JP5601878B2 - Route prediction device - Google Patents

Description

  The present invention relates to a route prediction apparatus that monitors the movement of a flight target and predicts a route ahead if observation data is missing due to the target being hidden in the shade.

The conventional route prediction apparatus also performs route prediction after missing observation data by using a tracking filter intended for target tracking when observation data is always obtained. A specific operation after missing observation data is to repeatedly input the previous predicted value to the tracking filter instead of the observation data and calculate the target position and speed at a future time.
For example, the target position and speed information at the current time and the terrain information are input to the fuzzy inference element, and the basic action that the target takes on the terrain (a collision with the terrain is avoided while the target maintains a low ground altitude). Etc.), a prediction unit for calculating the target position and speed at the next time is calculated.
Further, while the observation data is obtained, the learning adjustment unit is used to adjust the maneuvering amount calculated from the fuzzy inference so that the error is reduced by using the error between the observation data and the predicted value. By having two components, highly reliable route prediction is performed even when observation data is missing (see, for example, Patent Document 1).

Further, as a conventional technique that can be used for route prediction, there is a technology in which the target side plans an optimal flight route to a destination (an important base on the monitoring side). Hereinafter, this is referred to as “target side optimum route generation technology”. Since the route generated by the target-side optimal route generation technology is a basic action taken by the target with respect to terrain and monitoring radar, it can also be used as a predicted route when observation data is missing. is there.
The operation of the target-side optimum route generation technique is to generate an optimum route by discretizing the moving space (eg, dividing into a grid) and selecting a set of route points that optimizes a certain evaluation function by Dijkstra. That's it. The evaluation function used here is often modeled as a linear sum of the evaluation values of each element (movement distance, radar detectability, etc.) that should be optimized in the path generation. There are adjustment parameters (weighting coefficient values for each term). For example, when the target side performs the flight route planning, a route that has a short total movement distance and is difficult to be detected by the radar is determined as an optimum route, and a route that minimizes the evaluation function expressed by the following equation (1) is obtained.

    F (path) = (1−α) × total moving distance + α × detection risk from radar (1)

  Here, α is a trade-off adjustment parameter (see, for example, Non-Patent Document 1).

Japanese Patent No. 2749727

Shi Xiaoli, `` Optimization of Fighter Aircraft Evasive Trajectories for Radar Threats Avoidance '', In Proceedings of the 2007 IEEE International Conference on Control and Automation, P303-307

Since the conventional route prediction apparatus performs route prediction by accumulating prediction values when observation data is lacking, there is a problem in that the reliability of the prediction route decreases if this becomes longer. Here, in Patent Document 1, terrain information is used for route prediction based on the premise that the basic behavior of the target with respect to the terrain is “the route follows a route that avoids collision with the terrain while maintaining a low ground altitude”. Thereby, patent document 1 improves the reliability of a prediction path | route also at the time of observation data intermittent when a target is hidden in the shade.
However, Patent Document 1 does not assume that the target takes a path that is difficult for the monitoring radar to detect as the basic behavior of the target. (For example, when the target passes through a route that does not return from Sanin considering the radar coverage on the monitoring side), the reliability of the predicted route is reduced.

  On the other hand, in Non-Patent Document 1, the target-side optimal route generation technology, which is a conventional technology that can be used for target route prediction when observation data is lacking, is used as the basic behavior of a target. It is premised on “passing the route”. However, there is a problem that the reliability of the predicted path is lowered when the trade-off adjustment parameter value of each term of the target path evaluation function used at the time of path generation is inappropriate.

  The present invention has been made to solve the above-mentioned problems, and the basic behavior of the target is “considering terrain information, monitoring-side radar coverage information, etc., with an important base on the monitoring side as a destination. The target route is predicted by generating a target-side optimal route when observation data is missing, and the target-side optimal route is further assumed. By setting / adjusting the parameter value for trade-off adjustment of the target path evaluation function used for generating the target based on the obtained observation information, it is reliable even when the observation data is missing for a long time An object is to obtain a route prediction device capable of route prediction.

