JP2020027344A - Collision avoidance support device - Google Patents

Abstract

To provide a collision avoidance support device which makes return steering possible in addition to collision avoidance steering.SOLUTION: A preference degree calculation section 24 defines a space combining an option of a course change angle and an option of a speed change rate from the present state as a collision avoidance steering space and calculates, for each of positions in the collision avoidance steering space, a preference degree representing a subjective preference of a navigator with a selection of the position. A risk calculation section 26 calculates, for each of positions in the collision avoidance steering space, the risk of collision to the other ship in the case where an own ship performs collision avoidance steering, and a utility value calculation section 20 calculates a utility value by subtracting the collision risk from the preference degree. The preference degree calculation section 24 calculates a preference degree having a first peak value at a position in the collision avoidance steering space combining a course change angle to a way point and a speed change rate to a predetermined speed and having a second peak value at a position in the collision avoidance steering space combining a course change angle of original course maintenance and a speed change rate of original speed maintenance.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an escape assist device, for example, to a technique for calculating an optimal ship maneuvering method for avoiding collision with another ship by calculation.

  Non-Patent Literatures 1 and 2 disclose a system in which the current preference of a ship operator is represented by the utility value of the escape navigation space. The avoidance maneuvering space is a space in which the course change angle of the course and the speed change rate are combined, and the utility value is calculated for each position in this space. The utility value is obtained by subtracting an objective collision risk with respect to another ship according to the change in the course and the speed from the subjective preference of the operator due to the change in the course and the speed.

  Patent Document 1 discloses an evacuation assistance device that presents prediction information necessary for evacuation determination to a boat operator. The evacuation assistance device calculates the collision risk and the degree of blockage for each position on a plurality of routes that the ship can select in the future from the relationship with the predicted position of the other ship in the future, and calculates the degree of blockage for each position. Is displayed on the screen.

JP 9-22500 A

Nagasawa and three others, "The degree of difficulty in the avoidance maneuvering environment-II.-For evaluation by simulation-", Proceedings of the Nautical Society of Japan, 88, 1993, p. 137-144 Nagasawa, and three others, "Assessment of High Speed Navigation in a Congested Area by the Traffic Simulation", FAST '93, Vol. 2 (1993), p. 1349-1357

  For example, as a device for safely navigating a ship or saving manpower on a ship, an evacuation support device as disclosed in Patent Document 1 is known. The evacuation support device can play one of the main technologies required for autonomous navigation of a ship. Here, if the techniques of Non-Patent Documents 1 and 2 and Patent Document 1 are used, it is possible to automatically start an optimal avoiding maneuver. However, the technique does not particularly consider return ship maneuvering when returning to the original route after performing the escape ship maneuver. In particular, when considering application to autonomous navigation of a ship, some mechanism for performing return ship maneuver is required.

  The present invention has been made in view of such circumstances, and one of its objects is to provide an escape assist device that enables return ship maneuvering in addition to avoiding ship maneuvering.

  The above and other objects and novel features of the present invention will become apparent from the description of the present specification and the accompanying drawings.

  The outline of a typical embodiment of the invention disclosed in the present application will be briefly described as follows.

  The evacuation assistance device according to one embodiment includes a preference calculation unit, a risk calculation unit, and a utility value calculation unit. The preference calculation unit determines, for each position in the avoidance maneuvering space, a space obtained by combining options for changing the turning angle and options for changing the speed from the current state, and for each position in the avoiding maneuvering space, the ship operator's subjective Calculates the preference level that represents the favorite. The risk calculation unit calculates, for each position in the avoiding maneuvering space, the collision risk with respect to another ship when the own ship performs the avoiding maneuvering. The utility value calculation unit calculates a utility value by subtracting the collision risk from the preference for each position in the avoidance maneuvering space. Here, the preference degree calculation unit has a first peak value at a position in the avoidance maneuvering space obtained by combining the change angle to the way point and the speed change rate to the predetermined speed, and the change in the original course and the change in the original course. The preference having the second peak value at the position of the avoidance maneuvering space in combination with the speed change rate for maintaining the speed is calculated.

  According to the one embodiment, return ship maneuvering can be realized in addition to avoidance maneuvering.

It is a schematic diagram showing an example of composition around an escape assist device by one embodiment of the present invention. FIG. 2 is a diagram illustrating an example of a processing result of a preference degree calculation unit in FIG. 1. FIG. 3 is a diagram showing a relationship between a changing hand angle and a preference degree when the gear ratio is constant in FIG. 2. FIG. 3 is a diagram showing a relationship between a changing hand angle and a preference degree when the gear ratio is constant in FIG. 2. FIG. 2 is a diagram illustrating an example of a situation after the own ship has avoided another ship by the avoidance maneuver in the evacuation support device of FIG. 1. FIG. 6 is a diagram illustrating an example of a processing result of a utility value calculating unit in the situation of FIG. 5. FIG. 2 is a flowchart illustrating an example of a schematic processing content of a main part in the evacuation assistance device of FIG. 1. FIG. 2 is a schematic diagram showing an operation example when performing the evacuating maneuvering and returning maneuvering in the evacuation assisting device of FIG. 1. It is a figure showing an example of a situation where own ship and another ship may collide. It is a figure showing an example of the processing result of the preference calculation part in the escape assist device which is a comparative example of the present invention. FIG. 2 is a diagram illustrating an example of processing contents of a risk calculating unit in the evacuation assistance device of FIG. 1. FIG. 12 is a supplementary diagram illustrating a bumper area in FIG. 11. FIG. 4 is a diagram illustrating an example of a processing result of a risk calculator in the evacuation assistance device of FIG. 1. FIG. 4 is a diagram illustrating an example of a processing result of a utility value calculation unit in the evacuation assistance device of FIG. 1. FIG. 6 is a diagram illustrating an example of a situation after the own ship has avoided another ship by the avoidance maneuver in the escape assist device according to the comparative example of the present invention. FIG. 16 is a diagram illustrating an example of a processing result of the risk calculating unit in the situation of FIG. FIG. 16 is a diagram illustrating an example of a processing result of the utility value calculating unit in the situation of FIG. 15.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In all the drawings for describing the embodiments, the same members are denoted by the same reference numerals in principle, and the repeated description thereof will be omitted.

