KR102054713B1 - Collision avoidance method - Google Patents

Abstract

The present invention relates to a method for notifying a user of a collision risk within an activity radius of a virtual environment player and to a method for notifying a user of the collision risk within an activity radius of a virtual environment player, which estimates a future movement path using movement direction, movement speed, and start point (current position) of a player to notify a set of the possibility of a collision with an adjacent player. To this end, disclosed is a method for notifying the user of the collision risk within the activity radius of the virtual environment player, comprising the steps of: receiving a two dimensional absolute coordinate value of the player from a reference position transmission unit installed in an activity radius section of the player to store the received coordinate value at an absolute coordinate value time vector table in the order of a preset time interval based on the current time; calculating a movement directionality of the player based on the absolute coordinate value time vector table; calculating a probability value of the movement direction of the player in accordance with the movement directionality of the player at each of movement direction cases using the Markov probabilistic reasoning model, and primarily calculating an estimated path of the player in accordance with the movement direction probability value; calculating a relative speed coordinate value in accordance with the movement speed of the player by comparing a current coordinate of the player with a past coordinate thereof, and setting the latest relative speed coordinate value as a weighting factor; and adding the weighting factor to the movement direction probability value to calculate a final movement direction probability value, and secondarily calculating an estimated path of the player in accordance with the final movement direction probability value.

Description

Collision avoidance method of player in virtual environment {Collision avoidance method}

The present invention relates to a collision risk notification method of a player in a virtual environment, and more particularly, to predict a future moving path using a player's moving direction, a moving speed, and a starting point (the current position). It is about a collision risk notification method of a player in a virtual environment.

When each player experiences the virtual experience in the conventional virtual environment, various sensors are used to predict the movement of the player within a predetermined activity radius. In this virtual environment, the collision risk is notified based on the data values measured by various sensors when the collision risk between players. At this time, it is difficult to measure the collision risk between the players in various directions with the front sensor, the optical sensor has a problem that the tracking loss of the player is generated by the overlapping moving line to generate a shadow area.

In addition, when a large number of people experience a virtual experience within the same playground (Playground), the probability of a safety accident is increased by blocking the view of the player due to wearing the HMD helmet, collision by the existing sensor as described above It is difficult to cope with the situation.

Republic of Korea Patent Application Publication No. 10-2013-0095904 (Invention name: virtual environment management system and its server) Republic of Korea Patent Application Publication No. 10-2016-0080064 (Name of the invention: virtual sensor of the virtual environment)

Accordingly, an object of the present invention is to provide an invention capable of predicting a future movement path based on a moving direction, a moving speed, and a starting point (current position) of a player, which is created to solve the above problems.

However, the objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned will be clearly understood by those skilled in the art from the following description.

The above object of the present invention is to set a reference position that can generate the absolute coordinate value of the player by matching the measurement space and the virtual environment space of the reference position transmitter disposed in the active radius area of the player, the direction of movement of the player Calculating a player's prediction path using the movement speed and the current position, and informing a collision risk according to the calculation of the prediction path of each player. By providing.

The calculating of the player's prediction path may include: receiving the player's two-dimensional absolute coordinates from the reference position transmitter and storing the player's two-dimensional absolute coordinates in an absolute time value vector table based on a current time interval; Calculating a player's moving direction based on the temporal vector table; calculating a player's moving direction probability value according to the player's moving direction by the number of moving direction cases using a probability inference model, and predicting the player according to the moving direction probability value. Computing the path primarily, calculating the relative velocity coordinate value according to the player's movement speed by comparing the player's current coordinate and the past coordinate, calculating the player's moving direction based on the relative velocity coordinate value, Setting the most recent value of the velocity coordinate value as the weight, and the weighting direction Ryulgap applied to calculating the final moving direction and the probability value, and a step of calculating a predicted course of the player in accordance with the final direction of movement probability secondarily.

The absolute coordinate value time vector table also includes a player's current absolute coordinate value, a first past absolute coordinate value, and a second past absolute coordinate value.

