Alarm control device and alarm control method

Abstract

The present invention comprises: a determined difficulty degree calculation unit (13) that calculates a determined difficulty degree on the basis of difficulty degree calculation information for calculating the determined difficulty degree, the determined difficulty degree being the degree of difficulty in determining whether an alarm target event is occurring; and an overall determination unit (14) that, according to the determined difficulty degree calculated by the determined difficulty degree calculation unit (13), determines whether to execute or suspend an event occurrence / nonoccurrence determination as to whether or not an alarm target event is occurring, and when it is determined that the event occurrence / nonoccurrence determination is to be executed, executes the event occurrence / nonoccurrence determination on the basis of a moving body detection result, and performs alarm output control on the basis of the result of executing the event occurrence / nonoccurrence determination.

Description

Alarm control device and alarm control method

[0001] The present disclosure relates to an alarm control device and an alarm control method.

[0002]

[0003] Conventionally, a technology has been known in which a sensor, such as a radio wave sensor, transmits a signal toward a space (hereinafter referred to as a "target space") in which a moving object (hereinafter referred to as a "moving object"), including a living organism, may be present, and based on the result of detecting a moving object (hereinafter referred to as a "reflected signal"), the sensor determines that an event requiring an alarm (hereinafter referred to as an "alarm event") has occurred in the target space. For example, Patent Literature 1 discloses an in-vehicle monitoring device that includes a sensor that outputs millimeter-wave radio waves toward a vehicle cabin and detects the millimeter-wave reflected waves from objects inside the vehicle cabin, such as passengers or luggage, and a determination unit that determines the type of object inside the vehicle cabin based on the detection level of the millimeter-wave reflected waves by the sensor, and the determination unit determines whether the object inside the vehicle cabin is a child or luggage based on a trend in the change in the detection level of the millimeter-wave reflected waves by the sensor. When the determination unit determines that a child is present in the vehicle cabin when the passenger exits the vehicle, the in-vehicle monitoring device issues an alarm to the passenger exiting the vehicle that the child has been left behind.

[0003] JP 2022-182340 A

[0004] When attempting to determine whether an alarm-requiring event has occurred in a target space based on a moving object detection result obtained from a signal transmitted from a sensor to the target space and reflected by an object in the target space, such determination can be difficult depending on the conditions of the target space or the surrounding area. Therefore, prior art has had a problem of misjudging the occurrence of an alarm-requiring event, resulting in inappropriate alarm output control. The in-vehicle monitoring device disclosed in Patent Document 1 attempts to increase the accuracy of the object type determination by, for example, determining that an object in the vehicle is luggage if the detection level of the millimeter wave reflected by the sensor decreases over time compared to the detection level immediately after boarding, and determining that the object is a child if the detection level of the millimeter wave reflected by the sensor remains the same even after time has passed. However, the in-vehicle monitoring device has difficulty determining whether an object has been abandoned in a vehicle cabin when both a child and luggage are present, making it difficult to determine whether an object has been abandoned. As a result, a false alarm of abandoned object may be issued. Therefore, the above-mentioned in-vehicle monitoring device still does not solve the above-mentioned problem.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to provide an alarm control device that performs more accurate alarm output control than conventional technology in alarm output control that outputs an alarm when it is determined that an event requiring an alarm has occurred within a target space based on the detection result of a moving object obtained from a reflected signal obtained when a signal transmitted by a sensor toward the target space is reflected by an object within the target space.

[0006] The alarm control device according to the present disclosure is an alarm control device that outputs an alarm when it determines that an alarm-required event requiring the issuance of an alarm has occurred in the target space based on the detection result of a moving object obtained from a reflected signal obtained when a signal transmitted by a sensor toward the target space is reflected by an object in the target space, and is equipped with a judgment difficulty calculation unit that calculates the judgment difficulty based on difficulty calculation information for calculating the judgment difficulty, which is the degree of difficulty of determining whether or not an alarm-required event has occurred, and a comprehensive judgment unit that decides whether to perform an event occurrence determination to determine whether or not an alarm-required event has occurred, or to postpone performing the event occurrence determination, depending on the judgment difficulty calculated by the judgment difficulty calculation unit, and if it is decided to perform the event occurrence determination, performs the event occurrence determination based on the detection result of the moving object, and controls the output of the alarm based on the result of performing the event occurrence determination.

