A Redirecting Walking Method and Apparatus Based on Dynamic Alignment
By analyzing user motion information to select interaction targets, improving artificial potential fields and clustering methods, and dynamically adjusting the resultant force, the alignment problem in multi-user VR collaboration is solved, and the immersion and obstacle avoidance effects are improved.
Patent Information
- Application Number
- CN202311543744.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-20
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-11-20
AI Technical Summary
Existing redirection walking technology is not effectively applicable to multi-user VR collaborative applications. It cannot guarantee that the relative positions of multiple users are aligned in the physical and virtual environments, and it lacks user immersion and experience.
By analyzing user motion information to select the best potential interaction target, the traditional artificial potential field calculation of repulsive force is improved. Clustering methods are added to eliminate interference and dynamically adjust the resultant force to achieve user alignment in virtual and physical environments and reduce the probability of collision.
It improves the accuracy of dynamic alignment for multiple users, reduces the probability of collisions between users and obstacles, and enhances the user's immersion and interactive experience in the virtual environment.
Smart Images

Figure CN117572969B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of redirection walking, and particularly relates to a redirection walking method and apparatus based on dynamic alignment. Background Technology
[0002] In recent years, numerous virtual reality (VR) applications have broken down traditional boundaries, allowing users to explore virtual environments immersively by wearing head-mounted displays (HMDs). Compared to walking in place or flying, experiencing VR through natural walking can enhance the user's immersion when exploring virtual environments. However, allowing users to walk naturally in constrained physical environments is a challenge. To enable users to explore large virtual environments by walking naturally in constrained physical environments, redirected walking technology has been proposed. This technology subtly guides users along physical paths different from the virtual paths by slightly manipulating the virtual environment, avoiding collisions with unseen obstacles and thus providing natural walking in virtual reality. Therefore, improving the immersive experience of VR through redirected walking technology has received widespread attention and research, with many scholars conducting studies on redirected walking techniques such as perceptual threshold detection, reactive redirected walking algorithms, and pre-tested redirected walking algorithms.
[0003] With the development of redirection walking technology, multiple users can now roam virtual environments within a shared physical environment, breaking the limitations of VR's singularity. Multi-user redirection walking algorithms are mainly divided into two categories: those where multiple users are invisible in the virtual environment, and those where multiple users are visible. Existing algorithms focus on the former, aiming to reduce the probability of collisions between users and obstacles. These algorithms do not consider interactions between multiple users, making them unsuitable for multi-user VR collaborative applications and lacking user immersion and experience. Multi-user VR collaborative applications belong to the latter category. Currently, with the rise of the "metaverse," it is no longer uncommon for multiple users to experience a virtual environment together, such as in virtual social interactions, military training, and games. Achieving VR collaboration requires not only avoiding collisions between users but also ensuring that the relative positions of users are aligned in both the physical and virtual environments. Furthermore, most of the existing alignment-based redirection walking methods are designed for alignment between a single user and static interactive objects, and are not applicable to alignment between multiple users. Summary of the Invention
[0004] The present invention aims to overcome the above-mentioned shortcomings of the prior art and provides a multi-user redirection walking method and apparatus based on dynamic alignment.
[0005] When multiple users share a physical environment while roaming within the same virtual environment, if a user approaches and interacts with another user in the virtual environment, the algorithm aims to ensure that the same user can also approach and interact with the corresponding user in the physical environment. Simultaneously, to guarantee a positive user experience, the algorithm minimizes collisions between users and obstacles. This is a dynamic alignment adjustment process for multiple users. In a dynamic and unknown environment, the algorithm first analyzes the motion information of the current user and other users in the virtual environment to select the optimal potential interaction target for the current user, accelerating the alignment of the relative positions of subsequent users in both the physical and virtual environments and promoting interaction. Next, the algorithm improves upon traditional artificial potential fields by defining the repulsive force generated by the repulsive potential field acting on the current user as a function of the relative position and velocity of the obstacle relative to the user. The alignment guiding force acting on the user is then categorized and discussed based on the alignment of the user and the target's relative position and velocity in both the virtual and physical environments. Furthermore, the algorithm incorporates clustering methods to eliminate interfering repulsive forces generated by users surrounding the interaction target. Finally, to balance obstacle avoidance and dynamic alignment, the algorithm of this invention introduces dynamic parameters based on the virtual and physical distances between the user and the interactive target. This adaptively and dynamically adjusts the guiding force for alignment and the repulsive force to avoid collisions, resulting in smoother user-perceived manipulation. Ultimately, the algorithm redirects the user to the optimal direction, improving alignment accuracy, reducing the difference in relative position mapping between the user and the interactive target in the virtual and physical environments, and effectively avoiding obstacles, thus reducing the probability of collisions between the user and obstacles.