The calculating of the moving direction of the player may include: comparing the second past absolute coordinate value and the first past absolute coordinate value, respectively, and determining whether to increase or decrease the first past absolute coordinate value and the current absolute coordinate value. Determining whether the increase or decrease is performed, and calculating the moving direction in two cases according to the increase or decrease determination result.

In addition, the number of movement direction cases is made up of four movement direction cases on the assumption that the player moves in one of right, left, forward, and backward directions based on the movement directions in two cases.

In addition, the relative velocity coordinate value is composed of a plurality of relative direction vector values sequentially calculated according to time, the difference between the player's current coordinate value and the plurality of past coordinate values.

In addition, calculating the moving direction of the player based on the relative velocity coordinate value, comparing the plurality of relative direction vector values with each other to determine whether to increase or decrease, calculating the moving direction of the player according to the increase or decrease determination result And calculating the direction of movement of the player using the most recent values of the plurality of relative direction vector values.

The final moving direction probability value is calculated by applying a weight to each of the four moving direction cases.

In addition, the final moving direction probability value is calculated by applying a + weight value in the same direction as the player's moving direction calculated on the basis of the relative velocity coordinate value, and applying a -weighting value in a direction different from the player's moving direction. .

And calculating the player's movement prediction absolute coordinate value vector table by applying the most recent value of the relative velocity coordinate value to the player's current absolute coordinate value, based on each player's movement prediction absolute coordinate value vector table. Calculating a distance value between each player, and notifying a possibility of collision with the corresponding player when the calculated distance value is within a range of a predetermined value.

Further, the motion prediction absolute coordinate value time vector table is formed of future predicted coordinate values by sequentially adding the most recent value of the relative velocity coordinate value to the player's current absolute coordinate value in sequential order.

According to the present invention as described above it is possible to accurately predict the future movement path of the player only by the reference sensor disposed in the player's active area.

The following drawings, which are attached to this specification, illustrate one preferred embodiment of the present invention, and together with the detailed description thereof, serve to further understand the technical spirit of the present invention. It should not be construed as limited.
1 is a view showing that the reference sensor is arranged in four directions within the virtual environment space according to an embodiment of the present invention, and the measurement space of the virtual environment space and the reference sensor is matched.
2 and 3 are diagrams for deriving the Markov model probability according to an embodiment of the present invention,
4 is a diagram for deriving a detailed moving direction of a player according to an embodiment of the present invention;
5 is a diagram for calculating a distance from an adjacent player according to an embodiment of the present invention.

Hereinafter, with reference to the drawings will be described a preferred embodiment of the present invention. In addition, one Example described below does not unduly limit the content of this invention described in the Claim, and the whole structure demonstrated by this Embodiment is not necessarily required as a solution of this invention. In addition, the matters obvious to those skilled in the art and the art may be omitted, and the description of the omitted elements (methods) and functions may be sufficiently referred to without departing from the spirit of the present invention.

The collision risk notification method in the activity radius of the virtual environment player according to an embodiment of the present invention is sequentially performed as follows. First, the present invention is a device for notifying the collision through the location tracking of a HMD (a type of helmet worn by a user) using a UWB (Ultra Wide Band) in a multi-access virtual environment and a player's movement path prediction.

The present invention predicts the next moving position of the player using the direction and the speed which are the moving information of the past moving object (or the player). The position is estimated using the direction, velocity, and starting point (current position) of the most basic vector information. The prediction accuracy is calculated by grasping the degree of error in consideration of the prediction value and the distance direction to the actual value when the position was acquired in the past. The input value receives two-dimensional vector data of UWB, and extracts the predicted path using the past motion vector using the Markov model. To increase the accuracy of the predicted path, the player's velocity data is added and weighted to reduce the error of the predicted path.

Meanwhile, UWB is a reference sensor that is pre-arranged in the player's activity space. The reference sensor transmits the beacon signal, and checks the beacon signal arrival time in conjunction with the server and the HMD helmet to calculate the current position coordinates of each player.