[0007] According to the present disclosure, an alarm control device can perform more accurate alarm output control than conventional technology in outputting an alarm when it determines that an event requiring an alarm has occurred within a target space based on the detection result of a moving object obtained from a reflected signal obtained when a signal transmitted by a sensor toward the target space is reflected by an object within the target space.

[0008] 4A and 4B are diagrams illustrating an example of a moving object map generated by a moving object detection unit in the first embodiment.

[0024] FIG. 4B is a diagram illustrating an example of a mode of an alarm to be output, determined by an alarm output control unit according to the determination difficulty level in the first embodiment.

[0025] FIG. 4A is a diagram illustrating an example of a moving object map generated by a moving object detection unit in the first embodiment.

[0026] FIG. 4B is a diagram illustrating an example of a mode of an alarm to be output, determined by an alarm output control unit according to the determination difficulty level in the first embodiment.

[0027] FIG. 4B is a flowchart illustrating the operation of an alarm control device in the first embodiment.

[0028] FIG. 9A is a flowchart illustrating the details of the processing performed by the comprehensive determination unit in the comprehensive determination processing of step ST4 of FIG. 7.

[0029] FIG. 9B is a diagram illustrating an example of a hardware configuration of an alarm control device in the first embodiment.

[0009] Hereinafter, embodiments will be described in detail with reference to the drawings. Embodiment 1. Fig. 1 is a diagram showing an example of the configuration of an alarm control device 1 according to embodiment 1. The alarm control device 1 is connected to a sensor 2 and an alarm device 3, and the alarm control device 1, the sensor 2, and the alarm device 3 together constitute an alarm control system 4.

[0010] The alarm control device 1 outputs an alarm from the alarm device 3 when it determines that an event requiring an alarm has occurred in the target space (hereinafter referred to as an "alarm event") based on a detection result of a moving object (hereinafter referred to as a "moving object"), including a human or other living organism, in a space (hereinafter referred to as a "target space") where the moving object (hereinafter referred to as a "moving object"), including a living organism such as a human, may be present. The detection result is obtained from a signal (hereinafter referred to as a "reflected signal") that is reflected by an object in the target space. As described above, when attempting to determine whether an alarm event has occurred in the target space based on a detection result of a moving object obtained from a reflected signal of a signal transmitted by the sensor 2 toward the target space and reflected by an object in the target space, the determination may be difficult depending on the conditions of the target space or the surrounding area of ​​the target space. In response to this, the alarm control device 1 according to the first embodiment determines the degree of difficulty of determining whether an alarm event has occurred (hereinafter referred to as an "event occurrence determination"), and determines whether to execute or postpone the determination of whether an alarm event has occurred (hereinafter referred to as an "event occurrence determination") based on the determined determination difficulty. Specifically, when the warning control device 1 determines that the determination difficulty is high, it decides to suspend the determination of whether or not an event has occurred. This prevents a false alarm in which, when it is estimated that it is difficult to determine whether or not an event requiring an alarm has occurred, the warning control device 1 mistakenly determines that an event requiring an alarm has occurred, resulting in the output of an unnecessary alarm. More specifically, the determination difficulty is a degree that indicates how difficult it is to appropriately determine whether or not an event requiring an alarm has occurred.

[0011] In the following first embodiment, as an example, the target space is the interior of a vehicle, in other words, the vehicle interior, and the event to be warned is an infant being left behind. That is, the alarm control system 4 is applied to an infant being left behind detection system in a vehicle interior. The alarm control device 1 is also assumed to be mounted on the vehicle. A detailed configuration example of the alarm control device 1 will be described later.