[0006] The multi-user redirection walking method based on dynamic alignment of the present invention includes the following steps for each user participating in virtual roaming:
[0007] (1) Select the user’s best potential interaction target;
[0008] (2) Calculate the repulsive force on the user based on the relative position and relative velocity relationship between the user and the obstacle in the physical environment;
[0009] (3) Use the K-means clustering method to cluster other users besides the user to eliminate the repulsive force around the interaction target that interferes with the alignment of the user;
[0010] (4) Calculate the alignment guidance force on the user based on the mapping relationship between the user and the interaction target in the virtual environment and the physical environment, and the relative position and relative velocity of the user and the interaction target in the virtual environment;
[0011] (5) Calculate the resultant force of repulsive force and alignment guiding force dynamically based on the relative positions of the user and the interaction target in the virtual and physical environments;
[0012] (6) Guide the user to the optimal direction based on the combined force;
[0013] This invention proposes a multi-person redirection walking method based on dynamic alignment. This method can not only improve the accuracy of dynamic alignment and reduce the difference in relative position mapping between users and interactive targets in virtual and physical environments, but also effectively avoid obstacles and reduce the probability of collisions between users and obstacles.
[0014] A second aspect of the present invention relates to a multi-person redirection walking device based on dynamic alignment, comprising a memory and one or more processors, wherein the memory stores executable code, and the one or more processors execute the executable code to implement the multi-person redirection walking method based on dynamic alignment of the present invention.
[0015] A third aspect of the invention relates to a computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the dynamic alignment-based multi-person redirection walking method of the present invention.
[0016] This invention addresses the challenge of collaborative interaction among multiple users in the same physical and virtual environment by proposing an innovative multi-user redirection walking method based on dynamic alignment. This invention can select potential interaction targets in a dynamically unknown environment. By improving the traditional artificial potential field, it calculates the repulsive force acting on the current user, rationally sets the alignment guiding force, and incorporates a clustering algorithm to eliminate interfering forces. Finally, it dynamically adjusts the resultant force to guide the user's movement. This method not only effectively reduces the discrepancy in the relative position mapping between the user and the interaction target in the virtual and physical environments, improving alignment accuracy, but also effectively avoids obstacles, reducing collisions between the user and obstacles.
[0017] The advantages of this invention are: it can improve dynamic alignment accuracy, reduce the difference in relative position mapping between users and interactive targets in virtual and physical environments, and effectively avoid obstacles, reducing the probability of collisions between users and obstacles. Attached Figure Description
[0018] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation
[0019] The technical solution of the present invention will be further described below with reference to the accompanying drawings.
[0020] Example 1
[0021] This embodiment relates to a multi-user redirection roaming method based on dynamic alignment, which includes the following steps for each user participating in virtual roaming:
[0022] (1) Select the user’s best potential interaction target;
[0023] The motion information of users in the virtual environment is analyzed as shown in formula (1).
[0024]
[0025] in, v represents the relationship between the relative virtual velocity and relative virtual position of the current user and user j. v The virtual velocity vector representing the current user, The virtual velocity vector representing user j. This represents the virtual straight-line distance between the current user and user j. The virtual angle represents the relative virtual velocity and relative virtual position of user j relative to the current user. The smaller the value, the greater the likelihood that user j and the current user will interact in the virtual environment. and and The size is inversely proportional to, and The size is directly proportional to the size.
[0026] Then, the present invention can select the user's best potential interaction target according to formula (2).
[0027]
[0028] Here, target represents the potential target user for interaction, and U represents other users besides the current user.
[0029] (2) Calculate the repulsive force on the user based on the relative position and relative velocity relationship between the user and the obstacle in the physical environment.
