A game interaction system combined with limb movement sensing

Abstract

The present invention relates to a game interaction system combined with limb motion sensing. The system includes an environmental perception module that collects three-dimensional space data in real time to construct a dynamic environment model; a motion capture module that obtains limb motion data through multi-modal sensors and stores historical motion patterns; a game logic analysis module that analyzes the required motion sequence, target movement path, and motion requirement parameters; a motion prediction module that generates a predicted motion sequence and spatial coordinate trajectory; a collision risk assessment module that calculates the contact probability and danger level between the predicted motion and obstacles; and a dynamic guidance generation module that generates motion correction instructions based on the assessment results. The system provides a safe and smooth interaction experience for users in a complex dynamic environment through precise collision detection and intelligent motion adjustment, which not only meets the game goals but also effectively avoids collisions, significantly enhancing the immersion and operation safety, and is applicable to various limb motion sensing game scenarios.

Description

Technical Field

[0001] The present invention relates to the field of somatosensory game interaction, and specifically to a game interaction system combined with limb movement sensing. Background Art

[0002] In recent years, somatosensory game interaction technology has developed rapidly. By integrating motion capture, environmental perception, and virtual reality, it provides users with an immersive experience. Existing technologies generally use devices such as inertial sensors and depth cameras to achieve limb movement recognition, and combine environmental modeling technology to avoid physical obstacles. With the rise of the concepts of the metaverse and digital twin, higher requirements are put forward for the real-time performance, environmental adaptability, and safety protection capabilities of the system. The industry trend focuses on multi-modal data fusion, dynamic obstacle prediction, and personalized action guidance. However, there are still technical bottlenecks in accurate obstacle avoidance and action planning in complex scenarios for existing solutions:

[0003] Some systems attempt to combine game logic to analyze the action sequence that the user needs to execute, but they have insufficient ability to fuse the real-time position and movement state of obstacles in a dynamic environment, and do not fully consider the impact of the user's historical action patterns on future action prediction.

[0004] In a complex dynamic environment, it is impossible to accurately and real-time fuse static and dynamic obstacle information to build a comprehensive environmental model; the prediction of the user's future actions does not effectively combine historical action characteristics and game requirement parameters, resulting in poor adaptability of the predicted action sequence to the actual environment; the collision risk assessment only stays at a simple spatial position comparison, and a quantitative assessment model based on contact probability is not established, making it difficult to accurately identify high-risk actions; there is a lack of an intelligent action correction mechanism, and it is impossible to dynamically adjust the user's actions to avoid obstacles while ensuring the game goal, which may cause real collisions during the game and affect the experience safety and operation fluency. Summary of the Invention

[0005] The present invention aims at the technical problems existing in the prior art and provides a game interaction system combined with limb movement sensing.

[0006] The technical solution for the present invention to solve the above technical problems is as follows: A game interaction system combined with limb movement sensing, the system includes:

[0007] An environmental perception module, used to collect three-dimensional spatial data of the user's current environment in real time, identify the positions of obstacles and the movement trajectories of dynamic objects in the environment, and build a dynamic environmental model;

[0008] An action capture module, used to obtain the user's limb movement data in real time through multi-modal sensors, analyze the current action characteristics of the user, and store them as the user's historical action patterns;

[0009] A game logic analysis module, which is used to connect to the game content database, and according to the current game scene, parse the sequence of required action, the target movement path, and the game action requirement parameters that the user needs to complete within a preset time period in the future;

[0010] An action prediction module, which is used to generate the future predicted action sequence of the user and the spatial coordinate change trajectory of the corresponding body parts through a time series prediction algorithm based on the historical action patterns stored in the action capture module, the currently parsed current action features, and the action requirement parameters output by the game logic analysis module;

[0011] A collision risk assessment module, which is used to compare the spatial coordinate trajectory of the predicted action sequence with the dynamic environment model in real time, and calculate the contact probability and collision risk level of each predicted action with the obstacle;

[0012] A dynamic guidance generation module, which is used to generate an action correction instruction according to the collision risk assessment result, and generate the subsequent action guidance that meets the game goal and avoids obstacles.

[0013] As a further solution of the present invention, the execution steps of the environment perception module include:

[0014] Obtain environmental point cloud data and identify obstacle information, including the contours of static obstacles, and detect the real-time position and movement state of dynamic obstacles;

[0015] Fuse the information of the identified static and dynamic obstacles to establish a comprehensive dynamic environment model.

