack range, as well as information on other obstacles and cover; Based on the player's location and predicted attack range, the other obstacle information, and the cover information, a path planning algorithm is used to generate the optimal path for the virtual enemy to reach the nearest cover. Control the virtual enemy according to the optimal path; Acquire terrain information, distance information between the player and the virtual enemy, and the player's field of vision; The optimal timing and route for close-range assault are calculated based on the terrain information, distance information, and field of view. Control the virtual enemy to launch a surprise attack on the player according to the optimal close-range attack timing and optimal close-range attack path.

[0024] As described above, for players using ranged attacks, a path planning algorithm is used to generate the optimal path for virtual enemies to reach the nearest cover based on the player's position, predicted attack range, other obstacle information, and cover information. Furthermore, the optimal timing and path for close-range attacks are calculated based on terrain information, distance information, and field of vision. This allows virtual enemies to launch surprise attacks on players while effectively avoiding their attacks, thereby increasing the realism and challenge of the game.

[0025] Furthermore, controlling the virtual enemy using a second response strategy corresponding to the defensive type includes: The attack interval of the virtual enemy is shortened to obtain the shortened attack interval. Determine the most powerful weapon and skill combination pattern; The virtual enemy is controlled to randomly attack the player from different locations in order to obtain the player's defensive reaction information and resource allocation information, and the player's defensive vulnerabilities are determined based on the defensive reaction information and resource allocation information. Based on the aforementioned defensive vulnerabilities, the virtual enemy is controlled to attack the player using the most powerful weapon and skill combination mode according to the shortened attack interval.

[0026] As described above, when facing players who excel at defense, the game utilizes defensive vulnerabilities and controls virtual enemies to attack players using the most powerful weapons and skill combinations at shortened attack intervals. This increases the frequency and intensity of attacks on these players, making the game more challenging and enhancing its immersion and fun.

[0027] Furthermore, the method of controlling the virtual enemy using a first response strategy corresponding to the type of remote attack also includes: The player's attack methods and locations are monitored, and monitoring results are obtained. Based on the monitoring results, determine whether the player's attack method has changed to melee attack. If so, control the virtual enemy to perform melee defense. Based on the monitoring results, determine whether the player's position is gradually approaching the virtual enemy. If so, control the virtual enemy to counterattack.

[0028] As described above, while controlling virtual enemies to deal with players who use ranged attacks, the game also monitors the players' attack methods and positions. When the players switch to melee attacks, the game controls the virtual enemies to perform melee defense. As the players gradually approach the virtual enemies, the game controls the virtual enemies to counterattack. This allows the virtual enemies to interact with the players' actions in real time, improving the flexibility of virtual enemy control and enhancing the realism of the game.

[0029] The adaptive control method and system for virtual enemies described above are applicable to game scenarios, and are explained below through specific embodiments: Please refer to Figure 1 Embodiment 1 of the present invention is as follows: An adaptive control method for virtual enemies, comprising the following steps: S1. Collect players' historical action data.

[0030] In one optional implementation, the historical operation behavior data includes key press information, mouse click frequency, and mouse click location. The key press information corresponds to the release of different player skills, and the mouse click frequency and mouse click location can be used to determine the player's attack direction and method, such as aiming operations for ranged attacks. By collecting the player's historical operation behavior data, it can serve as a basis for judging the player's operating habits.

[0031] S2. Classify the players based on the historical operation behavior data to obtain the player types, specifically including S21-S23: S21. Analyze the key press information, mouse click frequency, and mouse click position to obtain the player's operating habits and skill usage frequency.

[0032] For example, in each game, the proportion of ranged attack skills used out of the total number of skill uses is determined based on key press information, mouse click frequency, and mouse click location. This helps determine the player's skill preference and usage frequency. Analyzing player strategy choices in the game, such as whether to launch an offensive, a defensive counter-attack, or employ flanking tactics in different scenarios, involves monitoring the player's character movement path, distance from enemies, and resource acquisition and utilization based on key press information, mouse click frequency, and mouse click location to infer the player's strategy, i.e., their operating habits.