《Outline around the escape assist device》
FIG. 1 is a schematic diagram showing a configuration example around an escape assist device according to an embodiment of the present invention. In FIG. 1, the ship information acquisition unit 11 uses various mechanisms to acquire information (own course, speed, and position) of own ship and information of one or more other ships existing within a predetermined range around the own ship ( Course, speed and position). The various mechanisms are typically one or a combination of an AIS (Automatic Identification System) 30, a radar 31, a camera 32, a direction sensor 35, a speed sensor 36, a GPS (Global Positioning System) 37, and the like.

  The AIS is a system for acquiring information on another ship or own ship by exchanging ship information such as the position, course, speed, and destination of the ship between the ships or between the ship and the land by wireless communication. The radar 31 and the camera 32 acquire information on other ships by detecting the relative course, speed, and position of each other ship existing around the own ship. The direction sensor 35 is a gyrocompass or the like, and detects a course of the own ship. The speed sensor 36 is an electromagnetic log, a Doppler log, or the like, and detects the speed of the own ship. The GPS 37 detects the position of the own ship.

  The evacuation assistance device 10 includes a utility value calculation unit 20, a route setting unit 22, and a marine vessel maneuvering instruction unit 23. The utility value calculator 20 includes a preference degree calculator 24 and a risk degree calculator 26. The evacuation assistance device 10 is realized by, for example, program processing using a computer system including a CPU (Central Processing Unit). However, the present invention is not necessarily limited to this. For example, a part of the escape assist device 10 may be realized by dedicated hardware.

  The route setting unit 22 stores a planned route on an electronic chart set in advance by a ship operator or the like. The scheduled route is a route connecting the departure point and the target point via a plurality of way points (WP), and the way point (WP) is typically a turning point. The electronic chart is stored in a nonvolatile storage device 14 such as a hard disk. Further, the route setting unit 22 notifies the preference degree calculation unit 24 of information (for example, a position or a course) of the waypoint (WP) based on the planned route stored in advance.

  The danger calculation unit 26 sets, for each position in the avoidance maneuvering space, the collision risk with respect to another ship when the own ship performs the avoidance maneuvering around the reference ship which is one of the own ship and the other ship. Is calculated using an elliptical bumper area (in other words, an exclusive area). The avoidance maneuvering space is a space in which options for changing the turning angle and options for changing the gear ratio are combined from the current state, and represent options (in other words, troubles) that can be taken when performing the avoidance maneuvering.

  The preference degree calculation unit 24 calculates, for each position in the avoidance maneuvering space, a preference degree representing a subjective preference of the ship operator accompanying the selection of each position. At this time, the preference degree calculation unit 24 calculates a preference degree having a peak value at a position for a way point (WP) in the avoiding and maneuvering space. The position for the way point (WP) is a position obtained by combining the changing angle to the way point (WP) and the speed change rate to a predetermined speed (for example, the original speed). In addition, the preference calculation unit 24 calculates a preference having a peak value at a position for maintaining the current state in the avoidance maneuvering space. The position for maintaining the current state is a position in which the changing course angle for maintaining the original course and the gear ratio for maintaining the original speed are combined. As described above, the preference degree calculation unit 24 calculates a preference degree having two peak values.

  The utility value calculating unit 20 calculates the utility value by subtracting the collision risk by the risk calculating unit 26 from the preference by the preference calculating unit 24 for each position in the avoidance maneuvering space. The ship maneuvering instruction unit 23 recognizes the position of the avoidance maneuvering space where the utility value from the utility value calculation unit 20 is maximum, and instructs the automatic steering device 15 to change the stylus angle and gear ratio corresponding to the position as an optimal ship maneuvering method. I do.

  The autopilot device 15 includes an autopilot that automatically controls the rudder to change the course toward the instructed turning angle, a device that automatically controls the engine output to shift to the instructed gear ratio, and the like. The autopilot device 15 receives own ship information (current course, speed, and position) from the ship information obtaining unit 11 and scheduled route information from the route setting unit 22, and responds to an instruction from the ship steering instruction unit 23. The autonomous navigation is performed toward the target point while maneuvering the boat as appropriate. The result display device 13 is a display or the like, and displays, for example, a three-dimensional graph (utility value at each position in the avoidance maneuvering space) representing a calculation result of the utility value by the utility value calculating unit 20.