[0016] As a further solution of the present invention, the execution steps of the game logic analysis module include:

[0017] Connect to the game content database, obtain the status information of the current game scene, identify the dynamic elements in the current game scene, parse the game logic, and determine the target tasks that the user needs to complete in the current scene and the corresponding action requirement parameters;

[0018] Based on the target tasks that need to be completed in the current scene and the corresponding action requirement parameters, and combined with the tasks that the user has completed currently, identify the action sequence that the game requires the user to execute within a preset time period in the future, which is defined as the required action sequence;

[0019] Based on the dynamic environment model and the required action sequence, analyze the required actions of the game and combine them to form a target movement path;

[0020] Read the game content database, use the required actions of the game as labels, identify the requirement parameters of each required action, including force, speed, and movement mode, and combine them with the target movement path.

[0021] As a further solution of the present invention, the execution steps of the action prediction module include:

[0022] Extract the user's historical action patterns, combine with the dynamic environment model, and identify the position coordinates of the user in the dynamic environment model at the current moment;

[0023] Read the action requirement parameters, combine with the current required action sequence and the position coordinates of the user in the dynamic environment model at the previous moment, establish a time series prediction algorithm, identify the user's predicted actions corresponding to each required action in the required action sequence, and combine them into a predicted action sequence;

[0024] According to the predicted action sequence, combine with the dynamic environment model and the position coordinates of the user in the dynamic environment model at the current moment, simulate the user's limb movements in the dynamic environment model, and combine them into a predicted movement trajectory. At the same time, identify the spatial coordinates corresponding to each user's predicted action in the predicted action sequence.

[0025] As a further solution of the present invention, the execution steps of the collision risk assessment module include:

[0026] Obtain the spatial coordinates and body part movement trajectories corresponding to each user's predicted action in the predicted action sequence. At the same time, obtain the contour coordinates of static obstacles, the real-time positions and movement state data of dynamic obstacles in the dynamic environment model, and establish a three-dimensional bounding box for static and dynamic obstacles;

[0027] Perform spatial mapping of the body part movement trajectories corresponding to the user's predicted actions in the dynamic environment model. For each user's predicted action, establish a three-dimensional bounding box of the human body containing the movement range of the human body part based on the spatial coordinates, and perform spatial position matching with the three-dimensional bounding box of the obstacles in the dynamic environment model;

[0028] According to the contour coordinates of static obstacles and the real-time positions and movement states of dynamic obstacles, calculate the spatial overlap degree between the spatial coordinates of the user's predicted actions at the corresponding time points and the three-dimensional bounding box of the obstacles, and based on the spatial overlap degree, construct a contact probability calculation model to calculate the contact probability between each user's predicted action and the obstacles;

[0029] Preset a collision danger level threshold interval, and according to the contact probability calculation results, compare and match the contact probabilities between each user's predicted action and the obstacles with the collision danger level threshold interval to determine the corresponding collision danger level;

[0030] Integrate the contact probabilities and collision danger levels corresponding to each user's predicted action to form a collision risk assessment result corresponding to the predicted action sequence.

[0031] As a further solution of the present invention, the execution steps of the dynamic guidance generation module include:

[0032] Analyze the collision risk assessment results, identify the predicted actions with high collision risk levels, and identify the action nodes and corresponding body parts that will come into contact with obstacles;

[0033] Combined with the game goal and the status information of the current game scene, analyze the role and necessity of each user's predicted action above the collision risk level threshold interval in the game logic, and determine whether the action can be adjusted without affecting the game goal;

[0034] For user predicted actions that are above the collision risk level threshold range and are adjustable and do not affect the game objectives, the motion planning algorithm is used to replan the spatial trajectory and motion parameters of the action while ensuring obstacle avoidance, and generate a revised action plan based on the dynamic environment model and the position and state information of the obstacles.

[0035] The revised action plan is integrated with the original required action sequence, and the execution order, action direction and action range of the required action sequence are adjusted according to the timing requirements and continuity of the game.

[0036] As a further solution of the present invention, the time series prediction algorithm is established to identify the user prediction action corresponding to each demand action in the demand action sequence and combine them into a prediction action sequence, specifically:

[0037] Prediction Spatial coordinates of time It is expressed as:

[0038] ;

[0039] in, express The user's position coordinates in the dynamic environment model at the moment, Represents the instantaneous velocity vector corresponding to the current demand action, Represents the acceleration vector corresponding to the current demand action, is a preset time interval;

[0040] Then for the first Action, its predicted position recursively calculates the cumulative motion of all previous actions:

[0041] .