The route prediction apparatus according to the present invention includes a sensor unit that observes a target and outputs observation data, a target information acquisition unit that outputs the input observation data as observation information of the target, and sequentially obtains the observation information obtained. Observation information storage unit for storing in series, terrain information database in which terrain information is stored in advance, monitoring side radar information database in which monitoring radar information is stored in advance, and monitoring side that is highly likely to be targeted by the target Trade-off between each element of the target side path evaluation function defined by the linear sum of the evaluation values of the elements that affect the optimality of the path for the above target, the monitoring side important base database in which the important base data is stored in advance Target side path evaluation function parameter value storage unit that stores parameter values for adjusting the target position and speed at the path generation start time, target side Based on the parameter values of the road evaluation function, the monitoring important base database, the monitoring radar information database, and the data stored in the topographic information database, the monitoring important base that optimizes the target side path evaluation function Target side optimal route generation unit that generates a route up to the target side optimal route, a series of the observation information of the target, the adjustment mode, and values of a plurality of parameters different in the target side route evaluation function, the target side optimal route generation unit To generate a target-side optimum route, and the parameter values used in the route most similar to the series of the target observation information among the plurality of different routes generated by the target-side optimum route generator Target-side path evaluation function parameters that are updated to the parameter values stored in the parameter-value storage unit The observation information stored in the observation information storage unit is monitored and analyzed, and while the observation data is obtained, the target side path evaluation function parameter value adjustment is performed based on the analysis result of the observation information. Is periodically executed, and while the observation data is missing, the parameter value stored in the target side path evaluation function parameter value storage unit is input to the target side optimal path generation unit and the target side optimal a path prediction apparatus including a path prediction system task execution unit to generate a route, the target-side route evaluation function parameter value adjustment unit, while the observed data is obtained, based on the analysis result of the observation information Determining whether or not to adjust the parameter value by focusing on some of the plurality of parameters of the target-side path evaluation function used in generating the target-side optimal path, and based on the determination Adjust all or some of the parameter values.

  The route predicting apparatus according to the present invention is configured such that the basic behavior of the target is “a route that optimizes a target-side route evaluation function that takes into account the topography, the monitoring-side radar coverage, etc. Based on the premise that the target side route evaluation function used to generate the target side optimal route is predicted by generating the target side optimal route when the observation data is missing. When setting and adjusting the parameter value of the trade-off adjustment parameter, the parameter value to be set and adjusted is narrowed down based on the analysis result of the observation information, so the parameter value can be set and adjusted efficiently, and the observation data There is an effect that it is possible to predict a reliable route even if it is missing over a long period of time.

It is a block diagram of the path | route prediction apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the timing which the path | route prediction system task execution part of FIG. 1 performs a target side path | route evaluation function parameter value adjustment part or a target side optimal path | route production | generation part. It is a figure which shows the example of the movement destination candidate point produced | generated in the target side optimal path | route production | generation part of FIG. It is a figure which shows the example of a setting of the adjustment mode input when the path | route prediction system task execution part of FIG. 1 performs the target side path | route evaluation function parameter value adjustment part. It is a figure which shows operation | movement of the target side path | route evaluation function parameter value adjustment part of FIG. It is a figure which shows the calculation method of the position error of the path | route produced | generated by the target side optimal path | route production | generation part in the target side path | route evaluation function parameter value adjustment part of FIG. 1, and a series of observation information.