[0033] S22. Determine the player's behavioral feature vector based on the operating habits and the frequency of skill usage.

[0034] For example, the behavioral feature vector of a player who frequently uses ranged attacks and prefers a specific skill combination can be defined as [1,0,0.8,0.2,...]. The values ​​at different positions represent different behavioral feature weights. For example, the first position indicates whether ranged attacks are mainly used (1 indicates yes, 0 indicates no), the second position indicates whether melee attacks are mainly used, and the subsequent values ​​indicate the usage frequency weights of different skills.

[0035] S23. Classify the player using a classification algorithm based on the behavioral feature vector to obtain the player's type.

[0036] The classification algorithms include decision trees, neural networks, etc. By classifying players based on behavioral feature vectors using classification algorithms, a better understanding of players' game style patterns can be achieved, enabling the development of corresponding virtual enemy strategies for different types of players.

[0037] S3. If the type is a remote attack type, then the first response strategy corresponding to the remote attack type is used to control the virtual enemy, specifically including S31-S36: S31. If the type is a remote attack type, then obtain the player's location and predicted attack range, other obstacle information and cover information.

[0038] The cover information includes the cover name and location; the cover can be a building, a large rock, or other object. The other obstacle information includes the locations of other obstacles.

[0039] S32. Based on the player's position and predicted attack range, the other obstacle information and the cover information, a path planning algorithm is used to generate the optimal path for the virtual enemy to reach the nearest cover.

[0040] The optimal path is the shortest path from the current location of the virtual enemy to the nearest cover, while avoiding the player's possible attack range and other obstacles.

[0041] S33. Control the virtual enemy according to the optimal path.

[0042] In one alternative implementation, after executing S33, the following may also be included: The system acquires a preset pose and controls the virtual enemy according to that pose. For example, when controlling the virtual enemy to hide behind a building, it will press its body close to the wall, exposing only a small portion of itself to the player, while simultaneously adjusting the camera angle to observe the player's movements. This ensures that the risk of being attacked by the player from a distance is minimized.

[0043] S34. Obtain terrain information, distance information between the player and the virtual enemy, and the player's field of vision.

[0044] S35. Calculate the optimal close-range assault timing and optimal close-range assault path based on the terrain information, the distance information, and the field of view.

[0045] The optimal time for a close-range ambush is during a short pause after an attack, during reload time, or in a blind spot. The optimal path for a close-range ambush is an irregular movement pattern, such as zigzag movement.

[0046] S36. Control the virtual enemy to launch a surprise attack on the player according to the optimal close-range attack timing and optimal close-range attack path. In this way, control the virtual enemy to take advantage of the player's short pause after attacking, reload time, or blind spots to quickly approach the player. During the approach, control the virtual enemy to use irregular movement methods, such as zigzag movement, to avoid the player's ranged attacks and increase the probability of successfully closing in.

[0047] Specifically, through visual and audio cues, the player controls virtual enemies to launch surprise attacks based on the optimal timing and path for close-range assaults, allowing the player to perceive the dynamic changes of the virtual enemies. For example, visual effects such as changes in the virtual enemy's behavior (from seeking cover to launching an attack), weapon switching displays, and audio cues such as different sounds emitted by the enemy when adopting different strategies (increased footsteps when approaching, special sound effects when launching powerful attacks), clearly inform the player that the virtual enemy is reacting to their actions, thereby increasing the game's tension and enjoyment.

[0048] During a surprise attack on the player, players can interact with the terrain and objects in the game environment, such as jumping over obstacles and bypassing traps. When they get close enough to the player, a melee attack is triggered, initiating close-quarters combat.