[0012] The sensor 2 transmits a signal toward the interior of the vehicle and receives a reflected signal that is reflected by an object within the vehicle. The signal transmitted from the sensor 2 may also reach the periphery of the vehicle interior, in other words, the periphery of the vehicle. That is, the sensor 2 may also receive a reflected signal that is reflected by an object outside the vehicle. Such objects include moving objects, such as human bodies. Examples of the sensor 2 include a radio wave sensor capable of transmitting and receiving radio waves, an ultrasonic sensor capable of transmitting and receiving ultrasonic waves, and a near-infrared sensor capable of transmitting and receiving near-infrared rays. This is merely an example, and the sensor 2 may be any sensor capable of transmitting a signal and receiving a reflected signal that is reflected by a moving object. In the first embodiment, the sensor 2 is mounted on the vehicle. Although the first embodiment includes a single sensor 2 as shown in FIG. 1 , this is merely an example. A plurality of sensors 2 may be connected to the alarm control device 1.

[0013] In the following first embodiment, the sensor 2 is assumed to be a radio wave sensor, as an example. The radio wave sensor is assumed to be a general radio wave sensor that detects objects. The sensing signal of the radio wave sensor can be modulated using various modulation methods, such as the well-known FM-CW (Frequency Modulation - Continuous Wave) method, Fast Chirp Modulation method, or pulse method. Here, the modulation method of the sensor 2 will be described using the Fast Chirp Modulation method as an example. In the Fast Chirp Modulation method, a radio wave transmitting / receiving unit (not shown) included in the sensor 2 periodically generates radio waves (chirps) whose frequency varies at a constant gradient. The radio wave transmitting / receiving unit transmits the generated radio waves multiple times via a transmitting antenna (not shown) toward a target space, in this case, the vehicle interior. When radio waves transmitted toward the vehicle interior reach an object within the range of the radio waves, a portion of the waves is reflected by the surface of the object and returns to the radio wave transmitter / receiver. Here, the object may be, for example, a person in the vehicle interior, luggage placed in the vehicle interior, or a vehicle structure, from which the radio waves are reflected. The radio wave transmitter / receiver receives the radio waves (reflected waves), which are signals reflected by the surface of the object, via a receiving antenna (not shown). The radio wave transmitter / receiver obtains the difference between the transmission frequency and the reception frequency, i.e., a beat signal, and performs AD conversion on the beat signal from an analog signal to a digital signal. The radio wave transmitter / receiver obtains the AD-converted waveform obtained during the AD conversion process as a received signal. The received signal received by the sensor 2 is output to the alarm control device 1. For example, the received signal obtained by the sensor 2 may be stored in a data storage unit (not shown), and the alarm control device 1 may acquire the received signal from the data storage unit.

[0014] The alarm device 3 outputs an alarm based on instructions from the alarm control device 1. The alarm device 3 may be, for example, an audio output device such as a horn mounted on a vehicle, a display device mounted on a vehicle, or a flashing device such as a turn signal mounted on a vehicle. Note that this is merely an example, and the alarm device 3 may be any device capable of outputting an alarm. For example, the alarm device 3 may be an audio output device, display device, or flashing device mounted on a mobile terminal carried by a person, or an audio output device, display device, or flashing device installed in a home, office, etc. The alarm device 3 can output various alarms. For example, if the alarm device 3 is an audio output device, the alarm device 3 can output an alarm sound or warning voice. Also, for example, if the alarm device 3 is a display device, the alarm device 3 can display a warning message. Also, for example, if the alarm device 3 is a flashing device, the alarm device 3 can output an alarm by flashing. Note that while only one alarm device 3 is illustrated in FIG. 1 , this is merely an example. Multiple alarm devices 3 can be connected to the alarm control device 1.

[0015] An example configuration of an alarm control device 1 according to the first embodiment will be described. As shown in FIG. 1 , the alarm control device 1 includes a moving object detection unit 11, a difficulty level calculation information acquisition unit 12, a judgment difficulty level calculation unit 13, and a comprehensive judgment unit 14. The judgment difficulty level calculation unit 13 includes an out-of-space detection unit 131, a vibration detection unit 132, an interference detection unit 133, a nearby object detection unit 134, and a moving object size detection unit 135. Note that, although the judgment difficulty level calculation unit 13 includes the out-of-space detection unit 131, the vibration detection unit 132, the interference detection unit 133, the nearby object detection unit 134, and the moving object size detection unit 135, this is merely an example. It is sufficient for the judgment difficulty level calculation unit 13 to include at least one of the out-of-space detection unit 131, the vibration detection unit 132, the interference detection unit 133, the nearby object detection unit 134, and the moving object size detection unit 135. The comprehensive determination unit 14 includes a determination necessity determination unit 141 , a determination execution unit 142 , and an alarm output control unit 143 .