[0042] As a further solution of the present invention, the three-dimensional bounding boxes of the static obstacles and the dynamic obstacles are established as follows:

[0043] For static obstacles:

[0044] The contour coordinates are composed of the set of vertices after processing the point cloud data and each vertex coordinate is ;

[0045] Then the calculation of the vertices of the obstacle bounding box for the static obstacle is:

[0046] ;

[0047] Among them, the obstacle bounding box is composed of the vertices:

[0048] , ;

[0049] uniquely determined;

[0050] For dynamic obstacles:

[0051] The real-time position is ;

[0052] The movement state includes the speed and the acceleration ;

[0053] Among them are respectively , , the speeds in three directions, are respectively , , the accelerations in three directions;

[0054] Then the calculation of the vertices of the obstacle bounding box for the dynamic obstacle at the current moment is:

[0055] ;

[0056] Among them, , , are respectively the fixed width, height and depth of the dynamic obstacle;

[0057] Among them, the obstacle bounding box is composed of the vertices:

[0058] , ;

[0059] uniquely determined.

[0060] As a further solution of the present invention, calculating the spatial overlap degree between the predicted action of the user at the corresponding time point and the three-dimensional bounding box of the obstacle specifically is:

[0061] Read the obstacle bounding box: ;

[0062] Construct the human body bounding box: ;

[0063] Calculate the length of the overlapping interval in a single axis:

[0064] ;

[0065] ;

[0066] ;

[0067] Among them, , , respectively represent the length of the overlapping interval between the corresponding axis ranges of the obstacle bounding box and the human body bounding box in the axis, axis, axis directions;

[0068] Calculate the ratio of the volume of the intersection of the human body bounding box and the obstacle bounding box to the volume of the union as the overlapping degree:

[0069] ;

[0070] Among them, represents the volume of the obstacle bounding box, represents the volume of the human body bounding box;

[0071] ;

[0072] ;

[0073] The specific method for calculating the contact probability between each user's predicted action and the obstacle is as follows:

[0074] Contact probability .

[0075] The beneficial effects of the present invention are:

[0076] The system realizes a complex collision detection mechanism. By establishing a three-dimensional bounding box, the system can accurately calculate the contact probability between the user's action and the obstacle. This technological breakthrough enables the system to provide safety protection in a real-time dynamic environment. Users can play games without obstacles, reducing the risk of accidental collisions.

[0077] Through the action correction instruction generation mechanism, it is ensured that when the user is about to collide, the action can be adjusted in a timely manner. This intelligent guidance not only improves the safety of the game but also enhances the user's immersion. The user can experience more natural and smooth operations in the game, and this feedback mechanism greatly improves the user's gaming experience.

[0078] This system creates a safe, flexible, and immersive gaming environment for users. Users can not only enjoy higher interactivity but also move freely in a real-time environment without worrying about potential collision risks. The intelligent and personalized features of the system enable each user to obtain a unique gaming experience and achieve higher participation and immersion in the game.

[0079] This game interaction system combined with limb motion sensing represents the future direction of game interaction technology, providing a safer, more personalized, and immersive gaming experience, demonstrating significant progress in intelligent, real-time feedback, and user experience design. These innovations and advancements drive the entire industry to a higher level and lay the foundation for more complex and interactive game scenarios in the future. Brief Description of the Drawings

[0080] Figure 1 It is a structural block diagram of a game interaction system combined with limb motion sensing provided by an embodiment of the present invention;

[0081] Figure 2 It is a flowchart of the execution steps of the environment perception module provided by an embodiment of the present invention;

[0082] Figure 3 It is a flowchart of the execution steps of the game logic analysis module provided by an embodiment of the present invention;

[0083] Figure 4 It is a flowchart of the execution steps of the action prediction module provided by an embodiment of the present invention;

[0084] Figure 5 It is a flowchart of the execution steps of the collision risk assessment module provided by an embodiment of the present invention;

[0085] Figure 6 It is a flowchart of the execution steps of the dynamic guidance generation module provided by an embodiment of the present invention. Detailed Embodiment

[0086] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present application.

[0087] In the description of the present application, the terms "first" and "second" are used only for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present application, "a plurality of" means two or more, unless otherwise specifically defined.

[0088] In the description of the present application, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present application is not necessarily construed as being more preferred or having more advantages than other embodiments. In order for any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope that conforms to the principles and features disclosed in the present application.

[0089] Figure 1 The structural block diagram of a game interaction system combined with limb motion sensing provided for an embodiment of the present invention is as Figure 1 shown, and the system includes:

[0090] An environment perception module 100, configured to collect three-dimensional space data of the environment where the user is located in real time, identify the positions of obstacles and the movement trajectories of dynamic objects in the environment, and construct a dynamic environment model;

[0091] This module identifies the contours of static obstacles in the environment by acquiring environmental point cloud data and using a laser scanner, a depth camera, or other three-dimensional sensing devices. These static obstacles may include walls, furniture, equipment, etc., and their accurate identification is crucial for the safety and feasibility of subsequent user actions. At the same time, the system also needs to monitor the positions and movement states of dynamic obstacles in real time, such as moving objects, pedestrians, etc., to ensure that the changes in the environment can be updated and reflected in a timely manner.