A preferred embodiment of a route prediction apparatus according to the present invention will be described below with reference to the drawings.
Embodiment 1 FIG.
FIG. 1 is a configuration diagram of a route prediction apparatus according to Embodiment 1 of the present invention.
The route prediction apparatus according to the first embodiment of the present invention includes a sensor unit 1 composed of a radar that observes a target and outputs observation data, and receives observation data as well as target observation information such as observation time, position, and speed. The target information acquisition unit 2 that outputs the observation information, the observation information storage unit 3 that stores the observation information sequentially obtained at a predetermined sampling period in time series, the terrain information database 4 in which the terrain information is stored in advance, the radar information on the monitoring side, for example The monitoring radar information database 5 in which the arrangement and coverage are stored in advance, the monitoring important base database 6 in which important monitoring bases that are likely to be targets are stored in advance, and the flight for the target Each element of the target-side path evaluation function defined by the linear sum of evaluation factors such as flight altitude and the detectability from the monitoring-side radar, which affects the optimality of the path Parameter value for adjusting the trade-off between, i.e. provided with a target-side route evaluation function parameter value storage unit 7 for storing the weighting coefficients of the respective terms.

  In addition, the route prediction apparatus according to Embodiment 1 of the present invention is the input route generation start time, the target position and speed at the route generation start time, the parameter value of the target side route evaluation function, and the monitoring side important Based on the data stored in the base database 6, the monitoring radar information database 5, and the topographic information database 4, the optimal path for the target side with the important base on the monitoring side as the destination Target-side route evaluation function parameter value storage unit 8 based on the target-side optimum route generation unit 8 that is generated as “a route to be converted” and a series of input target observation information and an adjustment mode (specification of parameters to be adjusted) In order to set and adjust the parameter values stored in the unit 7, the target-side optimal route generation unit 8 is executed and the parameter values of the target-side route evaluation function are different. A plurality of paths are generated, and the parameter values stored in the target-side path evaluation function parameter value storage unit 7 are updated with the parameter values used in the path most similar to the series of input target observation information. A target side path evaluation function parameter value adjustment unit 9 that repeats a plurality of times is provided.

  Further, the route prediction apparatus according to Embodiment 1 of the present invention uses the observation information stored in the observation information storage unit 3 based on the data stored in the monitoring radar information database 5 and the terrain information database 4. While monitoring and analyzing, while the observation data is obtained, the target side path evaluation function parameter value adjustment unit 9 is periodically executed in the adjustment mode in which the number of parameters to be adjusted is reduced as much as possible based on the analysis result of the observation information. If the observation data is missing, the target-side optimum route generation unit 8 is executed with the parameter value stored in the target-side route evaluation function parameter value storage unit 7, that is, the value adjusted based on the observation information as an input. The route prediction system task execution unit 10 outputs the obtained route to the outside of the route prediction device as a predicted route after lack of target observation data.

Next, a series of operations of the route prediction apparatus according to Embodiment 1 of the present invention will be described.
The sensor unit 1 observes a specific target at an arbitrary sampling interval. Data obtained by observation by the sensor unit 1 is processed by the target information acquisition unit 2 and converted into observation information including the target observation time, position, and velocity. The target observation information at each time obtained from the target information acquisition unit 2 is sequentially stored in the observation information storage unit 3 in time series.
The route prediction system task execution unit 10 monitors and analyzes the target observation information stored in the observation information storage unit 3 based on the data stored in the monitoring radar information database 5 and the terrain information database 4. The target side route evaluation function parameter value adjustment unit 9 and the target side optimum route generation unit 8 are executed.

FIG. 2 shows the timing at which the target side route evaluation function parameter value adjustment unit 9 and the target side optimal route generation unit 8 are executed by the route prediction system task execution unit 10.
The target-side path evaluation function parameter value adjustment unit 9 is at predetermined equal intervals while the target is observed by the sensor unit 1 and observation data is obtained, and after returning from a period in which the observation data is missing Immediately after that, a series of observation information newly acquired after execution of the previous target-side path evaluation function parameter value adjustment unit 9 and an adjustment mode are input and executed to adjust the parameter values. The adjustment mode is a mode for specifying parameters to be adjusted, and is set based on a series of analysis results of newly acquired observation information.