[0016] The moving object detection unit 11 extracts moving objects based on the received signals obtained by the sensor 2. The moving object detection unit 11 may extract moving objects based on the received signals obtained by the sensor 2 using a known signal processing technique. For example, if the sensor 2 is a radio wave sensor using the Fast Chirp Modulation method, the moving object detection unit 11 performs MTI processing, which applies a known MTI (Moving Target Indicator) filter to the received signals obtained by the sensor 2. By applying the MTI filter, reflection components with a sufficiently small relative velocity (e.g., reflection components with a velocity of zero) due to objects such as completely stationary structures, such as seats in a vehicle, are removed, and reflection components due to moving objects are extracted. In the MTI processing, the moving object detection unit 11 extracts components that do not change between chirps by, for example, obtaining the average value of the received signals for each chirp over a sufficiently long period, and then subtracts the extracted signal from the received signals to extract the reflection components of the moving objects. The longer the period over which the average is taken, the more reflection components with smaller movements can be extracted. The reflection components of the received signal from a moving object are expressed in terms of reflection intensity.

[0017] When the moving object detection unit 11 extracts a moving object, it performs frequency analysis on the received signal obtained by the sensor 2 to extract the distance, angle, speed, and other parameters to the moving object. After performing MTI, the moving object detection unit 11 can detect the distance (meters) of the moving object by, for example, performing a distance-direction FFT (Fast Fourier Transform). The moving object detection unit 11 can also identify the angle at which the moving object is located by, for example, performing the above-described signal processing on the received signal from each antenna in the sensor 2 and performing a known direction-of-arrival estimation process, such as DBF (Digital Beam Forming) processing. The moving object detection unit 11 can also calculate the speed of the moving object by utilizing the Doppler effect, which states that the frequency of a reflected wave from an object increases when the object is moving toward the radio wave transmitting and receiving unit of the sensor 2 and decreases when the object is moving away from the radio wave transmitting and receiving unit of the sensor 2. The moving object detection unit 11 detects, for example, changes in the frequency components of the received signal obtained by the sensor 2. Then, based on the principle of the Doppler effect, the moving object detection unit 11 calculates the speed of the moving object from the changes in the frequency components of the detected received signal. In this way, the moving object detection unit 11 can extract information such as the distance, angle, speed, and reflection intensity of the moving object (hereinafter referred to as "moving object information") from the received signal obtained by the sensor 2 based on various known methods or procedures.

[0018] The moving object detection unit 11 can also generate a distance profile based on the moving object information. In the first embodiment, the distance profile is information indicating the relationship between the distance from the sensor 2 and the reflection intensity of the reflected signal corresponding to that distance. That is, the distance profile reveals the power distribution in the distance direction. As described above, the moving object detection unit 11 performs known MTI processing or the like to extract the reflection components caused by the moving object. The moving object detection unit 11 can generate a distance profile based on the extracted reflection components. For example, the moving object detection unit 11 performs frequency analysis on the reflection components caused by the moving object by performing FFT or the like in the distance direction. For example, in a radio wave sensor using the Fast Chirp Modulation method, due to its principle, FFT results in the distance direction are obtained in the number of chirps multiplied by the number of receiving antennas. The moving object detection unit 11 performs known incoherent processing on the results of the frequency analysis to obtain the power distribution in the distance direction and generate a distance profile. In the distance profile, the relationship between the reflection intensity of the reflected signal from the moving object, i.e., the reflection component and the distance, is shown as a waveform. The details of the distance profile will be described later with reference to the drawings.