[0092] Once this information is collected, the system fuses the data of static and dynamic obstacles to establish a comprehensive dynamic environment model. This model not only contains the geometric information of the obstacles, but also the movement trajectories and dynamic change characteristics of the objects, and can dynamically adapt to the real-time changes of the user environment. Through such a dynamic model, the system can provide accurate environmental feedback to the user and effectively avoid potential collision risks in subsequent motion planning.

[0093] Through precise environmental data collection and dynamic model construction, this module ensures that the system can reflect the complex environment where the user is located in real time. This not only improves the user experience, enabling the user to play games in a safer and more immersive environment, but also enhances the system's adaptability to environmental changes, thereby improving the accuracy and safety of the user's movements. Through continuous environmental monitoring, the system can dynamically adjust the execution plan of the user's actions, ensuring that during the game, the user can interact without obstacles, reducing the risk of accidental collisions, and greatly enhancing the fun and sense of participation of the game.

[0094] As Figure 2 shown, the execution steps of the environmental perception module 100 include:

[0095] S110, obtaining environmental point cloud data and identifying obstacle information, including the contours of static obstacles, and detecting the real-time positions and movement states of dynamic obstacles;

[0096] S120, fusing the information of the identified static and dynamic obstacles to establish a comprehensive dynamic environment model.

[0097] The motion capture module 200 is used to obtain real-time user limb motion data through multi-modal sensors, analyze the current motion characteristics of the user, and store them as the user's historical motion patterns;

[0098] The game logic analysis module 300 is used to connect to the game content database and parse the sequence of required actions, the target movement path, and the game action requirement parameters that the user needs to complete within a preset time period according to the current game scenario;

[0099] Obtain the state information of the current game scenario, including the real-time states of various dynamic elements in the scenario, such as moving enemies, interactive items, the behaviors of NPCs (non-player characters), etc. By applying computer vision and environmental perception technologies, the system can efficiently identify and track these dynamic elements, thereby establishing a dynamic game environment view.

[0100] After identifying the dynamic elements of the current scenario, the system further analyzes the game logic to determine the target tasks that the user needs to complete in a specific scenario. This may be a complex multi-step task. For example, in a competitive game, the player needs to simultaneously avoid enemy attacks, search for items, and complete specific kill targets. The system not only identifies these tasks but also extracts the action requirement parameters required for each task, such as the required force, speed, and movement mode. For example, if the user needs to cross an obstacle area, the system may require the user to jump at a relatively fast speed while precisely controlling the jump height and landing posture.

[0101] After clarifying the target task, the system will analyze the user's historical behavior patterns in combination with the tasks already completed by the user, and predict the sequence of the next 5 to 10 actions to be performed. The prediction in this step is mainly to clarify the actions required for the subsequent system requirements, that is, the actions the user is about to perform, so as to provide a basis for the subsequent action analysis.

[0102] As Figure 3 shown, the execution steps of the game logic analysis module 300 include:

[0103] S310, connect to the game content database, obtain the status information of the current game scene, identify the dynamic elements in the current game scene, parse the game logic, and determine the target task to be completed by the user in the current scene and the corresponding action requirement parameters;

[0104] S320, based on the target task to be completed in the current scene and the corresponding action requirement parameters, in combination with the tasks already completed by the user, identify the sequence of actions that the game requires the user to perform within a preset future time period, defined as the required action sequence;

[0105] S330, based on the dynamic environment model and the required action sequence, analyze the required actions of the game and combine them to form the target movement path;

[0106] S340, read the game content database, use the required actions of the game as labels, identify the required parameters of each required action, including strength, speed, and movement mode, and combine them with the target movement path.

[0107] The action prediction module 400 is used to generate the user's future predicted action sequence and the spatial coordinate change trajectory of the corresponding body parts through a time series prediction algorithm based on the historical action patterns stored in the action capture module, the currently parsed current action features, and the action requirement parameters output by the game logic analysis module;

[0108] This module will extract the user's historical action patterns. This process uses machine learning and data mining techniques to extract features from the user's past action records, including action frequency, type, and success rate, etc. Combining with the dynamic environment model, the system can identify the user's position coordinates in the environment at the current moment. This position coordinate is updated in real time to ensure that the system makes predictions based on the user's actual state.

[0109] Next, the system reads the current action requirement parameters and combines them with the user's previous demand action sequence to establish a time series prediction algorithm. Through such a mathematical model, the system can predict the spatial coordinates of future actions and then generate a future action sequence. For example, if the user needs to move quickly and avoid obstacles in the current game, the system will, based on historical data and environmental changes, in real time identify the actions that the user may perform and the possible spatial coordinates in the next 5-10 action sequences.