[0049] When players frequently use ranged attacks, virtual enemies can seek cover to reduce their exposure to the player's ranged attack range in order to effectively counter the player's advantage. For example, in a shooting game, virtual enemies can quickly hide behind buildings or obstacles to avoid being directly hit by the player's ranged weapons. Simultaneously, virtual enemies can also employ close-range ambush strategies, exploiting the limitations of the player's ranged weapons in close combat to quickly close the distance and attempt to engage the player in close-range combat. This breaks the player's ranged attack advantage, increases their survivability and effectiveness in combat, and thus enhances the game's challenge for the player.

[0050] In one alternative implementation, S3 is performed concurrently with: The player's attack methods and locations are monitored, and monitoring results are obtained. Based on the monitoring results, determine whether the player's attack method has changed to melee attack. If so, control the virtual enemy to perform melee defense. Based on the monitoring results, determine whether the player's position is gradually approaching the virtual enemy. If so, control the virtual enemy to counterattack.

[0051] In this way, the virtual enemies can interact with the player's behavior in real time. For example, when the player changes their attack method or position, the virtual enemy's strategy can be adjusted accordingly. If the player switches from ranged attack to melee attack, the virtual enemy can immediately abandon cover and adopt a melee defense or counterattack strategy.

[0052] S4. If the type is a defensive type, then a second response strategy corresponding to the defensive type is used to control the virtual enemy, specifically including S41-S44: S41. If the type is a defensive type, then the attack interval of the virtual enemy is shortened to obtain a shortened attack interval.

[0053] The specific time reduction can be flexibly set according to the actual situation. For example, if the original attack interval was once every 5 seconds, it can now be shortened to once every 3 seconds.

[0054] S42. Determine the weapon with the strongest attack power and the skill combination mode.

[0055] For example, in strategy games, basic attack skills that were originally used alone may now be used in combination with skills that have armor-breaking or area-of-effect attacks to enhance attack power.

[0056] S43. Control the virtual enemy to randomly attack the player from different locations to obtain the player's defensive reaction information and resource allocation information, and determine the player's defensive vulnerabilities based on the defensive reaction information and resource allocation information. This method combines probing attacks and data analysis to gradually identify vulnerabilities in the player's defensive system and formulate targeted attack strategies.

[0057] S44. Based on the aforementioned defensive vulnerability, control the virtual enemy to attack the player using the most powerful weapon and skill combination mode according to the shortened attack interval.

[0058] Specifically, based on the defensive vulnerabilities, the virtual enemy is controlled to attack the player using the most powerful weapon and skill combination mode according to the shortened attack interval, through visual effects or sound cues.

[0059] If a player demonstrates strong defensive capabilities in the game, the virtual enemy will adjust accordingly using its intelligent algorithms. Specifically, the virtual enemy will increase the frequency and intensity of its attacks. By increasing the number of attacks, the virtual enemy aims to put more pressure on the player's defensive system, forcing the player to create defensive vulnerabilities under the constant barrage. Increased attack intensity may manifest as the use of more powerful weapons or more aggressive skill combinations. For example, in a strategy game, the virtual enemy might increase its troop deployment, accelerate the pace of its attack, or employ more destructive attack methods, such as concentrating firepower on the player's weak points to find and break through the player's defensive lines, thereby gaining an advantage in the game and increasing the difficulty of human-computer interaction.

[0060] The player's actions will continue to influence the next decision made by controlling the virtual enemy, forming a dynamic interactive loop that continuously drives the game forward and makes it highly replayable.

[0061] In one alternative implementation, the method may further include: collecting social interaction data of players in the game, such as the number of times they cooperate with other players and the frequency of communication. This data can reflect players' social skills and teamwork tendencies in the game, providing a more comprehensive basis for difficulty adjustment. It may also include analyzing players' performance data in different game modes, as players may have unique strengths or weaknesses in certain specific modes. A comprehensive analysis of performance data can provide a more accurate understanding of players' gaming abilities and growth trends.

[0062] In one alternative implementation, the game may further include: categorizing players into different developmental stages, such as beginner, intermediate, and advanced levels, based on their comprehensive data, and developing more detailed difficulty adjustment strategies for each stage. For example, for intermediate players, in addition to appropriately increasing the intelligence and number of virtual enemies, some levels with special rules or mechanisms can be introduced, requiring players to use specific strategies to pass, further enhancing the game's challenge. For advanced players, game elements with randomness and unpredictability can be designed, such as randomly generated maps and random enemy skill combinations, ensuring players constantly face new challenges and preventing the game from becoming monotonous.