First, as shown in FIG. 1, the matching between the measurement space of the reference sensor (or UWB) and the virtual environment space is set. Based on the absolute coordinates of the reference sensor, when the player is positioned at the coordinate (0,0), the X axis increases, the R decreases, and the L, Y increases. The decreasing side is defined as B.

Next, the server receives the player's two-dimensional absolute coordinate value from the reference sensor (or reference position transmitter) installed in the player's active radius zone, and stores the absolute two-dimensional absolute coordinate value based on the present time in the absolute coordinate time vector table. do. That is, for example, when storing the absolute coordinate value of the player every 0.5 seconds, up to (-1 second, -0.5 second, 0 second) data based on the current time is stored in the time vector table. Of course, this absolute coordinate value vector table is constantly updated every 0.5. The absolute coordinate value time vector table of a player is defined as pa1 [(0.5, 1.0), (1.3, 1.2), (1.0, 2.1)] as an example and will be described based on this. The absolute coordinate value time vector table is a coordinate value generated under the matching of FIG. 1.

Next, the player's moving direction is calculated based on the absolute coordinate time vector table. The player's direction of movement is defined as R for increasing X axis, L for decreasing X axis, F for increasing Y, and B for decreasing Y, so pa1 [(R, F), ( R, F), (L, F)]. That is, the X axis is "R" because it increased from 0.5-> 1.3 and "L" because it decreased to 1.3-> 1.0. In addition, since the Y-axis increased from 1.0-> 1.2-> 2.1, it is "F". However, the immediately preceding data of the absolute coordinate time vector table is not defined, so pal [(R, F)] is arbitrary data for explanation. To calculate the direction of movement, pa1 [(R, F), (R, F), (L, F)] is [(R) | (R) | (L), (F) | (F) | (F )], And based on this, the number of cases in the moving direction is calculated as follows.

Next, by using the Markov probability inference model, the player's moving direction probability value according to the player's moving direction is calculated for each number of moving direction cases, and the player's prediction path according to the moving direction probability value is first calculated. First, the number of cases in the moving direction can be divided as follows based on [(R) | (R) | (L), (F) | (F) | (F)].

1) (R) | (R) | (L)-> (L)

2) (R) | (R) | (L)-> (R)

3) (F) | (F) | (F)-> (F)

4) (F) | (F) | (F)-> (B)

The number of cases in the moving direction can be divided into four as mentioned above, and the probability of moving direction must be calculated for each number of cases in the moving direction. Before calculating the direction probability value, the Markov model probability must be derived. The Markov model probability derivation is Markov model probabilities are derived from per1 (F1, F2, B1, B2) and per1-1 (L1, L2, R1, R2). per1 (F1, F2, B1, B2) is a vector table that stores the probabilities before and after every 120 seconds as an example, and per1-1 (L1, L2, R1, R2) is an example of right and left updating every 120 seconds. Vector table for storing probabilities. This will be described with reference to FIGS. 2 and 3 to derive the Markov model probability.

First, referring to FIG. 2, the derivation data is derived as follows.

Historical data Current data Derived data F F F1 F B B1 B F F2 B B B2 L L L1 L R R1 R L L2 R R R2

When the above table is described, it is R2 when (R)-> (R) and F1 when (F)-> (F). An example for deriving Markov model probabilities through the table will be described below. For example, in the case of (R)-> (L)-> (R), (R)-> (L) * (L)-> (R) = L2 * R1. In the case of (L)-> (R)-> (L), (L)-> (R) * (R)-> (L) = R1 * L2.

Derived data formulation method is to define the value established for 120 seconds as a probability, and the method to obtain this method receives the absolute coordinate value every 0.5 seconds to calculate the relative coordinate value. In this case, the relative coordinate value is represented by R when the X axis increases, L when the X axis increases, F when the Y axis increases, and B when the Y axis increases.