[0110] Through a highly intelligent time series prediction algorithm, it is possible to provide users with a forward-looking gaming experience. This ability to predict in advance not only improves the smoothness of the game but also elevates the user experience to a new level. Through prediction and simulation, the system can effectively reduce the risk of collisions and increase the user's sense of security and immersion.

[0111] For example, in some action games, users may need to continuously perform a series of rapidly changing actions. Through the prediction mechanism, the system can identify potential risks in these actions in advance. This proactive feedback mechanism not only enhances the interaction between the user and the game but also improves the overall quality of the gaming experience, enabling users to remain flexible and agile in a constantly changing environment.

[0112] As Figure 4 shown, the execution steps of the action prediction module 400 include:

[0113] S410, extract the user's historical action patterns, combine them with the dynamic environment model, and identify the position coordinates of the user in the dynamic environment model at the current moment;

[0114] S420, read the action requirement parameters, combine them with the current demand action sequence and the position coordinates of the user in the dynamic environment model at the previous moment, establish a time series prediction algorithm, identify the predicted actions of the user corresponding to each demand action in the demand action sequence, and combine them into a predicted action sequence;

[0115] In this step, specifically:

[0116] Predict the spatial coordinates at the th moment which is expressed as:

[0117] ;

[0118] where represents the position coordinates of the user in the dynamic environment model at the th moment, represents the instantaneous velocity vector corresponding to the current demand action, represents the acceleration vector corresponding to the current demand action, is the preset time interval;

[0119] For the th action in the demand action sequence, its predicted position recursively accumulates the movements of all previous actions:

[0120] .

[0121] S430. According to the predicted action sequence, combined with the dynamic environment model and the position coordinates of the user in the dynamic environment model at the current moment, simulate the limb movements of the user in the dynamic environment model, and combine them into a predicted movement trajectory. At the same time, identify the spatial coordinates corresponding to each user predicted action in the predicted action sequence.

[0122] The collision risk assessment module 500 is used to compare the spatial coordinate trajectory of the predicted action sequence with the dynamic environment model in real time, and calculate the contact probability and collision risk level between each predicted action and the obstacle;

[0123] For static obstacles, the system will identify the boundary points of the obstacles through point cloud data processing to form an accurate three-dimensional bounding box. The calculation formula of this bounding box shows how to determine its spatial range through the geometric information of the static obstacle (such as the point coordinates in the vertex set G). This enables the system to effectively locate the obstacle in three-dimensional space and provide a reliable reference framework for subsequent dynamic collision detection.

[0124] For dynamic obstacles, the system adopts a similar strategy to update its position and motion state in real time. The implementation of this part depends on sensor data and motion models to ensure that no matter how the obstacle moves, the system can always accurately reflect its position in space. This dynamic update mechanism greatly improves the authenticity and responsiveness of the game environment, enabling users to obtain the most realistic environmental feedback when performing actions.

[0125] Next, the system will perform a spatial mapping of the predicted action trajectory of the user with the bounding boxes of these obstacles. By calculating the movement trajectory of the body part corresponding to the user action, the system can establish a three-dimensional bounding box of the human body and match it with the bounding box of the obstacle. The key to this step is that it can calculate the contact probability between the user action and the obstacle through the spatial overlap degree, and then evaluate the collision risk level. At this time, the system will calculate the spatial overlap situation between the spatial coordinates of each predicted action at the corresponding time point and the obstacle bounding box to form a contact probability calculation model.

[0126] After determining the contact probability of each user's predicted action, the system compares it with a preset collision risk level threshold. This evaluation process helps determine which actions have a high collision risk, thereby allowing the system to perform precise risk management. For actions whose contact probability exceeds the safety threshold, the system can respond in a timely manner and generate corresponding collision risk assessment results.

[0127] By implementing real-time dynamic detection, it ensures that users are always in a safe environment during the game. This proactive collision detection mechanism not only enhances the user's sense of security but also greatly enhances the immersion and engagement of the game. Users can freely perform complex actions without worrying about collision risks, so they can focus more on game strategies and operations.

[0128] In a virtual reality combat game, users need to shuttle between enemies and attack. By combining the predicted future action sequence of the user obtained from the action prediction module and the spatial coordinate change trajectory of the corresponding body parts, and executing the collision detection mechanism, the system can real-time predict the relationship between the user's movement path and the enemy's position, and when the user is about to enter a high-risk area, issue a warning in a timely manner or adjust their preset actions. This not only ensures the smoothness of the game but also improves the user's participation and the fun of the game experience.