[0030] In a multi-user shared physical environment, there are not only static obstacles such as physical boundaries, but also dynamic obstacles such as other dynamic users. The calculation of the repulsive force of obstacles on users is divided into two parts: the repulsive potential field function of relative position and the repulsive potential field function of relative velocity.
[0031] 2.1 Calculate the relative positional repulsive force experienced by the current user.
[0032] The repulsive potential field function of the physical boundary i relative to the current user's position is established as shown in formula (3).
[0033]
[0034] Among them, U repp1i The function representing the repulsive potential field function generated by the physical boundary i relative to the current user's position. ρ represents the straight-line distance from the current user to the physical boundary i, with the vector direction pointing from the physical boundary i to the current user's physical position. ρ represents the maximum range of influence of each obstacle on the current user's repulsive force.
[0035] The relative positional repulsive potential field function generated by other non-interactive target users j to the current user is established as shown in formula (4).
[0036]
[0037] Among them, U repp2j This represents the repulsive potential field function representing the relative position of other non-interactive target users j with respect to the current user. This represents the physical straight-line distance between the current user and user j, with the vector direction pointing from the physical location of user j to the physical location of the current user.
[0038] Then, U repp1i and U repp2j Differentiate each position separately, and sum the results of the relative position repulsion forces on the current user from each physical boundary and other non-interactive target users to obtain the relative position repulsion sum, as shown in formula (5).
[0039]
[0040] Where p represents the current user's physical location, O represents all physical boundaries, and R represents all users other than the current user and the target user. F repp1i F represents the repulsive force generated by the physical boundary i relative to the current user's position in the repulsive potential field. repp1i The direction F points from the physical boundary i to the current user's physical location. repp2j F represents the repulsive force generated by the repulsive potential field of other non-interactive target users j relative to the current user's position. repp2j The direction is from the physical location of user j to the current user's physical location, F repp This represents the total repulsive force currently experienced by the user.
[0041] 2.2 Calculate the relative velocity repulsion force experienced by the current user.
[0042] In the physical environment, other non-interactive target users are regarded as dynamic obstacles, and the relative velocity repulsion potential field function of other non-interactive target users j to the current user is established as shown in formula (6).
[0043]
[0044] Among them, v p The physical velocity vector representing the current user. The physical velocity vector representing user j. The physical angle represents the relative physical velocity and relative physical position of user j relative to the current user.
[0045] Then F repvj The differential with respect to position is shown in formula (7).
[0046]
[0047] Among them, F repvj F represents the repulsive force generated by user j relative to the current user's velocity repulsive potential field. repvj The direction is perpendicular to the line connecting user j and the current user's relative physical position.
[0048] Finally, the sum of the repulsive forces generated by the relative velocity repulsive potential fields of all other non-interactive target users is calculated to obtain the relative velocity repulsive force sum, as shown in formula (8).
[0049] F repv =∑ j∈R F repvj (8)
[0050] Among them, F repv This represents the sum of the relative velocity repulsion forces currently experienced by the user.
[0051] One point to note is that when When the relative physical velocity and relative physical position of the dynamic obstacle relative to the current user are greater than 90 degrees, the user is moving away from the obstacle. Therefore, this invention does not consider the motion information of the obstacle, so the user does not need to be subject to the relative velocity repulsion force generated by the obstacle.
[0052] 2.3 Calculate the net repulsive force acting on the current user.
[0053] The relative positional repulsion force generated by static and dynamic obstacles and the relative velocity repulsion force generated by dynamic obstacles are calculated by summing up as shown in formula (9).
[0054] F rep =F repp +kF repv (9)
[0055] Among them, F rep This represents the resultant force of the relative positional repulsion and the relative velocity repulsion experienced by the current user, where k represents the proportionality coefficient of the relative velocity repulsion.
[0056] (3) Use the K-means method to cluster other users besides the user to eliminate the repulsive force around the interaction target that interferes with the alignment of the user;
[0057] In an artificial potential field, clustering methods are used to group all users except the target user, eliminating the repulsive forces generated by other users near the interaction target in the physical environment. This eliminates the interference of these repulsive forces in guiding the user's alignment. The steps are as follows: Using the distance-based K-means clustering method, all users in the physical environment are divided into different clusters, where the scan length of the K-means method is 2, and the minimum number of users in each cluster is 1. The cluster containing the interaction target is selected, and other users in this cluster are considered as interference terms and excluded when calculating the repulsive force.