[0063] In one optional implementation, the system may further include: real-time tracking of various performance metrics of players in the game, such as time to complete tasks, number of errors, and success rate in dealing with different enemies. Dynamic analysis of these performance metrics allows for timely detection of new players' skill improvement and game adaptation. When a new player is detected as adapting to the game, a guidance system is activated. This system provides targeted game skill tips and strategic suggestions based on the new player's current skill level, helping them better cope with the gradually increasing difficulty of the game environment. After new players have adapted to the game, a reasonable difficulty curve is designed to ensure a smooth transition in difficulty, avoiding sudden increases that could lead to frustration. For example, it could begin by fine-tuning the attributes and behaviors of virtual enemies. As players' abilities further improve, new types of virtual enemies and level mechanics can be gradually introduced, but the increase in difficulty should be moderate each time, allowing players to gradually adapt to new challenges.

[0064] Please refer to Figure 2 Embodiment two of the present invention is as follows: An adaptive control system for a virtual enemy includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the adaptive control method for the virtual enemy in Embodiment 1.

[0065] In summary, this invention provides an adaptive control method and system for virtual enemies. It collects historical player action data, categorizes players based on this data to determine player types. If the player type is a ranged attack type, a first response strategy corresponding to this type is used to control the virtual enemy. If the player type is a defensive type, a second response strategy corresponding to this type is used. This dynamically adjusts the response strategy based on the player's actions, ensuring the virtual enemy remains challenging and prompting players to continuously adjust their actions, increasing the game's replayability and effectively enhancing its fun and challenge. Furthermore, the historical action data includes key press signals... By analyzing player behavior data such as information, mouse click frequency, and mouse click location, behavioral feature vectors are determined. Classification algorithms are then used to categorize players based on these feature vectors, enabling more accurate and effective identification of player types. This allows for the application of different strategies to control virtual enemies based on different player types. Furthermore, while controlling virtual enemies to counter ranged attacks, the game also monitors player attack methods and positions. When a player switches to melee attacks, the virtual enemy is controlled to perform melee defense; as the player approaches the virtual enemy, the virtual enemy is controlled to counterattack. This allows for real-time interaction between the virtual enemy and the player's actions, increasing the flexibility of virtual enemy control and enhancing the game's realism.

[0066] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. An adaptive control method for virtual enemies, characterized in that, Including the following steps: Collect players' historical action data; The players are categorized based on their historical operational behavior data to determine their player types; If the type is a remote attack type, then the first response strategy corresponding to the remote attack type is used to control the virtual enemy; If the type is a defensive type, then a second response strategy corresponding to the defensive type is used to control the virtual enemy.

2. The adaptive control method for a virtual enemy according to claim 1, characterized in that, The historical operation data includes key press information, mouse click frequency, and mouse click location; The player classification based on the historical operation behavior data, resulting in player types, includes: By analyzing the key press information, mouse click frequency, and mouse click location, the player's operating habits and skill usage frequency can be obtained. The player's behavioral feature vector is determined based on the operating habits and the frequency of skill usage. The player is classified using a classification algorithm based on the behavioral feature vector to obtain the player's type.

3. The adaptive control method for a virtual enemy according to claim 1, characterized in that, The control of the virtual enemy using a first response strategy corresponding to the type of remote attack includes: Obtain the player's location and predicted attack range, as well as information on other obstacles and cover; Based on the player's location and predicted attack range, the other obstacle information, and the cover information, a path planning algorithm is used to generate the optimal path for the virtual enemy to reach the nearest cover. Control the virtual enemy according to the optimal path; Acquire terrain information, distance information between the player and the virtual enemy, and the player's field of vision; The optimal timing and route for close-range assault are calculated based on the terrain information, distance information, and field of view. Control the virtual enemy to launch a surprise attack on the player according to the optimal close-range attack timing and optimal close-range attack path.