[0129] As Figure 5 shown, the execution steps of the collision risk assessment module 500 include:

[0130] S510, obtain the spatial coordinates and body part movement trajectories corresponding to each user's predicted action in the predicted action sequence, and at the same time obtain the contour coordinates of static obstacles, the real-time positions and movement state data of dynamic obstacles in the dynamic environment model, and establish a three-dimensional bounding box for static and dynamic obstacles;

[0131] In this step, the establishment of the three-dimensional bounding box for static and dynamic obstacles is specifically:

[0132] For static obstacles:

[0133] The contour coordinates are composed of the vertex set after point cloud data processing and each vertex coordinate is ;

[0134] Then the calculation of the vertices of the bounding box of the static obstacle is:

[0135] ;

[0136] Among them, the bounding box of the obstacle is composed of vertices:

[0137] , ;

[0138] uniquely determined;

[0139] For dynamic obstacles:

[0140] The real-time position is ;

[0141] The movement state includes speed and acceleration ;

[0142] wherein are respectively , , the speeds in three directions, are respectively , , the accelerations in three directions;

[0143] Then the vertices of the obstacle bounding box of the dynamic obstacle at the current moment are calculated as:

[0144] ;

[0145] wherein, , , are respectively the fixed width, height and depth of the dynamic obstacle;

[0146] wherein, the obstacle bounding box is composed of the vertices:

[0147] , ;

[0148] uniquely determined.

[0149] S520, perform spatial mapping of the movement trajectory of the body part corresponding to the user's predicted action in the dynamic environment model. For each user's predicted action, establish a three-dimensional human bounding box including the movement range of the human body part based on the spatial coordinates, and perform spatial position matching with the three-dimensional obstacle bounding box in the dynamic environment model;

[0150] S530, calculate the spatial overlap degree between the spatial coordinates of the user's predicted action at the corresponding time point and the three-dimensional obstacle bounding box according to the contour coordinates of the static obstacle and the real-time position and movement state of the dynamic obstacle, and construct a contact probability calculation model based on the spatial overlap degree to calculate the contact probability between each user's predicted action and the obstacle;

[0151] In this step, the calculation of the spatial overlap degree between the spatial coordinates of the user's predicted action at the corresponding time point and the three-dimensional obstacle bounding box is specifically:

[0152] Read the obstacle bounding box: ;

[0153] Construct the human body bounding box: ;

[0154] Calculate the length of the overlapping interval in a single axis:

[0155] ;

[0156] ;

[0157] ;

[0158] Among them, , , respectively represent the length of the overlapping interval between the corresponding axis ranges of the obstacle bounding box and the human body bounding box in the axis, axis, axis directions;

[0159] Calculate the ratio of the volume of the intersection of the human body bounding box and the obstacle bounding box to the volume of the union as the overlap degree:

[0160] ;

[0161] Among them, represents the volume of the obstacle bounding box, represents the volume of the human body bounding box;

[0162] ;

[0163] ;

[0164] The specific method for calculating the contact probability between each user's predicted action and the obstacle is as follows:

[0165] Contact probability .

[0166] S540, a preset collision risk level threshold interval. According to the calculation result of the contact probability, compare and match the contact probability between each user's predicted action and the obstacle with the collision risk level threshold interval to determine the corresponding collision risk level;

[0167] S550, integrate the contact probability and the collision risk level corresponding to each user's predicted action to form a collision risk assessment result corresponding to the predicted action sequence.

[0168] The dynamic guidance generation module 600 is used to generate an action correction instruction according to the collision risk assessment result and generate a subsequent action guidance that meets the game goal and avoids obstacles.

[0169] This module will conduct an in-depth analysis of the collision risk assessment results to identify predicted actions with a high collision risk level. These actions may face the risk of contact with static or dynamic obstacles due to the user's movement path or execution method. In this step, the system not only identifies potential dangers, but also specifically identifies those action nodes and corresponding body parts that come into contact with obstacles. For example, if the user's knee is likely to contact the wall next to him when performing a high leg lift, the system will accurately identify this risk.

[0170] Next, combining the game goal and the status information of the current game scene, the system analyzes the role and necessity of each high-risk action in the game logic. This analysis process ensures that the system can judge which actions can be adjusted and which are indispensable. For example, if an action is necessary to complete a task, the system will give priority to adjusting it instead of simply deleting or replacing the action. This judgment relies on the understanding of the game logic, allowing the system to maintain the consistency of the game experience.

[0171] For high-risk actions that are adjustable and do not affect the game objectives, the system will use motion planning algorithms to re-plan the spatial trajectory and motion parameters of the action based on the dynamic environment model and the position and state information of obstacles. This process ensures that when adjusting the action, the user can still achieve the goal and effectively avoid obstacles. For example, if the user's action requires moving from one area to another, the system may suggest that the user move sideways instead of moving straight forward to reduce the risk of collision with obstacles, or by expanding the user's movement feedback, only guiding the user to move in small segments, but feedback to the game is a large segment of displacement.