Referring to FIG. 3, as an example, if the data of 0.5 second ago is (1.0,1.2) and the current data is (1.2,1.5), the X axis increases by 0.2 and the Y axis increases by 0.3, so that [(R) (F)]. After 0.5 second again, the data before 0.5 second is (1.2,1.5) and the current data is (1.3,1.2), so the X-axis increases by 0.1 and the Y-axis decreases by 0.3 to [(R) (B)]. Therefore, the X-axis becomes [(R)-> (R)] and adds 1 to R2. And Y-axis becomes [(F)-> (B)] and adds 1 to B1. In addition, if 0.5 second passes and becomes (1.3,1.2) and the current data is (1.2,1.0), the X axis decreases by 0.1 and the Y axis decreases by 0.2, [(L) (B)]. Compared with (B), (R)-> (L) adds 1 to L2 and (B)-> (B) adds 1 to B2. Divide the total number of times obtained by this addition. That is, since it is obtained every 0.5 seconds for 120 seconds, the total number of times is divided by 240 to derive the Markov model probability.

The Markov model probability value is derived through the calculation method as described above, and it is assumed that per1 (0.8, 0.3, 0.2, 0.7) and per1-1 (0.45, 0.6, 0.55, 0.4) for the purpose of the present invention.

Matrix representation per1 matrix
(F1 B1
F2 B2) per1-1 matrix
(L1 R1
L2 R2) (0.8 0.2
0.3 0.7) (0.45 0.55
0.6 0.4)

Therefore, with reference to the hypothesized Markov model probability values per1 and per1-1, the movement direction probability values for each number of cases in the movement direction are as follows.

1) (R) | (R) | (L)-> (L) = 0.4 * 0.6 * 0.45 = 0.108

2) (R) | (R) | (L)-> (R) = 0.4 * 0.6 * 0.55 = 0.132

3) (F) | (F) | (F)-> (F) = 0.8 * 0.8 * 0.8 = 0.512

4) (F) | (F) | (F)-> (B) = 0.8 * 0.8 * 0.2 = 0.128

The player's next prediction path is determined as [(R, F)], but if the player's historical data is insufficient, there may be an error. Increase To increase the accuracy of the prediction path, the following steps are taken.

Next, the relative speed coordinate value according to the player's moving speed is calculated by comparing the current coordinate with the past coordinate of the player, and the most recent value of the relative speed coordinate value is set as the weight. Based on the above-described embodiment, suppose pa1 [(0.5,1.0), (1.3,1.2), (1.0,2.1)] and set the player's real time position in the unit of 0.1 seconds as Van = (Va1, Va2, Va3, Save as Va4). Based on p1, Va1 current speed is set as (vc5, vy5), and (vx1, vy1) is -0.4 seconds before (vx2, vy2) -0.3 seconds before (vx3, vy3) -0.2 seconds before, (vx4 , vy4) -0.1 seconds ago, the position from the past to the present in chronological order is stored in the vector table.

 Measure the velocity (relative direction vector) of Va1 ((vx1, vy1), (vx2, vy2), (vx3, vy3), (vx4, vy4), (vx5, vy5)). As an example, the player's absolute coordinates are every 0.1 seconds: Va1 [(1.2,1.1), (1.3,1.2), (1.4,1.1), (1.2,1.5), (1.1,1.9), (1.0,2.1) ], V1 [(0.1,0.1), (0.1, -0.1), (-0.2,0.4), (-0.1,0.4), (-0.1,0.2)]], and then V1 [(R, F), (R, B), (L, F), (L, F), (L, F)]. The most recent movement direction is "[(-0.1,0.2)]" and therefore "[(L, F)]". Since the player is currently heading towards L and F, the player weights + a (-0.1,0.2) to the probability of the expected direction (L) (F) and calculates the detailed direction.

Calculate the detailed travel direction of [(L) (F)] ((F, B) a = tanΘ) = x (| -0.1 |) / y (| 0.2 |) as shown in FIG. It is calculated that the axis moves in the direction of 26 degrees. Accordingly, the current relative velocity value (relative direction vector) is added as a probability (by adding the current velocity to the probability, the velocity in the current traveling direction is an immediate process, and thus the velocity is increased as if the probability is increased). .