<< Details of each part of the escape support device (comparative example) >>
Here, prior to description of the evacuation assistance device of the embodiment, an evacuation assistance device as a comparative example will be described. The evacuation assistance device as a comparative example includes a preference calculation unit (referred to as 24 ') that is not notified of waypoint (WP) information in the configuration example of FIG. Unlike the preference calculation unit 24 of the embodiment, the preference calculation unit (24 ') calculates a preference having a peak value only at a position for maintaining the current state in the avoidance maneuvering space.

  FIG. 9 is a diagram illustrating an example of a situation where the own ship and another ship may collide. In the example of FIG. 9, the own ship OS is traveling at a predetermined speed toward the north, and the other ship TS is traveling at a predetermined speed toward the west. If the own ship OS and the other ship TS continue to advance as they are, there is a possibility that a collision occurs after a predetermined time (the closest time (Tcpa)). In this case, the own ship OS looking at the other ship TS to the right is obliged to avoid collision. In such a case, the own ship OS searches for an avoidance maneuvering method for efficiently avoiding a collision by using the escape assist device.

  FIG. 10 is a diagram illustrating an example of a processing result of the preference calculation unit in the evacuation assistance device according to the comparative example of the present invention. In FIG. 10, the horizontal axis is the turning angle [deg], and indicates the range from the left turning course “−60 [deg]” to the right turning course “60 [deg]” with the original course being 0 [deg]. The vertical axis indicates the transmission rate [%], and indicates the range from 0 [%] to 100 [%] with the original speed set to 100 [%]. The avoidance maneuvering space described above is a space indicated by the horizontal axis and the vertical axis, and is a space representing options for avoidance maneuvering.

On the other hand, the height axis indicates the degree of preference, and indicates that the preference increases as the value approaches “1.0” and decreases as the value approaches “0”. Usually, when there is no obstacle such as another ship or land area, the ship operator subjectively prefers to maintain the original course and the original speed, and prefers not to change the course and the speed as much as possible. The original course preference Pb _org (i, j) shown in FIG. 10 is determined by the arithmetic expression shown in Expression (1), reflecting such a subjective opinion of the boat operator.
Pb _org (i, j) = Pb _org (i, 0) × Pb _org (0, j) (1)

As shown in FIG. 10, the original course preference Pb _org (i, j) calculated by the equation (1) is a position for maintaining the current state (variation angle “0 [deg]”, speed change rate “100 [ %] ") Has a peak value PK2. The preference calculation unit (24 ′) calculates the original course preference Pb _org (i, j) of Expression (1) for each combination of the turning angle (i) and the speed change rate (j).

Here, each term in the equation (1) is determined by an exponential function shown in the changing hand preference Pb _org (i, 0) in the equation (2) and the shift preference Pb _org (0, j) in the equation (3). Can be The changing-point preference Pb _org (i, 0) represents the preference when the changing rate is changed from 0 [deg] to “ΔCo” when the transmission rate is set to 100 [%]. On the other hand, the shift preference Pb _org (0, j) represents the preference when the change rate is changed from 100 [%] by “ΔV” when the changing hand angle is 0 [deg]. A coefficient equation (2) and "a _c" and in the formula (3) "a _v" is set in advance, "a _c" may be set to different values in a time when the right veering and left veering .
_{Pb org (i, 0) =} exp (-a c × ΔCo) ... (2)
_{Pb org (0, j) =} exp (-a v × ΔV) ... (3)

  FIG. 11 is a diagram for explaining an example of the processing content of the risk calculating section in the evacuation assistance device of FIG. 1, and FIG. 12 is a supplementary diagram for explaining the bumper region in FIG. FIG. 13 is a diagram illustrating an example of a processing result of the risk calculating unit in the evacuation assistance device of FIG. 1. For example, in the case as shown in FIG. 9, the risk calculating section 26 generates a processing result as shown in FIG. In FIG. 13, the horizontal axis and the vertical axis are the same evacuated maneuvering spaces as in FIG. The height axis indicates the degree of collision risk, and indicates that the risk increases as the distance approaches “1.0” and decreases as the distance approaches “0”.

  The risk calculation unit 26 calculates the collision risk for each position in the avoidance maneuvering space in FIG. 13 using a method as shown in FIG. FIG. 11 shows a relative relationship between the own ship OS and the other ship TS. The own ship OS is located at coordinates (Xo, Yo) on the absolute axis, and travels at a predetermined speed to a predetermined course. On the other hand, the other ship TS is located at the coordinates (Xt, Yt) on the absolute axis, and proceeds to a predetermined course at the speed Vt.

  Here, one of the own ship OS and the other ship TS is a reference ship, and the other is a target ship. The position of the reference vessel is determined at the origin on the relative axis, the course of the reference vessel is determined on the Y axis on the relative axis, and the axis orthogonal thereto is defined on the X axis on the relative axis. In this example, the reference ship is the other ship TS and the target ship is the own ship OS. However, since there is a relative relationship, the target ship may be the other ship TS and the reference ship may be the own ship OS.

  On such a relative axis, the own ship OS, which is the target ship, advances with a relative speed Vr to a relative course 43 obtained by a composite vector of the vector of the speed Vo and the inverse vector of the speed Vt. The position, course, and speed of the own ship OS are determined by reflecting, in the own ship information from the ship information obtaining unit 11 in FIG. 1, the changing hand angle and the gear ratio based on the position of the avoidance maneuvering space. In addition, the position, course, and speed of the other vessel TS are determined based on the other vessel information from the vessel information acquisition unit 11 in FIG. 1 assuming that the other vessel TS maintains the course and the speed.