  When the observation data is missing, the target-side optimum route generation unit 8 is the parameter stored in the target-side route evaluation function parameter value storage unit 7 immediately before the observation data is missing. The value, that is, the parameter value adjusted by the target side route evaluation function parameter value adjusting unit 9, is input and executed to generate the target side route. Then, the route data obtained from the target-side optimum route generation unit 8 is sent to the route prediction system task execution unit 10, and this route is used as the target prediction route after the observation data is missing from the route prediction system task execution unit 10. Output to the outside of the prediction device.

Further, the operation of the target side optimum route generation unit 8 will be described in detail.
Here, repeatedly generating a plurality of target destination candidate points for each arbitrary time interval ΔT and selecting a destination candidate point with a good evaluation value of the target side path evaluation function as a target destination at the next time Then, the target side optimum route is generated. Hereinafter, the position and speed of the target and the time when the target holds the position and speed are referred to as “target state”.

Specifically, the following processing is performed.
In step 1, the route generation start time, the target position at the route generation start time, and the speed given as inputs when the target side optimum route generation unit 8 is executed are set as the “target state at the current time”. Further, the parameter value of the target side path evaluation function given as an input when executing the target side optimal path generation unit 8 is set in the target side path evaluation function.
In step 2, a plurality of destination candidate points are generated after the time interval ΔT has elapsed with respect to the target state at the route generation start time. The target state of the destination candidate point is the time when the target exists at the destination candidate point (the time obtained by adding ΔT to the current time t), and the position and speed of the destination candidate point.
FIG. 3 is a diagram showing an example of destination candidate points generated by the target-side optimal route generation unit 8 provided in the route prediction device according to Embodiment 1 of the present invention.
The maximum angle at which the target can turn is calculated from the target speed at the current time t and the time interval ΔT. As shown in FIG. 3, “moving destination candidate points without turning” and “ In addition to “movement destination candidate points when turning at the maximum turning angle”, a plurality of “movement destination candidate points when turning at a turning angle smaller than the maximum turning angle” is generated. The target position and speed at the next time t + ΔT are calculated from the turning angle of the movement destination candidate point and the position and speed at the current time.

  In step 3, the destination candidate point generated in step 2 is evaluated by the target side path evaluation function, and the destination candidate point with the best evaluation value is selected as the destination at the next time t + ΔT. An example of the target side path evaluation function f is shown in Expression (2). The value of each term is calculated based on data stored in the topographic information database 4, the monitoring radar information database 5, and the monitoring important base database 6.

    f (destination point) = α × horizontal distance to destination + β × terrain altitude + γ × detection risk from radar (2)

The coefficients α, β, and γ in Equation (2) are trade-off adjustment parameters, and each term in which α, β, and γ are coefficients has an optimum trade-off relationship, and depends on the specific gravity of α, β, and γ. The generated route is as follows.
α and γ have a trade-off relationship between the shortest route to the destination or the route that is difficult to be detected by the radar. When the value of α is increased, a route that places more importance on the shortest route is obtained. On the other hand, if the value of γ is increased, the path becomes more focused on the detection ability from the radar.

  For α and β, when performing a flight that maintains a low ground altitude, when avoiding collision with the terrain, emphasis is placed on terrain following, that is, not deviating from the destination direction. Focus on avoiding terrain collisions by climbing along the slope, or avoiding terrain, that is, flying at low altitudes. There is a trade-off between avoiding collisions. When the value of α is increased, a terrain following type route is obtained. Conversely, if the value of β is increased, a terrain avoidance type route is obtained.

In step 4, if the target condition of the destination candidate point selected in step 3 satisfies the end condition that the destination is close to the monitoring important base, that is, an arbitrarily set distance, then A set of all the movement destination candidate points (target states) selected in (1) is output as a generation route, and the processing of the target side optimum route generation unit 8 is terminated.
If the end condition is not satisfied, ΔT is added to the current time, the target state at the current time is set as the target state of the selected destination candidate point, and the process returns to step 2.