[0058] (4) Calculate the alignment guidance force on the user based on the mapping relationship between the user and the interaction target in the virtual environment and the physical environment, and the relative position and relative velocity of the user and the interaction target in the virtual environment;
[0059] Establish alignment guidance to help users and interaction targets interact, as shown in Equation (10).
[0060] F ali =k a (10)
[0061] Among them, F ali Represents alignment guidance force, k a k represents the alignment guiding force vector. a Its value is a constant.
[0062] About k a The direction is determined by classification and discussion. If the angle between the line connecting the relative physical velocity direction and the relative position of the user and the interactive target is greater than 90°, or if the virtual distance between the user and the interactive target is less than the physical distance between them, then k a The direction points from the user's physical location to the physical location of the interaction target, aiming to allow the user to adjust to move closer to the interaction target in the fastest way possible in the physical environment, thus mirroring the scenario of the user moving closer to the interaction target in the virtual environment. In other cases, the problem is transformed into finding the optimal solution, obtaining k. a The direction is to align the relative physical position and relative physical velocity of the user and the interactive target in the physical environment with the relative virtual position and relative virtual velocity of the user and the interactive target in the virtual environment.
[0063] The relative virtual speed and relative virtual position relationship between the current user and the interactive target in the virtual environment can be summarized as follows: As shown in formula (11).
[0064]
[0065] in, The virtual velocity vector representing the target. This represents the virtual straight-line distance between the current user and the target. This represents the virtual angle between the target's relative virtual velocity and relative virtual position with respect to the current user.
[0066] The optimal relative physical velocity and relative physical position relationship between the user and the interaction target in the physical environment can be summarized as follows: As shown in formulas (12) and (13).
[0067]
[0068]
[0069] in, This represents the optimal physical velocity vector for the user that needs to be determined. The physical velocity vector representing the target. This represents the physical straight-line distance between the current user and the target. This represents the physical angle between the target's relative physical velocity and relative physical position with respect to the current user.
[0070] Next, the problem is transformed into solving how to obtain an optimal physical velocity, as shown in Equation (14).
[0071]
[0072] Alignment of the mapping of the relative positions and relative velocities of users and interaction targets in the physical and virtual environments, i.e. and The difference is minimized, as shown in formula (15). Finally, the optimal physical velocity direction for the user is defined as k. a The direction.
[0073] Min (x,y) fun(x,y)stx 2 +y 2 =||v p || 2 (15)
[0074] (5) The resultant force of the repulsive force and the alignment guiding force on the user is dynamically calculated based on the relative position of the user and the interaction target in the virtual environment and the physical environment, as shown in formula (16).
[0075]
[0076] Where F is defined as the net force acting on the user, γ represents the repulsive force proportionality coefficient, and l p This represents the side length of the square physical environment.
[0077] (6) Guide the user to the optimal direction based on the combined force.
[0078] The direction of the resultant force vector guides the user to take a step forward in the physical space.
[0079] Example 2
[0080] This embodiment relates to a multi-person redirection walking device based on dynamic alignment, including a memory and one or more processors. The memory stores executable code, and when the one or more processors execute the executable code, they are used to implement the multi-person redirection walking method based on dynamic alignment of Embodiment 1.
[0081] Example 3
[0082] This embodiment relates to a computer-readable storage medium storing a program that, when executed by a processor, implements the multi-person redirection walking method based on dynamic alignment as described in Embodiment 1.
[0083] The embodiments described in this specification are merely examples of implementations of the inventive concept. The scope of protection of this invention should not be considered as limited to the specific forms stated in the embodiments. The scope of protection of this invention also extends to equivalent technical means that can be conceived by those skilled in the art based on the inventive concept.
