Allocation determination apparatus, allocation determination method, and computer-readable medium
When determining, from a plurality of sensors ( 10 ), a sensor ( 10 ) for observing a plurality of moving objects ( 20 ), based on positions of the plurality of moving objects ( 20 ), an allocation determination apparatus ( 2000 ) executes annealing on an allocation determination model of which value is larger as the number of the moving objects ( 20 ) that are not observed by any of the plurality of sensors ( 10 ) is larger, and of which value is smaller as the number of the moving objects ( 20 ) that are not observed by any of the plurality of sensors ( 10 ) is smaller, and thereby determines allocation of the moving object ( 20 ) to each of the sensors ( 10 ) in a case in which the value of the allocation determination model decreases.
1 . An allocation determination apparatus comprising:
at least one memory storing instructions; and
at least one processor that is configured to execute the instructions to:
generate a model on which annealing is executed to allocate, for each of a plurality of moving objects, a sensor that observes the moving object from a plurality of the sensors based on a position of each one of the moving objects,
wherein a value of the model is larger as a number of the moving objects that are not observed by any of the plurality of sensors is larger, and
wherein a value of the model is smaller as the number of the moving objects that are not observed by any of the plurality of sensors is smaller;
wherein a direction in which the sensor can observe is divided into a plurality of sections,
wherein the value of the model is larger as the section to be observed by the sensor is farther from the section being observed by the sensor, and
wherein the value of the model is smaller as the section to be observed by the sensor is closer to the section being observed by the sensor.
2 . The allocation determination apparatus according to claim 1 ,
wherein the model includes variables that represent whether or not each of the moving objects is observed by the sensor, and
wherein the annealing on the model is executed to determine values of the variables.
3 . The allocation determination apparatus according to claim 1 ,
wherein a direction in which the sensor can observe is divided into a plurality of sections, and
wherein the value of the model is larger as a number of the sensors that are allocated to the moving object that is present in the section that is not observed by the sensor is larger.
4 . The allocation determination apparatus according to claim 3 ,
wherein the value of the model is smaller as a number of the sections observed by the sensor is closer to 1, and
wherein the value of the model is larger as the number of the sections observed by the sensor is farther from 1.
5 . The allocation determination apparatus according to claim 1 ,
wherein the at least one processor is configured to further execute:
estimating a position of each of the moving objects at a time point at which allocation of the moving object being determined by the annealing is reflected; and
reflecting the estimated position of each of the moving objects in the model.
6 . The allocation determination apparatus according to claim 1 ,
wherein the value of the model is larger as a difference between the sensors being allocated to each of the moving objects and the sensors to be allocated to each of the moving objects is larger, and
wherein the value of the model is smaller as the difference is smaller.
7 . The allocation determination apparatus according to claim 1 ,
wherein the value of the model is smaller as duration of the sensors being allocated to the moving objects is longer while the value of the model is larger as the duration is shorter.
8 . An allocation determination method, which is performed by a computer, comprising: generating a model on which annealing is executed to allocate, for each of a plurality of moving objects, a sensor that observes the moving object from a plurality of the sensors based on a position of each one of the moving objects,
wherein a value of the model is larger as a number of the moving objects that are not observed by any of the plurality of sensors is larger, and
wherein a value of the model is smaller as the number of the moving objects that are not observed by any of the plurality of sensors is smaller,
wherein a direction in which the sensor can observe is divided into a plurality of sections,
wherein the value of the model is larger as the section to be observed by the sensor is farther from the section being observed by the sensor, and
wherein the value of the model is smaller as the section to be observed by the sensor is closer to the section being observed by the sensor.
9 . A non-transitory computer-readable medium storing a program causing a computer to execute: generating a model on which annealing is executed to allocate, for each of a plurality of moving objects, a sensor that observes the moving object from a plurality of the sensors based on a position of each one of the moving objects,
wherein a value of the model is larger as the number of the moving objects that are not observed by any of the plurality of sensors is larger, and
wherein a value of the model is smaller as the number of the moving objects that are not observed by any of the plurality of sensors is smaller,
wherein a direction in which the sensor can observe is divided into a plurality of sections,
wherein the value of the model is larger as the section to be observed by the sensor is farther from the section being observed by the sensor, and
wherein the value of the model is smaller as the section to be observed by the sensor is closer to the section being observed by the sensor.
10 . The allocation determination method according to claim 8 ,
wherein the model includes variables that represent whether or not each of the moving objects is observed by the sensor, and
wherein the annealing on the model is executed to determine values of the variables.
11 . The allocation determination method according to claim 8 ,
wherein a direction in which the sensor can observe is divided into a plurality of sections, and
wherein the value of the model is larger as a number of the sensors that are allocated to the moving object that is present in the section that is not observed by the sensor is larger.
12 . The allocation determination method according to claim 11 ,
wherein the value of the model is smaller as a number of the sections observed by the sensor is closer to 1, and
wherein the value of the model is larger as the number of the sections observed by the sensor is farther from 1.
13 . The allocation determination method according to claim 8 , further comprising:
estimating a position of each of the moving objects at a time point at which allocation of the moving object being determined by the annealing is reflected; and
reflecting the estimated position of each of the moving objects in the model.
14 . The medium according to claim 9 ,
wherein the model includes variables that represent whether or not each of the moving objects is observed by the sensor, and
wherein the annealing on the model is executed to determine values of the variables.
15 . The medium according to claim 9 ,
wherein a direction in which the sensor can observe is divided into a plurality of sections, and
wherein the value of the model is larger as a number of the sensors that are allocated to the moving object that is present in the section that is not observed by the sensor is larger.
16 . The medium according to claim 15 ,
wherein the value of the model is smaller as a number of the sections observed by the sensor is closer to 1, and
wherein the value of the model is larger as the number of the sections observed by the sensor is farther from 1.
17 . The medium according to claim 9 , the program causes the computer to further execute:
estimating a position of each of the moving objects at a time point at which allocation of the moving object being determined by the annealing is reflected; and
reflecting the estimated position of each of the moving objects in the model.