Next, a method for setting an adjustment mode that is given as an input when the target side route evaluation function parameter value adjustment unit 9 is executed by the route prediction system task execution unit 10 will be described.
An adjustment mode is set so as to narrow down the parameters to be adjusted as much as possible according to a series of analysis results of observation information newly acquired since the target side path evaluation function parameter value adjustment unit 9 was executed last time.
For example, in the case of the target side path evaluation function of Expression (2), the adjustment mode is set as follows according to a series of analysis results of the observation information. Here, FIG. 4 shows a setting example of the adjustment mode for the target-side path evaluation function of Expression (2).

When the series of analysis results of newly acquired observation information is “immediately after initial detection of the target”, parameter adjustment processing has not been performed, so “parameter value initialization mode” is set to the adjustment mode. .
If the series of analysis results of newly acquired observation information is "after passing through a place where the terrain is flat to some extent", it is predicted that the target flew without being affected by the terrain. It is difficult to adjust the specific gravity of α and β based on the above. Therefore, “specific gravity adjustment mode of α and γ” for adjusting the specific gravity focused on α and γ is set as the adjustment mode.

If a series of analysis results of newly acquired observation information is “after going out of sight of a mountain and returning from the same mountain (or a nearby mountain) from sight”, the target strongly influences the topography. Since it is predicted that the flight was received, it is difficult to adjust α and γ based on this observation information. Therefore, “specific gravity adjustment mode of α, β” for adjusting the specific gravity limited to α and β is set as the adjustment mode.
If a series of analysis results of newly acquired observation information is "after passing through a complicated terrain" or "after missing data for a long time", the terrain and monitoring radar Since it is difficult to understand how the flight has been affected, the “specific gravity adjustment mode of α, β, γ” for adjusting the specific gravity of all α, β, γ without adjusting the parameters to be adjusted is set as the adjustment mode.

Next, the operation of the target side path evaluation function parameter value adjustment unit 9 will be described in detail. FIG. 5 is an image diagram showing the operation of the target-side path evaluation function parameter value adjustment unit 9 provided in the path prediction apparatus according to Embodiment 1 of the present invention.
First, the parameter value stored in the target-side path evaluation function parameter value storage unit 7 is read, and the value is set as the reference parameter used in the parameter adjustment process. Then, the following adjustment process for the reference parameter is repeated a plurality of times according to the series of input observation information and the set adjustment mode.

In step 11, according to the adjustment mode set, neighborhood parameters P ₁ to P of the reference parameter P _{K N,} i.e. multiple generating parameters consisting reference parameter P values were slightly changing the value of _K. For example, in the case of the target-side path evaluation function of Expression (2), the reference parameter P _K is a set of parameter values P _K = {α _K , β _K , γ _K } for each term, and the neighborhood depends on the adjustment mode. The parameters P _{1 to} P _N are generated as follows.
In the parameter value initialization mode, a plurality of combinations of various values of α, β, and γ are generated as neighborhood parameters.
In the specific gravity adjustment mode of α and β, the reference parameter values are used as they are for α and γ, and a plurality of combinations of α, β, and γ in which only β is increased or decreased with respect to the reference value are generated as neighborhood parameters.
In the specific gravity adjustment mode for α and γ, the reference parameter values are used as they are for α and β, and a plurality of combinations of α, β, and γ in which only γ is increased or decreased with respect to the reference value are generated as neighborhood parameters.
In the specific gravity adjustment mode of α, β, and γ, α uses the reference parameter value as it is, and generates a plurality of combinations of α, β, and γ by increasing and decreasing β and γ with respect to the reference value as neighboring parameters.

In step 12, the target position and velocity at the earliest time in the series of input observation information, each of the reference parameter P _K and the plurality of neighboring parameters P _{K1 to} P _KN generated in step 11 are set. As an input, the target side optimal route generation unit 8 generates routes R _K and R _{K1 to} R _KN and obtains the target side optimal route when each parameter value is set in the target side route evaluation function.