Claims
1. A multi-user redirection walking method based on dynamic alignment, for each user participating in virtual roaming, includes the following steps: (1) Select the user's best potential interaction target; (2) Calculate the repulsive force on the user based on the relative position and relative velocity relationship between the user and the obstacle in the physical environment; (3) Use the K-means clustering method to cluster other users besides the user to eliminate the repulsive force around the interaction target that interferes with the alignment of the user; (4) Calculate the alignment guidance force on the user based on the mapping relationship between the user and the interaction target in the virtual environment and the physical environment, and the relative position and relative velocity of the user and the interaction target in the virtual environment; (5) Calculate the resultant force of repulsive force and alignment guiding force dynamically based on the relative positions of the user and the interaction target in the virtual and physical environments; (6) Guide the user to the optimal direction based on the combined force.
2. The multi-person redirection walking method based on dynamic alignment as described in claim 1, characterized in that: Step (1) specifically includes: The motion information of users in the virtual environment is analyzed as shown in formula (1): (1) in, Represents the current user and users The relationship between relative virtual velocity and relative virtual position, The virtual velocity vector representing the current user, On behalf of users The virtual velocity vector, Represents the current user and users The virtual straight-line distance between them On behalf of users The virtual angle between the relative virtual speed and relative virtual position of the current user; The smaller the value, the better for users. The greater the likelihood of interaction between the current user and the virtual environment; and and The size is inversely proportional to, and The size is directly proportional to the size; Then, select the user's best potential interaction target according to formula (2): (2) in, Representing potential target users for interaction, This represents users other than the current user.
3. The multi-person redirection walking method based on dynamic alignment as described in claim 1, characterized in that: Step (2) specifically includes: In a multi-user shared physical environment, there are not only static obstacles but also dynamic obstacles; the calculation of the repulsive force generated by the obstacles on the user is divided into two parts, namely the repulsive potential field function of relative position and the repulsive potential field function of relative velocity. 2.1 Calculate the relative positional repulsive force experienced by the current user; Establish physical boundaries The repulsive potential field function relative to the current user's position is shown in formula (3): (3) in, Represents physical boundary The repulsive potential field function generated for the relative position of the current user, Represents the current user to the physical boundary The straight-line distance, the vector direction is determined by the physical boundary. Points to the current user's physical location. This represents the maximum range of influence of the obstacle that exerts a repulsive force on the current user. Establish other non-interactive target users The repulsive potential field function generated relative to the current user's position is shown in formula (4): (4) in, Representing other non-interactive target users The repulsive potential field function for the relative position of the current user, Represents the current user and users The physical straight-line distance between them, and the vector direction, are determined by the user. The physical location points to the current user's physical location; Then, and Differentiate each position separately, and sum the results of the relative positional repulsive forces acting on the current user from each physical boundary and other non-interactive target users to obtain the sum of the relative positional repulsive forces, as shown in formula (5): (5) Among them, subscript Represents the current user's physical location. Represents all physical boundaries. This represents users other than the current user and the target user of the interaction. Represents physical boundary The repulsive force generated by the repulsive potential field relative to the current user's position. The direction from the physical boundary Points to the current user's physical location. Representing other non-interactive target users The repulsive force generated by the repulsive potential field relative to the current user's position. The direction from the user The physical location points to the current user's physical location. This represents the total repulsive force currently experienced by the user; 2.2 Calculate the relative velocity repulsion force acting on the current user; In the physical environment, other non-interactive target users are treated as dynamic obstacles, and other non-interactive target users are established. The repulsive potential field function of the relative velocity with respect to the current user is shown in formula (6): (6) in, The physical velocity vector representing the current user. On behalf of users The physical velocity vector, On behalf of users The physical angle between the current user's relative physical speed and relative physical position; Then The differential with respect to position is shown in formula (7): (7) in, On behalf of users The repulsive force generated by the relative velocity repulsive potential field of the current user The direction is perpendicular to the user Connect to the current user's relative physical location; Finally, the sum of the repulsive forces generated by the relative velocity repulsive potential fields of all other non-interactive target users is calculated to obtain the relative velocity repulsive force sum, as shown in formula (8): (8) in, This represents the sum of the relative velocity repulsive forces currently acting on the user; when When the value is <0, the relative physical velocity and relative physical position of the dynamic obstacle relative to the current user are greater than 90 degrees. At this time, the user is moving away from the obstacle. Therefore, the motion information of the obstacle is not considered, and the user does not need to be subject to the relative velocity repulsion force generated by the obstacle. 2.3 Calculate the net repulsive force acting on the current user; The cumulative calculation of the relative positional repulsion force generated by static and dynamic obstacles and the relative velocity repulsion force generated by dynamic obstacles is shown in formula (9): (9) in, This represents the resultant force of the relative positional repulsion and the relative velocity repulsion experienced by the current user. This represents the proportionality coefficient of relative velocity repulsion.