Next, the above-described weight is added to the movement direction probability value to calculate the final movement direction probability value, and the player's prediction path according to the final movement direction probability value is secondarily calculated. In more detail, finally, the final moving direction probability value is calculated for each number of moving direction cases, and weighting according to the player's speed is applied as follows. In the above description, the direction corresponding to the speed is "[(L, F)]" and "[(-0.1,0.2)]". Therefore, L is +0.1, R is -0.1, F is +0.2, and B is -0.2.

1) (R) | (R) | (L)-> (L) = 0.4 * 0.6 * 0.45 = 0.108 + 0.1 = 0.208

2) (R) | (R) | (L)-> (R) = 0.4 * 0.6 * 0.55 = 0.132-0.1 = 0.032

3) (F) | (F) | (F)-> (F) = 0.8 * 0.8 * 0.8 = 0.512 + 0.2 = 0.712

4) (F) | (F) | (F)-> (B) = 0.8 * 0.8 * 0.2 = 0.128-0.2 = -0.072

Therefore, it can be calculated that the prediction path of the final player is -26 degrees in the (L, F) direction on the (F, B) axis. The prediction path calculation of the player as described above can be applied to all players playing in the virtual environment.

Next, perform the following steps to calculate the risk of collision between adjacent players. A player's movement prediction absolute coordinate value time vector table is calculated by adding the most recent value of the relative velocity coordinate value to the player's current absolute coordinate value in chronological order. That is, the most recent value of the relative velocity coordinate value of the player is "[(-0.1,0.2)]" as before, and the future value is continuously added by adding "[(-0.1,0.2)]" to the player's current absolute coordinate value. The absolute predicted coordinates of are stored in the time vector table. The player's movement prediction absolute coordinate value The time vector table is the future predicted absolute coordinate value based on the current absolute coordinate value. Therefore, if the player's current absolute coordinate value is (1.0,2.1), then the "[(-0.1,0.2)]" value is continuously added to obtain 30 future moving predictive absolute coordinate values as follows. Here, the "30" vector tables are for acquiring up to 3 seconds of data every 0.1 from the present time to the future, and may be changed as needed as needed.

PR1 [30] = {(1.0,2.1), (0.9,2.3), (0.8,2.5) .......}

Therefore, the motion prediction absolute coordinate value time vector tables of the players 1,2,3,4 can be obtained as PR2 [30], PR3 [30], and PR4 [30], respectively.

Next, the distance to the adjacent players is calculated with reference to FIG. 5 based on the obtained motion prediction absolute coordinate value time vector tables PR1 [30], PR2 [30], PR3 [30], and PR4 [30], respectively. As an example, the distance between player 1 and player 2 is calculated as follows.

distance (distance) = (PR1 (x1, y1), PR2 [i] (x1, y1)) = sqrt ((PR1 (x1) -PR2 (x1)) ^ 2 + (PR1 (y1)-PR2 (y1) )) ^ 2)

The above-described distance value is an expression for calculating distance values according to the current coordinate values of player 1 and player 2, and data after 0.1 second from the current coordinate value can be calculated using the above equation, and the total distance value of 30 times Can be found between Player 1 and Player 2.

Next, when any one or many of the calculated 30 distance values of the first player and the second player are within a preset range, the possibility of collision with the player is notified. As an example, if the distance value between the player 1 and the player 2 after 0.3 seconds is a risk of collision prediction, it is notified that "possible collision with the player after 0.3".

The calculation and control of the data described above may be performed by a server or a controller disposed in a virtual environment.

In the following description of the present invention, those skilled in the art and those skilled in the art may omit descriptions, and descriptions of such omitted components (methods) and functions may be sufficiently referred to without departing from the technical spirit of the present invention. Could be.

Description of the configuration and functions of the above-described parts have been described separately from each other for convenience of description, any one configuration and function may be implemented by integrating into other components, or may be implemented more subdivided as necessary.