In step 13, the route most similar to the series of input observation information is selected from the routes R _K and R _{1 to} R _N generated in step 12. The similar route indicates, for example, a route having the smallest position error with respect to a series of observation information. The position error in this case is calculated by the following process. FIG. 6 shows the calculation method.
Here, the route generated by the target-side optimal route generation unit 8 is a set of destination candidate points R = {r ₁ , r ₂ ,..., R _n } selected as the destination, and the observation information The series is OB = {ob ₁ , ob ₂ ,..., Ob _m }. Each r _j (j = 1, 2,..., N) and ob _i (i = 1, 2,..., M} are observed in the case of ob _i in addition to the target position and speed. In the case of time, r _j , information on the time at which the target exists at that position is held.

In step 13-1, r _j and ob _i are associated with each other using time. In the example of the target side optimal route generation unit 8, the generated route has time information. However, if the route generated by the target side optimal route generation unit does not have time information, r _j and r _{j The} time is calculated by calculating the distance of ₋₁ and dividing by the target speed.
In step 13-2, the position error (distance) of the associated r _j and ob _i is calculated.
In step 13-3, the total sum of the position errors calculated in step 13-2 is obtained and set as the position error between the path generated by the target side optimal path generation unit 8 and the series of observation information.
In step 14, the reference parameter value is updated with the parameter value used to generate the route selected in step 13 (the route most similar to the series of observation routes). When an arbitrary number of repetitions designated in advance is reached, the target side path evaluation function is updated by updating the parameter value stored in the target side path evaluation function parameter value storage unit 7 with the reference parameter value (adjusted parameter value). The operation of the parameter value adjustment unit 9 is terminated. If not, return to Step 11.

  In the route prediction apparatus according to the first embodiment of the present invention, the basic behavior of the target is “optimize the target side route evaluation function considering the topography, the monitoring radar coverage, etc., with the important base on the monitoring side as the destination. Based on the premise of `` flying a route like this, '' target route prediction is performed by generating a target route optimally by generating a target route optimal when observation data is missing, and used in generating the target route optimal When setting and adjusting parameter values for trade-off adjustment parameters for each element of the function, the number of parameters to be set and adjusted is reduced based on the analysis results of observation information, so parameter values can be set and adjusted efficiently. Thus, there is an effect that a reliable route prediction is possible even if observation data is missing for a long period of time.

Embodiment 2. FIG.
Since the route prediction device according to Embodiment 2 of the present invention is different from the route prediction device according to Embodiment 1 of the present invention in that the target-side optimum route generation unit is the same except for the above, the same reference numerals are used for the same parts. Is added and explanation is omitted.
The target side optimum route generation unit according to the second embodiment of the present invention implements the present invention except that a condition that “the target can be detected again by the monitoring radar” is added to the end condition of the route generation to be used. This is the same as the target-side optimum route generation unit 8 according to the first embodiment.

  The route prediction apparatus according to the second embodiment of the present invention immediately predicts a predicted route even when a situation in which the target can be detected by the monitoring radar is predicted before the target arrives at the destination (monitoring important base). Is output to the outside, and there is an effect that the real-time property of information acquisition of “target reappearance time, position and speed”, which is important for the purpose of route prediction, is improved.

Embodiment 3 FIG.
The route prediction device according to Embodiment 3 of the present invention is different from the route prediction device according to Embodiment 1 of the present invention, in the target-side optimum route generation unit and the target-side route evaluation function parameter value adjustment unit, and is otherwise the same. Therefore, the same reference numerals are added to the same parts, and the description is omitted.
The target-side optimal route generation unit according to the third embodiment of the present invention inputs the route generation end time and sets the route generation end condition as “when the target flies from the current time to the generated route. This is the same as the target-side optimum route generation unit 8 according to the first embodiment of the present invention, except that the condition that “the end time is reached” is added.