  Here, an elliptical bumper area (exclusive area) 40 is set around the other vessel TS as the reference vessel. FIG. 12 shows a bumper region 40 having a minor axis A and a major axis B. The size of the bumper region 40 is determined based on information on two ships, for example, the total length and speed of the own ship OS and the total length and speed of the other ship TS. The position of the other ship TS is set not at the center of the bumper area 40 but at a position shifted to the lower left by the distance C and the distance D from the center reflecting the traffic rules of the ship.

  Using such a bumper area 40, as shown in FIG. 11, risk functions 41 and 42 are defined so as to sequentially change from the maximum value to the minimum value from the position of the other ship TS toward the outer periphery of the bumper area 40. Is done. Specifically, the risk function 41 is for the Y axis (relative axis), and has a maximum value “1.0” at the origin coordinate of the Y axis (relative axis) which is the position of the other ship TS. The minimum value (for example, “0”) is obtained at the maximum and minimum coordinates of the Y axis (relative axis) where the outer periphery is located. Similarly, the risk function 42 is for the X axis (relative axis), has a maximum value of “1.0” at the origin coordinate of the X axis (relative axis), and has the X axis (relative axis) where the outer periphery of the bumper region 40 is located. ) Is the minimum value (for example, “0”) at the maximum coordinates and the minimum coordinates.

  The risk calculator 26 calculates the risk of collision by calculating the values of the risk functions 41 and 42 according to the position in the bumper area 40 through which the ship OS (target ship) passes in the future. Specifically, the risk calculating unit 26 calculates the value of the risk function 41 corresponding to the coordinates of the intersection with the Y axis (relative axis) when the own ship OS advances on the relative course 43, as the Y axis collision risk Ry. The value of the risk function 42 corresponding to the coordinates of the intersection with the X-axis (relative axis) is calculated as the X-axis collision risk Rx.

Then, as shown in equation (4), the risk degree calculation unit 26 performs a predetermined weighting on the larger value of the Y-axis collision risk degree Ry or the X-axis collision risk degree Rx, so that the collision risk degree R ( i, j) are calculated. In Equation (4), “Tcpa” is the closest time, and “Wtcpa” is a predetermined time. In equation (4), weighting is performed such that the shorter the closest time (Tcpa), the higher the collision risk R (i, j).
R (i, j) = max (Rx, Ry) × (1−Tcpa / Wtcpa) (4)

Actually, there is a case where q (where q is an integer of 2 or more) other TS exists in a predetermined range, instead of one other TS. Accordingly, the target other ship TS (that is, the other ship TS having the largest influence) may be different for each position in the avoidance maneuvering space in FIG. Therefore, in actuality, instead of the collision risk R (i, j) in Expression (4), the collision risk Rk (i, j) in Expression (5) is used. The collision risk Rk (i, j) of the equation (5) is calculated for each position in the avoidance maneuvering space by the collision risk R (i, j) with respect to the other ship TS having the largest influence among one or a plurality of other ships TS. j).

FIG. 14 is a diagram illustrating an example of a processing result of the utility value calculation unit in the evacuation assistance device of FIG. 1. In FIG. 14, the horizontal axis and the vertical axis are the same evacuated maneuvering space as in FIG. The height axis is a utility value, and indicates that the effect increases as the value approaches "1.0" and decreases as the value approaches "0". The utility value calculating section 20 calculates the risk level calculating section 26 from the preference level Pb (i, j) of the preference level calculating section for each position in the avoidance maneuvering space of FIG. The utility value u (i, j) is calculated by subtracting the collision risk Rk (i, j) in equation (5). Here, the preference degree Pb (i, j) corresponds to the original course preference degree Pb _org (i, j) of the equation (1) by the preference degree calculation unit (24 ′) in the evacuation support device as the comparative example. I do. Further, the collision risk Rk (i, j) is weighted with a predetermined coefficient α in detail.
u (i, j) = Pb (i, j) −α × Rk (i, j) (6)

  Along with such an operation, the utility value in FIG. 14 is obtained by subtracting the collision risk in FIG. 13 from the preference in FIG. The optimal avoiding and maneuvering method is a method in which the utility value u (i, j) is maximized. In the example of FIG. 14, the right turning of the steering angle 18 [deg] is performed while maintaining the speed. The ship maneuvering instruction section 23 in FIG. 1 instructs the automatic steering device 15 on the changing hand angle and the speed change rate corresponding to the position of the avoidance maneuvering space where the utility value is maximum as the optimum avoiding maneuvering method.

《Problems of the evacuation support device (comparative example)》
FIG. 15 is a diagram illustrating an example of a situation after the own ship has avoided another ship by the avoidance maneuver in the escape assist device according to the comparative example of the present invention. FIG. 16 is a diagram illustrating an example of a processing result of the risk calculating section in the situation of FIG. 15, and FIG. 17 is a diagram illustrating an example of a processing result of the utility value calculating section in the situation of FIG. 15.