[0172] After generating the revised action plan, the system will merge it with the original required action sequence. At this time, the system will also consider the timing requirements and coherence of the game, and adjust the execution order, action direction, and action range of the required action sequence. This comprehensive adjustment ensures that when users perform subsequent actions, they can complete the game tasks smoothly and naturally without feeling abrupt or uncoordinated.

[0173] Through intelligent dynamic feedback and real-time adjustments, users can operate more safely and confidently in complex gaming environments. First, the system can promptly identify and reduce collision risks, ensure the safety of users in the game, and reduce the chance of accidents. Second, by understanding the user's behavior and game logic, the system ensures that the adjusted actions will not affect the user's game goals, thereby improving the overall gaming experience.

[0174] For example, in an action adventure game, if a user is about to collide with a moving enemy while running, the system will immediately calculate the best avoidance plan, guiding the user to move laterally or jump, ensuring that the user can not only avoid the collision but also continue to move towards the goal. This real-time feedback mechanism greatly enhances the user's immersion and sense of participation, enabling the user to enjoy higher safety and freedom while experiencing the fun of the game.

[0175] As Figure 6 shown, the execution steps of the dynamic guidance generation module 600 include:

[0176] S610, analyze the collision risk assessment result, identify the predicted actions with a high collision risk level, and identify the action nodes and corresponding body parts that will come into contact with obstacles;

[0177] S620, combine the game goal and the state information of the current game scene, analyze the role and necessity of each user's predicted action in the game logic that is higher than the collision risk level threshold interval, and determine whether the action can be adjusted without affecting the game goal;

[0178] S630, for the user's predicted actions that are adjustable and do not affect the game goal and are higher than the collision risk level threshold interval, based on the dynamic environment model and the position and state information of the obstacles, use the motion planning algorithm to re-plan the spatial trajectory and motion parameters of the action on the premise of avoiding obstacles, and generate a corrected action plan;

[0179] S640, fuse the corrected action plan with the original required action sequence, and adjust the execution order, action direction, and action range of the required action sequence according to the timing requirements and coherence of the game.

[0180] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0181] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0182] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0183] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0184] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0185] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0186] Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A game interaction system combined with limb movement sensing, characterized in that, The system includes: An environmental perception module, which is used to collect three-dimensional space data of the user's environment in real time, identify the positions of obstacles and the movement trajectories of dynamic objects in the environment, and construct a dynamic environment model; An action capture module, which is used to obtain the user's limb action data in real time through multi-modal sensors, analyze the current action characteristics of the user, and store them as the user's historical action patterns; A game logic analysis module, which is used to connect to the game content database, and according to the current game scenario, analyze the sequence of required actions, the target movement path, and the game action requirement parameters that the user needs to complete within a preset time period in the future; An action prediction module, which is used to generate the user's future predicted action sequence and the spatial coordinate change trajectory of the corresponding body parts through a time series prediction algorithm based on the historical action patterns stored in the action capture module, the current action characteristics analyzed in real time, and the action requirement parameters output by the game logic analysis module; A collision risk assessment module, which is used to compare the spatial coordinate trajectory of the predicted action sequence with the dynamic environment model in real time, and calculate the contact probability between each predicted action and the obstacle and the collision risk level; A dynamic guidance generation module, which is used to generate an action correction instruction according to the collision risk assessment result, and generate subsequent action guidance that meets the game goal and avoids obstacles; The execution steps of the action prediction module include: Extract the user's historical action patterns, combine with the dynamic environment model, and identify the position coordinates of the user in the dynamic environment model at the current moment; Read the action requirement parameters, combine with the current required action sequence and the position coordinates of the user in the dynamic environment model at the previous moment, establish a time series prediction algorithm, identify the user's predicted actions corresponding to each required action in the required action sequence, and combine them into a predicted action sequence; According to the predicted action sequence, combine with the dynamic environment model and the position coordinates of the user in the dynamic environment model at the current moment, simulate the user's limb movement in the dynamic environment model, and combine it into a predicted movement trajectory, and at the same time identify the spatial coordinates corresponding to each user's predicted action in the predicted action sequence; The establishment of the time series prediction algorithm, the identification of the user's predicted actions corresponding to each required action in the required action sequence, and the combination into a predicted action sequence are specifically: Predict the spatial coordinates at a moment, expressed as: ； Among them, represents the position coordinates of the user in the dynamic environment model at a moment, represents the instantaneous velocity vector corresponding to the current required action, represents the acceleration vector corresponding to the current required action, is a preset time interval; For the th action in the demand action sequence, its predicted position recursively accumulates the movements of all previous actions: 。 2. The system according to claim 1, wherein The execution steps of the environmental perception module include: Obtain environmental point cloud data and identify obstacle information, including the outlines of static obstacles, and detect the real-time positions and movement states of dynamic obstacles; Fuse the information of the identified static and dynamic obstacles to establish a comprehensive dynamic environment model.