  The target side route evaluation function parameter value adjustment unit according to the third embodiment of the present invention executes the target side optimal route generation unit and sets the parameter value set in the target side route evaluation function in order to use the parameter adjustment processing Generate multiple paths with different. The route generated at this time does not require a route to the destination, and only a route corresponding to a series of observation information given as an input to the target side route evaluation function parameter value adjustment unit. Therefore, when the target-side optimal route generation unit is executed, the latest target observation time in the series of observation information is input as the route generation end time. Except for the above, the operation of the target-side path evaluation function parameter value adjustment unit is the same as that of the first embodiment.

  In the route prediction apparatus according to Embodiment 3 of the present invention, the target-side route evaluation function parameter value adjusting unit executes the target-side optimal route generation unit by specifying only necessary portions, thereby executing the target-side route evaluation function parameter. The parameter adjustment processing of the value adjustment unit is made efficient, and there is an effect that the real-time property is improved.

Embodiment 4 FIG.
Since the route prediction device according to Embodiment 4 of the present invention is different from the route prediction device according to Embodiment 1 of the present invention in that the route prediction system task execution unit is the same, the same reference numerals are used for the same parts. Is added and explanation is omitted.
The route prediction system task execution unit according to the fourth embodiment of the present invention outputs a predicted route that can be executed by the target side optimal route generation unit 8 when the observation data is missing, but the observation data continues to be missing. If so, the target side route evaluation function parameter value adjustment unit 9 is executed again, and the parameter value stored in the target side route evaluation function parameter value storage unit 7 is adjusted. In this case, the series of observation information input to the target-side path evaluation function parameter value adjustment unit 9 is a series of observation information input last time or in the past. Then, the parameter value after readjustment stored in the target-side route evaluation function parameter value storage unit 7 is input to the target-side optimal route generation unit 8, the optimal route generation processing is executed, and the obtained route is used as the predicted route. Output to the outside again. Except for the above, the operation of the route prediction system task execution unit is the same as that of the first embodiment.

  In the route prediction device according to the fourth embodiment of the present invention, the route prediction system task execution unit outputs the predicted route when the observation data is missing, and then continues the target side route when the observation data is missing. The parameter value of the evaluation function is further adjusted based on the observation information, and the route generated by the parameter value after the adjustment is output to the outside as the prediction route again, so that the effect of providing a more reliable prediction route is obtained. is there.

Embodiment 5 FIG.
In the route prediction device according to Embodiment 5 of the present invention, the user sets the parameter value stored in the target side route evaluation function parameter value storage unit 7 from the outside in the route prediction device according to Embodiment 1 of the present invention. Since the functions that can be added are different and the others are the same, the same parts are denoted by the same reference numerals and the description thereof is omitted.
In the route prediction apparatus according to Embodiment 5 of the present invention, the initial value of the parameter value of the target-side route evaluation function can be set by the user, and the parameter adjustment process can be performed interactively with the user, so that the reliability is higher. There is an effect that a route can be predicted.

Embodiment 6 FIG.
The route prediction apparatus according to Embodiment 6 of the present invention differs from the route prediction apparatus according to Embodiment 1 of the present invention in the route prediction system task execution unit, and is otherwise the same. Is added and explanation is omitted.
The route prediction system task execution unit according to Embodiment 6 of the present invention executes the target side optimal route generation unit 8 with the observation data missing from the route prediction system task execution unit according to Embodiment 1 of the present invention. When a predicted route is generated, a plurality of routes are generated using a plurality of different parameter values, and the rest is the same.
The different parameter values are, for example, as follows.
A parameter near the parameter value stored in the target-side path evaluation function parameter value storage unit 7.
A parameter value input by adding a function that allows the user to set a parameter value that the user wants to additionally display as a predicted route in the route prediction system task execution unit.

  The route prediction device according to Embodiment 6 of the present invention outputs a plurality of routes having different parameter values to the target-side route evaluation function used by the target-side optimum route generation unit 8 as a predicted route. Has an effect that it is possible to assume various cases as a route when the target observation data is missing.