  In FIG. 15, as a result of the own ship OS performing the avoidance maneuver, the passage of the other ship TS2 has been completed in front of the own ship OS. Since the original course of the own ship OS deviates from the course to the way point WP due to the evacuating maneuver, the own ship OS is originally at the stage where the passage of the other ship TS2 is completed (or a stage close thereto). Wants to return to the waypoint WP. However, the evacuation assistance device according to the comparative example does not have a mechanism for performing such a return ship maneuver, and the traveling to the original course is continued as it is.

  Therefore, for example, a system is conceivable in which a mechanism for separately starting the return maneuver when the collision risk in the avoidance maneuver space becomes zero is separately provided. However, as shown in FIG. 15, for example, when there is another ship TS <b> 1 that is proceeding substantially parallel to the own ship OS in front of the own ship OS, as shown in FIG. The collision risk associated with TS1 remains. As a result, in the situation as shown in FIG. 15, it is difficult for the own ship OS to perform the return maneuvering by the other ship TS1 although the ship OS can originally perform the return maneuvering regardless of the other ship TS1. Become. That is, the marine vessel maneuvering instruction unit 23 instructs the autopilot device 15 to maintain the current marine vessel maneuvering method corresponding to the peak value PK2 based on the result of FIG.

  Such a problem is not limited to the situation as shown in FIG. 15, and may similarly occur particularly in a sea area where the degree of congestion of a ship is high, such as in a bay. That is, in a sea area where the degree of congestion is high, the situation itself where the risk of collision becomes zero hardly occurs. On the other hand, for example, a method is also conceivable in which, when the degree of collision danger in a predetermined range of the avoidance maneuvering space becomes equal to or lower than a predetermined level, return maneuvering is started. However, it is not easy to find a proper condition at this time. Therefore, it is useful to realize return boat maneuvering using the evacuation support device 10 of the embodiment shown in FIG.

<< Details of each part of the escape support device (embodiment) >>
FIG. 2 is a diagram illustrating an example of a processing result of the preference degree calculation unit in FIG. As shown in FIG. 2, the preference degree calculation unit 24 of FIG. 1 performs, similarly to the case in FIG. The preference having the peak value PK2 at the position for maintaining the current state, which is the position obtained by combining the rate (ie, 100 [%]), is calculated. In addition, the preference degree calculation unit 24 calculates the preference having the peak value PK1 at the position for the way point WP, which is a position obtained by combining the change point angle Cwp to the way point WP and the speed change rate to the predetermined speed in the avoiding and maneuvering space. Calculate the degree. In this example, the predetermined speed is the original speed (speed change rate 100%). Here, the peak value PK1 is set to a value larger than the peak value PK2.

FIGS. 3 and 4 are diagrams showing the relationship between the changing hand angle and the preference when the gear ratio is constant in FIG. Here, a case where the gear ratio is 100 [%] (maintenance speed maintained) is exemplified. FIG. 3 shows the original course preference Pb _org (i, 0) and the waypoint WP preference Pb _wp (i, 0). The original course preference Pb _org (i, j) is calculated by the above-described equation (1). By using the equation (1), as shown in FIG. 2, the original course preference Pb _org (Corg, j) at the time of the course change Corg (0 [deg]) of the original course maintenance is obtained.

The preference Pb _wp (Cwp, j) for the way point WP at the time of the course change angle Cwp to the way point WP shown in FIG. 2 is calculated using the original course preference Pb _org (Corg, j) by using the equation ( 7). In Expression (7), “β” is a coefficient that determines in advance how much the peak value PK1 is larger than the peak value PK2, and is a coefficient that satisfies “β> 1”. Note that the course change angle Cwp is determined based on, for example, the information (for example, the position) of the way point WP from the route setting unit 22 in FIG. 1 and the own ship information (for example, the position and the course) from the ship information acquisition unit 11. Can be
Pb _wp (Cwp, j) = β × Pb _org (Corg, j) (7)

On the other hand, the preference point Pb _wp (i, j) for the way point WP in the entire evacuating maneuvering space is calculated by the equation (8) or the equation (9). Equation (8) is applied when the stylus angle φi is smaller than the stapling angle Cwp to the way point WP (when φi <Cwp), and is applied in a range where the stapling angle φi is a left stapler than the stapling angle Cwp. . On the other hand, the equation (9) is applied when the turning angle φi is equal to or larger than the changing angle Cwp to the way point WP (when φi ≧ Cwp), and the changing angle φi is equal to the changing angle Cwp or is smaller than the changing angle Cwp. Applies within the range of right turn.
Pb _wp (i, j) = exp (a ₁ × φi + b ₁ ) (8)
Pb _wp (i, j) = exp (a ₂ × i + b ₂ ) (9)

“A ₁ ” in Expression (8) is calculated by Expression (10). In Expression (10), “Cmin” is the minimum value of the changing course angle in the avoidance maneuvering space (in other words, the maximum left changing course angle (here, −60 [deg])). “Z ₁ ” is the original course preference Pb _org (Cmin, j) at the course change angle Cmin, and “Z ₂ ” is the way point WP at the course change angle Cwp shown in the equation (7). The preference level Pb _wp (Cwp, j). “A ₁ ” presupposes that “Z ₁ ” is obtained for the changing needle angle Cmin, and determines the slope of the exponential function that obtains “Z ₂ ” for the changing needle angle Cwp. On the other hand, “b ₁ ” in Expression (8) is calculated by Expression (11). "B _1", in the formula (8), veering angle φi is "Cmin" way point WP for preference Pb _wp (i, j) in the case of a factor for adjusting such that the "Z _1" is there.
a ₁ = {1 / (Cwp−Cmin)} × log (Z ₂ / Z ₁ ) (10)
b ₁ = log (Z ₁ ) − (Cmin × a ₁ ) (11)