4. The adaptive control method for a virtual enemy according to claim 1, characterized in that, The control of the virtual enemy using a second response strategy corresponding to the defensive type includes: The attack interval of the virtual enemy is shortened to obtain the shortened attack interval. Determine the most powerful weapon and skill combination pattern; The virtual enemy is controlled to randomly attack the player from different locations in order to obtain the player's defensive reaction information and resource allocation information, and the player's defensive vulnerabilities are determined based on the defensive reaction information and resource allocation information. Based on the aforementioned defensive vulnerabilities, the virtual enemy is controlled to attack the player using the most powerful weapon and skill combination mode according to the shortened attack interval.

5. The adaptive control method for a virtual enemy according to claim 1, characterized in that, The method of controlling the virtual enemy using a first response strategy corresponding to the type of remote attack also includes: The player's attack methods and locations are monitored, and monitoring results are obtained. Based on the monitoring results, determine whether the player's attack method has changed to melee attack. If so, control the virtual enemy to perform melee defense. Based on the monitoring results, determine whether the player's position is gradually approaching the virtual enemy. If so, control the virtual enemy to counterattack.

6. An adaptive control system for a virtual enemy, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it performs the following steps: Collect players' historical action data; The players are categorized based on their historical operational behavior data to determine their player types; If the type is a remote attack type, then the first response strategy corresponding to the remote attack type is used to control the virtual enemy; If the type is a defensive type, then a second response strategy corresponding to the defensive type is used to control the virtual enemy.

7. The adaptive control system for a virtual enemy according to claim 6, characterized in that, The historical operation data includes key press information, mouse click frequency, and mouse click location; The player classification based on the historical operation behavior data, resulting in player types, includes: By analyzing the key press information, mouse click frequency, and mouse click location, the player's operating habits and skill usage frequency can be obtained. The player's behavioral feature vector is determined based on the operating habits and the frequency of skill usage. The player is classified using a classification algorithm based on the behavioral feature vector to obtain the player's type.

8. The adaptive control system for a virtual enemy according to claim 6, characterized in that, The control of the virtual enemy using a first response strategy corresponding to the type of remote attack includes: Obtain the player's location and predicted attack range, as well as information on other obstacles and cover; Based on the player's location and predicted attack range, the other obstacle information, and the cover information, a path planning algorithm is used to generate the optimal path for the virtual enemy to reach the nearest cover. Control the virtual enemy according to the optimal path; Acquire terrain information, distance information between the player and the virtual enemy, and the player's field of vision; The optimal timing and route for close-range assault are calculated based on the terrain information, distance information, and field of view. Control the virtual enemy to launch a surprise attack on the player according to the optimal close-range attack timing and optimal close-range attack path.

9. The adaptive control system for a virtual enemy according to claim 6, characterized in that, The control of the virtual enemy using a second response strategy corresponding to the defensive type includes: The attack interval of the virtual enemy is shortened to obtain the shortened attack interval. Determine the most powerful weapon and skill combination pattern; The virtual enemy is controlled to randomly attack the player from different locations in order to obtain the player's defensive reaction information and resource allocation information, and the player's defensive vulnerabilities are determined based on the defensive reaction information and resource allocation information. Based on the aforementioned defensive vulnerabilities, the virtual enemy is controlled to attack the player using the most powerful weapon and skill combination mode according to the shortened attack interval.

10. An adaptive control system for a virtual enemy according to claim 6, characterized in that, The method of controlling the virtual enemy using a first response strategy corresponding to the type of remote attack also includes: The player's attack methods and locations are monitored, and monitoring results are obtained. Based on the monitoring results, determine whether the player's attack method has changed to melee attack. If so, control the virtual enemy to perform melee defense. Based on the monitoring results, determine whether the player's position is gradually approaching the virtual enemy. If so, control the virtual enemy to counterattack.

Images