As mentioned above, although demonstrated with reference to one Embodiment of this invention, this invention is not limited to this, A various deformation | transformation and an application are possible. That is, those skilled in the art will readily appreciate that many modifications are possible without departing from the spirit of the invention. In addition, when it is determined that the detailed description of the known function and its configuration or the coupling relationship for each configuration of the present invention may unnecessarily obscure the subject matter of the present invention, it should be noted that the detailed description is omitted. something to do.

Claims (11)

Setting a reference position for generating an absolute coordinate value of the player by matching the measurement space of the reference position transmitter and the virtual environment space arranged in the activity radius area of the player,
Calculating a player's prediction path using the player's moving direction, moving speed, and current position, and
Informing the collision risk according to each player's prediction path calculation,
Computing the prediction path of the player,
Receiving the two-dimensional absolute coordinate values of the player from the reference position transmitter and storing the two-dimensional absolute coordinate values in an absolute time value vector table based on a present time interval;
Calculating a moving direction of the player based on the absolute coordinate time vector table;
Calculating a probability value of the player's movement direction according to the movement direction of the player for each number of cases in the movement direction using a probability inference model, and primarily calculating a player's prediction path according to the movement direction probability value;
The relative speed coordinate value according to the moving speed of the player is calculated by comparing the current coordinate with the past coordinate of the player, the direction of movement of the player is calculated based on the relative speed coordinate value, and the most recent value of the relative speed coordinate value. Setting the weight to, and
Calculating a final moving direction probability value by applying the weight to the moving direction probability value, and secondly calculating a player's prediction path based on the final moving direction probability value. Notification method.
delete The method of claim 1,
The absolute coordinate value time vector table,
A collision risk notification method of a player in a virtual environment, comprising a player's current absolute coordinate value, a first past absolute coordinate value, and a second past absolute coordinate value.
The method of claim 3, wherein
Calculating the moving direction of the player,
Comparing the second past absolute coordinate values with the first past absolute coordinate values, respectively, and determining whether to increase or decrease them;
Comparing the first past absolute coordinate value and the present absolute coordinate value to determine whether to increase or decrease each; and
And calculating a moving direction in two cases according to the result of the determination of increase or decrease.
The method of claim 4, wherein
The number of cases in the moving direction is
A collision risk notification method of a player in a virtual environment, characterized in that the number of the four moving direction assuming that the player moves in any one of the right, left, forward, backward based on the movement direction of the two cases. .
The method of claim 5,
The relative velocity coordinate value is,
A collision risk notification method for a player in a virtual environment, comprising a plurality of relative direction vector values sequentially calculated according to time, the difference between a player's current coordinate value and a plurality of past coordinate values.
The method of claim 6,
Calculating a moving direction of the player based on the relative speed coordinate value,
Comparing the plurality of relative direction vector values with each other to determine whether there is increase or decrease;
Calculating the moving direction of the player according to the result of the increase or decrease judgment; and
And calculating a moving direction angle of the player using the most recent values of the plurality of relative direction vector values.
The method of claim 7, wherein
The final moving direction probability value,
The collision risk notification method of a player in a virtual environment, characterized in that calculated by applying a weight to each of the number of the four moving direction case.
The method of claim 8,
The final moving direction probability value,
Under the virtual environment, it is calculated by applying a + weight value in the same direction as the player's moving direction calculated based on the relative velocity coordinate value, and applying a -weight value in a direction different from the player's moving direction. How to notify a player about a collision risk.
The method of claim 1,
Calculating a player's movement prediction absolute coordinate value time vector table by applying the most recent value of the relative velocity coordinate value to the player's current absolute coordinate value,
Calculating a distance value between each player based on the moving prediction absolute coordinate value time vector table of each player, and
And informing the possibility of collision with the corresponding player when the calculated distance value is within a range of a predetermined value.
The method of claim 10,
The motion prediction absolute coordinate value time vector table,
Collision risk notification method of the player in a virtual environment, characterized in that the future prediction coordinates by adding the most recent value of the relative velocity coordinate value in sequential order to the current absolute coordinate value of the player.

Images