4. The multi-person redirection walking method based on dynamic alignment as described in claim 1, characterized in that: Step (3) specifically includes: In an artificial potential field, clustering methods are used to cluster other users besides the target user, eliminating the repulsive forces generated by other users near the interaction target in the physical environment, thereby eliminating the interference of these repulsive forces in aligning and guiding the user. This includes: using a distance-based K-means clustering method to divide all users in the physical environment into different clusters, where the scan length of the K-means method is 2, and the minimum number of users in each cluster is 1; selecting the cluster where the interaction target is located, and treating other users in that cluster besides the interaction target as interference terms, and excluding them when calculating the repulsive force.
5. The multi-person redirection walking method based on dynamic alignment as described in claim 1, characterized in that: Step (4) specifically includes: Establish alignment guidance to help users and interaction targets interact, as shown in formula (10): (10) in, Represents alignment guidance. Represents the alignment guiding force vector. The value is a constant. about The direction is determined by classification and discussion. If the angle between the line connecting the relative physical velocity direction and the relative position of the user and the interactive target is greater than 90°, or if the virtual distance between the user and the interactive target is less than the physical distance between them, then... The direction is from the user's physical location to the physical location of the interaction target, aiming to enable the user to adjust to be closer to the interaction target in the fastest way in the physical environment, so as to match the scenario of the user approaching the interaction target in the virtual environment; in other cases, the problem is transformed into solving for the optimal solution, and thus obtaining... The direction; to align the relative physical position and relative physical velocity of the user and the interactive target in the physical environment with the relative virtual position and relative virtual velocity of the user and the interactive target in the virtual environment; The relative virtual speed and relative virtual position relationship between the current user and the interactive target in the virtual environment can be summarized as follows: As shown in formula (11): (11) in, The virtual velocity vector, Representing the current user and The virtual straight-line distance between them The virtual angle between the relative virtual speed and relative virtual position of the current user; The optimal relative physical velocity and relative physical position relationship between the user and the interaction target in the physical environment can be summarized as follows: As shown in formulas (12) and (13): (12) (13) in, This represents the optimal physical velocity vector for the user that needs to be determined. The physical velocity vector, Representing the current user and The physical straight-line distance between them The physical angle between the current user's relative physical speed and relative physical position; Next, the problem is transformed into solving how to obtain an optimal physical velocity, as shown in formula (14): (14) Alignment of the mapping of the relative positions and relative velocities of users and interaction targets in the physical and virtual environments, i.e. and The difference is minimized, as shown in formula (15). Finally, the optimal physical velocity direction for the user is defined as... Direction: (15) 6. The multi-person redirection walking method based on dynamic alignment as described in claim 1, characterized in that: Step (5) specifically includes: dynamically calculating the resultant force of the repulsive force and the alignment guiding force experienced by the user based on the relative position of the user and the interaction target in the virtual environment and the physical environment, as shown in formula (16): (16) in, Defined as the combined force experienced by the user. Represents the repulsive force proportionality coefficient. This represents the side length of the square physical environment.
7. The multi-person redirection walking method based on dynamic alignment as described in claim 1, characterized in that: Step (6) specifically includes: guiding the user to take a step forward in the physical space according to the direction of the resultant force vector.
8. A multi-person redirection walking device based on dynamic alignment, characterized in that, The device includes a memory and one or more processors, wherein the memory stores executable code, and the one or more processors execute the executable code to implement the multi-person redirection walking method based on dynamic alignment as described in any one of claims 1-7.
9. A computer-readable storage medium, characterized in that, It stores a program that, when executed by a processor, implements the multi-person redirection walking method based on dynamic alignment as described in any one of claims 1-6.
Citation Information
Patent Citations
Large-scale virtual scene roaming method based on multi-person redirection
CN112150634A
Multi-user redirection walking method based on optimal mutual benefit collision avoidance
CN116755551A