Similarly, “a ₂ ” in Expression (9) is calculated by Expression (12). In the equation (12), “Cmax” is the maximum value of the changing course angle in the avoidance maneuvering space (in other words, the maximum right changing course angle (here, +60 [deg])). “Z ₃ ” is the original course preference Pb _org (Cmax, j) at the course change angle Cmax. “A ₂ ” presupposes that “Z ₂ ” is obtained with respect to the stylus angle Cwp, and determines the slope of the exponential function such that “Z ₃ ” is obtained with respect to the stylus angle Cmax. On the other hand, “b ₂ ” in Expression (9) is calculated by Expression (13). "B _2", in the formula (9), veering way point WP for preference when angle φi is _{"Cwp" Pb wp (i,} j) is a factor for adjusting such that the "Z _2" is there.
a ₂ = {1 / (Cmax−Cwp)} × log (Z ₃ / Z ₂ ) (12)
b ₂ = log (Z ₂ ) − (Cwp × a ₂ ) (13)

As shown in FIGS. 2 and 3, the preference degree calculation unit 24 in FIG. 1 calculates an exponential function that decreases from the peak value PK1 as the distance from the position for the way point WP increases in the avoidance maneuvering space. The exponential function corresponds to the preference value Pb _wp (i, j) for the way point WP in Expression (8) or Expression (9). Further, as shown in FIG. 2 and FIG. 3, the preference degree calculation unit 24 calculates an exponential function that decreases from the peak value PK2 as the distance from the position for maintaining the current state increases in the avoidance maneuvering space. The exponential function corresponds to the original course preference Pb _org (i, j) in equation (1).

Based on this calculation result, the preference calculation unit 24 calculates the two exponential functions (waypoint WP preference Pb _wp (i, j) and original course preference, as shown in FIGS. 2 and 3). Pb _org (i, j)). Then, as shown in FIGS. 2 and 4, the preference calculating unit 24 calculates the value of the _waypoint WP preference Pb _wp (i, j) or the original course preference for each position in the avoidance maneuvering space. The final selection degree Pb (i, j) is calculated by using the larger one of the values of Pb _org (i, j). That is, the preference degree calculation unit 24 calculates Expression (14).
Pb (i, j) = max (Pb _wp (i, j), Pb _org (i, j)) (14)

  FIG. 5 is a diagram illustrating an example of a situation after the own ship has evaded another ship by the avoidance maneuver in the escape assist device of FIG. 1. FIG. 6 is a diagram illustrating an example of a processing result of the utility value calculating unit in the situation of FIG. FIG. 5 shows the same situation as in FIG. Accordingly, the risk calculating unit 26 calculates a collision risk as shown in FIG. On the other hand, the utility value calculation unit 20 calculates the collision risk (Rk (i, j)) and the preference (Pb (i, j)) as shown in FIG. To calculate the utility value u (i, j) based on equation (6). As a result, a utility value in the escape navigation space as shown in FIG. 6 is obtained.

  The ship maneuvering instruction unit 23 of FIG. 1 differs from the case of FIG. 17 described above based on the utility value of FIG. 6 in that the ship maneuvering method corresponding to the peak value PK1 (that is, the changing angle Cwp to the way point WP and maintaining the original speed). An instruction is given to the autopilot device 15. As a result, in the situation shown in FIG. 5, unlike the case of FIG. 15, the own ship OS can perform the return ship maneuver toward the way point WP.

<< Operation of Evacuation Support Device (Embodiment) >>
FIG. 7 is a flowchart showing an example of a schematic processing content of a main part in the evacuation assistance device of FIG. The escape assist device 10 repeatedly executes the flow of FIG. 7 for each predetermined control cycle. In FIG. 7, the evacuation assistance device 10 acquires own ship information (the course, speed, and position) from the ship information acquisition unit 11 and information on other ships (the course, speed, and position) existing within a predetermined range (FIG. 7). Step S101). Next, the avoidance assistance device 10 causes the preference degree calculation unit 24 to acquire the information of the way point WP from the route setting unit 22 (Step S102).

  Subsequently, the evacuation assistance device 10 causes the preference level calculation unit 24 to calculate the preference level Pb (i, j) of Expression (14) having two peak values PK1 and PK2 as shown in FIG. 2 ( Step S103). Next, the evacuation assistance device 10 causes the risk calculating unit 26 to calculate the collision risk Rk (i, j) in Expression (5) (Step S104). Subsequently, the escape assist device 10 uses the preference value Pb (i, j) calculated in step S103 and the collision risk degree Rk (i, j) calculated in step S104 in the utility value calculating unit 20. To calculate the utility value u (i, j) shown in equation (6) (step S105). After that, the escape assist device 10 causes the ship maneuvering instruction unit 23 to determine an optimum ship maneuvering method (change of angle and speed change rate) based on the utility value u (i, j) calculated in step S105 (step S106).