3. The system according to claim 2, characterized in that, The execution steps of the game logic analysis module include: Connect to the game content database, obtain the status information of the current game scenario, identify the dynamic elements in the current game scenario, analyze the game logic, and determine the target tasks that the user needs to complete in the current scenario and the corresponding action requirement parameters; Based on the target tasks that need to be completed in the current scenario and the corresponding action requirement parameters, combine with the tasks that the user has completed currently, and identify the action sequence that the game requires the user to execute within a preset time period in the future, which is defined as the required action sequence; Analyze the required actions of the game based on the dynamic environment model and the sequence of required actions, and combine them to form a target movement path. Read the game content database, use the required actions of the game as labels to identify the required parameters of each required action, including force, speed, and movement mode, and combine them with the target movement path.

4. The system according to claim 3, wherein The execution steps of the collision risk assessment module include: Obtain the spatial coordinates and body part movement trajectories corresponding to each user's predicted action in the predicted action sequence. At the same time, obtain the contour coordinates of static obstacles, the real-time positions and movement state data of dynamic obstacles in the dynamic environment model, and establish a three-dimensional obstacle bounding box for static and dynamic obstacles. Perform spatial mapping of the body part movement trajectories corresponding to the user's predicted actions in the dynamic environment model. For each user's predicted action, establish a three-dimensional human bounding box containing the movement range of the human body part based on the spatial coordinates, and perform spatial position matching with the three-dimensional obstacle bounding box in the dynamic environment model. According to the contour coordinates of static obstacles and the real-time positions and movement states of dynamic obstacles, calculate the spatial overlap degree between the spatial coordinates of the user's predicted action at the corresponding time point and the three-dimensional obstacle bounding box, and based on the spatial overlap degree, construct a contact probability calculation model to calculate the contact probability between each user's predicted action and the obstacle. Preset a threshold range for the collision danger level. According to the calculation results of the contact probability, compare and match the contact probability between each user's predicted action and the obstacle with the threshold range of the collision danger level to determine the corresponding collision danger level. Integrate the contact probabilities and collision danger levels corresponding to each user's predicted action to form a collision risk assessment result corresponding to the predicted action sequence.

5. The system according to claim 4, wherein The execution steps of the dynamic guidance generation module include: Analyze the collision risk assessment result, identify the predicted actions with a high collision danger level, and identify the action nodes and corresponding body parts that will come into contact with the obstacles. Combined with the game goal and the state information of the current game scene, analyze the role and necessity of each user's predicted action above the threshold range of the collision danger level in the game logic, and determine whether the action can be adjusted without affecting the game goal. For the user's predicted actions above the threshold range of the collision danger level that can be adjusted and do not affect the game goal, based on the dynamic environment model and the position and state information of the obstacles, use the motion planning algorithm to re-plan the spatial trajectory and motion parameters of the action on the premise of ensuring avoiding obstacles, and generate a modified action plan. Fuse the modified action plan with the original sequence of required actions, and adjust the execution order, action direction, and action range of the sequence of required actions according to the timing requirements and coherence of the game.

6. The system according to claim 4, wherein The establishment of the three-dimensional obstacle bounding box for static and dynamic obstacles is specifically as follows: For static obstacles: The contour coordinates are composed of the vertex set after point cloud data processing and each vertex coordinate is ; Then the vertex calculation of the obstacle bounding box of the static obstacle is: ； Among them, the obstacle bounding box is uniquely determined by the vertices: ， ； Unique determination; For dynamic obstacles: The real-time position is ; The moving state includes speed and acceleration ; Among them are respectively , , the velocities in three directions, are respectively , , the accelerations in three directions; Then the vertex calculation of the obstacle bounding box of the dynamic obstacle at the current moment is: ； Among them, , , are the fixed width, height, and depth of the dynamic obstacle, respectively; Among them, the obstacle bounding box at a moment consists of vertices: ， ； Unique determination.

7. The system according to claim 4, characterized in that Calculating the spatial overlap degree between the predicted action of the user at the corresponding time point and the three-dimensional bounding box of the obstacle, specifically: Read obstacle bounding box: ; Construct a human body bounding box: ; Calculating the length of the single-axis overlap interval: ； ； ； Among them, , , respectively represent the lengths of the overlapping intervals between the corresponding axis ranges of the obstacle bounding box and the human body bounding box in the axis, axis, axis directions; Calculating the ratio of the volume of the intersection of the human bounding box and the obstacle bounding box to the volume of the union as the overlap degree: ； Among them, represents the volume of the obstacle bounding box, represents the volume of the human body bounding box; ； ； Calculating the contact probability between each predicted action of the user and the obstacle, specifically: Contact probability 。

Images