DESCRIPTION OF SYMBOLS 1 Sensor part, 2 Target information acquisition part, 3 Observation information storage part, 4 Terrain information database, 5 Monitoring side radar information database, 6 Monitoring side important base database, 7 Target side path evaluation function parameter value storage part, 8 Target side optimal A route generation unit; 9 a target side route evaluation function parameter value adjustment unit; and 10 a route prediction system task execution unit.

Claims (6)

Sensor unit that observes the target and outputs the observation data,
A target information acquisition unit that outputs the input observation data as observation information of the target;
An observation information storage unit for storing the observation information obtained sequentially in time series;
Topographic information database in which topographic information is stored in advance,
Monitoring radar information database in which monitoring radar information is stored in advance,
A monitoring important site database in which important data on the monitoring side that is highly likely to be targeted by the above target is stored in advance,
A target-side path evaluation function that stores a parameter value for adjusting a trade-off between each of the above-described elements of a target-side path evaluation function defined by a linear sum of evaluation values of elements that affect the optimality of the path for the target. Parameter value storage,
Based on the data stored in the target position and speed at the start time of route generation, the value of the parameter of the target side route evaluation function, the monitoring important base database and the monitoring radar information database, and the topographic information database A target-side optimal route generation unit that generates a route to the monitoring-side important base that optimizes the target-side route evaluation function as a target-side optimal route;
A series of the target observation information, the adjustment mode, and values of a plurality of parameters having different target side path evaluation functions are input to the target side optimal path generation unit to generate the target side optimal path and the target side optimal path generation The parameter value used in the path most similar to the series of the target observation information among the plurality of different paths generated by the unit is changed to the parameter value stored in the target side path evaluation function parameter value storage unit. Target side path evaluation function parameter value adjustment unit to be updated,
The observation information stored in the observation information storage unit is monitored and analyzed, and while the observation data is obtained, the target side path evaluation function parameter value adjustment unit is periodically operated based on the analysis result of the observation information. When the observation data is missing, the parameter value stored in the target side path evaluation function parameter value storage unit is input to the target side optimal path generation unit to generate the target side optimal path. a path prediction apparatus including a path prediction system task execution unit,
The target side path evaluation function parameter value adjustment unit is configured to provide a plurality of target side path evaluation functions to be used for generating the target side optimal path based on the analysis result of the observation information while the observation data is obtained. A route prediction apparatus that determines whether or not to adjust a parameter value by focusing on a part of the parameters, and adjusts all or a part of the parameter values based on the determination . The target side optimum route generating unit, when the upper Symbol goals have Tsu Do a detectable again in the monitoring side radar path prediction apparatus according to claim 1, characterized in that to terminate the route generation utilized. The target side optimum route generating section, when the target raw form path reaches a set route generated end time when flying from the current time, and terminates the route generation utilized,
The target side route evaluation function parameter value adjustment unit inputs a route generation end time to the target side optimum route generation unit so as to generate only a route part necessary for parameter adjustment processing. 1. The route prediction apparatus according to 1.   If the observation data is missing even after outputting the predicted route, the route prediction system task execution unit adjusts the parameter value by executing the target side route evaluation function parameter value adjustment unit again, and the parameter value The route prediction apparatus according to claim 1, wherein the route obtained by executing the target side optimum route generation unit is output to the outside again as a predicted route.   The route prediction apparatus according to claim 1, further comprising a function that allows a user to set a parameter value stored in the target-side route evaluation function parameter value storage unit from outside. When the route prediction system task execution unit lacks observation data and executes the target side optimal route generation unit to obtain a prediction route, the parameter value stored in the target side route evaluation function parameter value storage unit And a plurality of parameter values for generating a plurality of routes different from the parameter values are input to the target-side optimal route generation unit and a plurality of routes generated by the target-side optimal route generation unit are obtained as predicted routes The route prediction device according to claim 1, wherein the route prediction device outputs the route.

Images