FIG. 8 is a schematic diagram showing an operation example when performing the evacuating maneuvering and returning maneuvering in the evacuation assisting device of FIG. 1. In FIG. 8, at time t = t0, no other ship TS exists in a predetermined range around the own ship OS, and the own ship OS is moving toward the way point WP. In this case, the changing point angle (0 [deg]) corresponding to the peak value PK1 of the way point WP preference Pb _{wp and} the changing point angle (0 [deg]) corresponding to the peak value PK2 of the original course preference Pb _org are: Become equal. Also, the collision risk Rk becomes zero. As a result, the utility value is determined based on the _waypoint WP preference _Pbwp , and maneuvering to the original course is maintained.

At time t = t1, another ship TS that is traveling to the left exists at the front right of the ship OS. Accordingly, in this example, the collision risk Rk occurs in a range on the left turning side of the right turning angle “θ1”. The utility value is obtained by subtracting the collision risk Rk from the _waypoint WP preference _Pbwp including the original course preference _Pborg . As a result, the utility value becomes maximum at the right turning angle “θ1”, and the right turning course with the changing angle “θ1” is determined as the optimal avoiding and maneuvering method.

At time t = t2, when the right veering is made of veering angle ".theta.1", the preference Pb _wp for way point WP, shifted by veering angle ".theta.1" left veering side from the state of the time t = t1. Further, the collision risk Rk also shifts from the state at time t = t1 to the left turning side by the turning angle “θ1”. As a result, the utility value becomes maximum at the course change angle (0 [deg]) of the original course maintenance, and after the right course change of the course change angle “θ1” is performed, the avoiding maneuvering toward the original course is performed.

At time t = t3, the evacuating maneuver of own ship OS and the progress of the other ship TS to the left have progressed to some extent. Along with this, the collision risk Rk becomes smaller as compared with the case at time t = t2, and the utility value becomes larger by the smaller the collision risk Rk. In addition, the turning angle “−θ2” to the way point WP is shifted to the left turning side by an amount corresponding to the advance of the avoiding maneuver as compared with the case at time t = t2. Here, the range of the changing angle at which the collision risk Rk exists includes the changing angle “−θ2” to the way point WP. The veering angle "-.theta.2", there is a peak value PK1 waypoint WP for appetite Pb _wp. Therefore, the influence of the peak value PK1 on the utility value increases as the collision risk Rk decreases. However, at this stage, the utility value is still the maximum at the changing course angle for maintaining the original course.

  At time t = t4, the ship's own ship OS's evacuating maneuver and the proceeding to the left side of the other ship TS are further advanced. Accordingly, the collision risk Rk becomes smaller and the utility value becomes larger as compared with the case at time t = t3. Further, the course change angle “−θ3” to the way point WP is shifted to the left course change side by an amount corresponding to the advance of the avoiding maneuver as compared with the case at time t = t3. At time t = t4, the collision risk Rk is smaller than the difference value between the peak value PK1 and the peak value PK2. As a result, the utility value at the changing course angle “−θ3” becomes larger than the utility value at the changing course angle of maintaining the original course. As a result, the left-hand changing course of the changing course angle of “−θ3” is determined as the optimum returning boat maneuvering method, and based on this, the returning boat maneuvering toward the way point WP is performed.

<< Further details of the preference calculation unit >>
For example, as shown in FIG. 2 and FIG. 3, the waypoint WP preference corresponding to the maximum course change angle (right change course angle of 60 [deg] and left change course angle of −60 [deg]) in the avoidance maneuvering space. The value of Pb _wp (i, j) is equal to the value of the original course preference Pb _org (i, j) corresponding to the maximum course change angle. That is, the preference calculation unit 24 sets the value of the preference Pb _wp (i, j) for the waypoint WP and the value of the preference Pb _org (i, j) for the original course at the position of the foot of the mountain. Then, the inclination or the like of the way point WP preference level Pb _wp (i, j) is appropriately adjusted.

If “Pb _wp (i, j) <Pb _org (i, j)” at the base of FIG. 3, the inclination of the _waypoint WP preference Pb _wp (i, j) is the case of FIG. It becomes sharper than. This means that the options for maneuvering with the _waypoint WP preference level Pb _wp (i, j) are substantially reduced in advance. Conversely, if “Pb _wp (i, j)> Pb _org (i, j)” at the foot position in FIG. 3, the two types of preference degrees intersect, for example, at a predetermined right turning angle. At a right turning angle larger than this intersection, the preference point for waypoint WP Pb _wp (i, j) instead of the original course preference Pb _org (i, j) is applied as the preference in FIG. In order to avoid such an unfavorable situation, it is desirable that the preference calculation unit 24 adjusts the values of the bases of the two types of preference to be equal.

  Further, for example, there is a case where the changing angle Cwp to the way point WP in FIG. 2 does not fall within the range from the maximum left changing angle “Cmin” to the maximum right changing angle “Cmax”. The preference degree calculation unit 24 regards the changing course angle Cwp as “Cmin” when the actual changing course angle Cwp is a left changing course angle larger than “Cmin”, and determines the right changing course angle when the actual changing course angle Cwp is larger than “Cmax”. , The preference angle is calculated by regarding the changing needle angle Cwp as “Cmax”. More specifically, the preference calculation unit 24 calculates the preference in the range of “Cmin” to “Cmax”. Therefore, for example, when the actual changing angle Cwp is the left changing angle larger than “Cmin”, if the changing angle Cwp is applied as it is, the peak value PK1 does not exist in the range of “Cmin” to “Cmax”. The operation shown in FIG. 8 becomes difficult.

<< Main effects of the embodiment >>
As described above, by using the evacuation support device of the embodiment, typically, it becomes possible to realize return ship maneuvering in addition to evacuation maneuvering. Here, the preference level in FIG. 2 is based on the assumption that the ship maneuvers while avoiding the return to the way point WP, and starts the return maneuver when the collision risk in the way point WP direction is sufficiently reduced. It can be said that it reflects the actual situation. By using such a preference, return maneuvering can be started when the collision risk in the direction of the way point WP is reduced to a sufficient level instead of zero, so that efficiency can be improved in terms of time and economic loss. It becomes possible to plan.

Here, as another modified example, for example, a method using only the _waypoint WP preference Pbwp in FIG. 2 can be considered. In this case, for example, at time t = t3 in FIG. 8, the changing hand angle is determined based on the thick broken line corresponding to the _waypoint WP preference _Pbwp . However, in this case, for example, as can be seen from the situation at time t = t3 in FIG. 8, the level of the collision risk Rk at the start of restoration maneuvering various parameters (inclination or Pb _wp, positions of θ2, etc.) And there is a possibility that safety at the time of return maneuvering may not be sufficiently ensured.

  On the other hand, if the preference degree of FIG. 2 is used, as shown at time t = t4 in FIG. 8, the level of the collision risk at the time of starting the return ship maneuver is determined by the difference value between the peak value PK1 and the peak value PK2 (equation). (7) β), and safety at the time of return ship maneuvering can be sufficiently ensured. Further, from the viewpoint of the actual situation of the ship operator, the ship operator usually maintains the original course and the original speed until the safety can be ensured, while keeping in mind the return to the way point WP during the avoiding maneuver. Have an idea. The preference having two peak values PK1 and PK2 as shown in FIG. 2 reflects the actual situation of such a boat operator.

  In FIG. 2, the peak value PK1 is set at the position where the gear ratio is 100% (maintenance speed maintained), but is not necessarily limited to this position. For example, a ship operator may prefer to accelerate slightly when starting a return ship maneuver. The peak value PK1 may be set, for example, at a position with a gear ratio greater than 100%, reflecting the actual situation of the boat operator.

  As described above, the invention made by the inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and can be variously modified without departing from the gist of the invention.

  For example, the formulas for calculating the preference, the collision risk, and the utility value are not necessarily limited to the above-described formulas, and may be appropriately changed without departing from the gist.

Reference Signs List 10 Evacuation support device 11 Ship information acquisition unit 13 Result display device 14 Electronic chart 15 Autopilot device 20 Utility value calculation unit 22 Route setting unit 23 Ship steering instruction unit 24 Preference calculation unit 26 Risk calculation unit 30 AIS
31 radar 32 camera 35 bearing sensor 36 speed sensor 37 GPS
40 Bumper area 41, 42 Risk function 43 Relative course OS Own ship TS Other ship PK Peak value WP Waypoint

Claims (8)

A space that combines the options of the changing angle and the speed change rate from the current situation is defined as an evacuation maneuvering space. A preference calculation unit for calculating the degree,
For each position of the avoiding maneuvering space, a risk calculating unit that calculates a collision risk with respect to another ship when the own ship performs the avoiding maneuvering,
An utility value calculating unit that calculates an utility value by subtracting the collision risk degree from the preference degree for each position in the avoidance maneuvering space;
Has,
The preference degree calculating unit has a first peak value at a first position that is a position in the avoidance maneuvering space obtained by combining the turning angle to a way point and the speed change rate to a predetermined speed. Calculating the preference degree having a second peak value at a second position that is a position in the avoidance maneuvering space in which the course change angle of course keeping and the speed change rate of original speed maintaining are combined;
Evacuation support device. The evacuation support device according to claim 1,
The first peak value is greater than the second peak value;
Evacuation support device. The evacuation support device according to claim 2,
The preference calculation unit includes a first exponential function that decreases from the first peak value as the distance from the first position increases, and a second exponential function that decreases as the distance from the second position increases. A second exponential function falling from the peak value is superimposed, and for each position in the escape maneuvering space, the larger value of the first exponential function or the value of the second exponential function is calculated. Calculate the selection degree by using
Evacuation support device. The evacuation support device according to claim 3,
The value of the first exponential function corresponding to the maximum steepness angle in the avoiding maneuvering space is equal to the value of the second exponential function corresponding to the maximum staggering angle in the avoiding maneuvering space,
Evacuation support device. The evacuation support device according to claim 1,
The predetermined speed at the first position is an original speed;
Evacuation support device. The evacuation support device according to claim 1,
Furthermore, a route setting unit that notifies the preference point calculation unit of the waypoint information based on the planned route stored in advance,
Evacuation support device. The evacuation support device according to claim 1,
The preference calculating unit, when the change angle to the way point is larger than the maximum change angle in the avoidance maneuvering space, considers the change angle to the way point to be the maximum change angle. Calculate preference,
Evacuation support device. The evacuation support device according to claim 1,
Further, a ship maneuvering instruction unit that instructs the changing hand angle and the speed change rate corresponding to the position of the avoidance maneuvering space where the utility value is maximum as a maneuvering method,
Evacuation support device.