[0019] The moving object detection unit 11 can also generate a distance profile for each angle of arrival of the reflected signal, for example. For example, if the sensor 2 is a radio wave sensor having multiple receiving antennas or multiple radio wave transmitters and receivers installed so as to be able to receive reflected waves from different directions of arrival, the moving object detection unit 11 can extract, as described above, reflected components caused by a moving object for each angle of arrival of the reflected wave from the reflected signals received by the multiple radio wave transmitters and receivers. In this case, the moving object detection unit 11 extracts the reflected components caused by the moving object for each angle of arrival of the reflected signal by using, for example, known signal processing such as MTI processing. Then, the moving object detection unit 11 generates a distance profile for each angle of arrival of the reflected signal based on the extracted reflected components caused by the moving object for each angle of arrival of the reflected signal.

[0020] The moving object detection unit 11 can also generate a moving object map based on the moving object information. In the first embodiment, the moving object map is information representing the moving objects present in a target space, in this case, the vehicle cabin, as a three-dimensional spatial distribution. In other words, the moving object map represents the distribution of the area in the vehicle cabin where the moving objects exist in three dimensions. In more detail, the moving object map represents the minute movements of the moving objects in the vehicle cabin, in other words, the objects that reflected the radio waves transmitted by the sensor 2, using multiple grids that correspond to the reflection points of the radio waves in three-dimensional space. Because there are multiple reflection points on the surface of the object, the moving object detection unit 11 can generate a three-dimensional spatial distribution of the object based on the moving object information extracted by signal processing of the reflected waves from the multiple reflection points. Details of the moving object map will be described later using drawings.

[0021] In this way, the moving object detection unit 11 extracts moving objects using known signal processing techniques from the received signals obtained by the sensor 2, and can obtain moving object information such as the distance, angle, speed, and reflection intensity of the moving object, a distance profile, or a moving object map.

[0022] The moving object detection unit 11 outputs the moving object detection result to the difficulty calculation information acquisition unit 12 and the comprehensive judgment unit 14. The moving object detection result includes information indicating whether a moving object has been detected, and, if a moving object has been detected, information about the detected moving object (hereinafter referred to as "detected moving object related information"). The detected moving object related information includes moving object information, a distance profile, or a moving object map. Note that even if a moving object is detected, the detected moving object related information does not need to include all of the moving object information, the distance profile, and the moving object map. The signal processing that the moving object detection unit 11 can execute is appropriately selected and implemented depending on the target to be observed. The moving object detection unit 11 obtains detected moving object related information including moving object information, a distance profile, or a moving object map by executing the implemented signal processing. If the moving object detection unit 11 does not detect the presence of a moving object, more specifically, if the result of performing known signal processing on the received signal obtained by the sensor 2 is that no moving object is extracted, the moving object detection unit 11 outputs information indicating that no moving object has been detected to the difficulty level calculation information acquisition unit 12 and the overall judgment unit 14 as the moving object detection result.

[0023] The difficulty level calculation information acquisition unit 12 acquires a moving object detection result from the moving object detection unit 11. Furthermore, when the difficulty level calculation information acquisition unit 12 determines that the moving object detection unit 11 has detected the presence of a moving object based on the moving object detection result acquired from the moving object detection unit 11, the difficulty level calculation information acquisition unit 12 acquires information for calculating the difficulty level of judgment (hereinafter referred to as "difficulty level calculation information"). The judgment difficulty level is calculated by the judgment difficulty calculation unit 13. Details of the judgment difficulty calculation unit 13 will be described later. For example, the function of the difficulty level calculation information acquisition unit 12 may be possessed by the judgment difficulty calculation unit 13. In this case, it is not essential for the alarm control device 1 to be provided with the difficulty level calculation information acquisition unit 12.

[0024] For example, the difficulty level calculation information acquisition unit 12 acquires, as difficulty level calculation information, moving object information, a distance profile, or a moving object map from the moving object detection unit 11. The moving object information, distance profile, or moving object map is included in the moving object detection result output from the moving object detection unit 11. Furthermore, for example, the difficulty level calculation information acquisition unit 12 acquires, as difficulty level calculation information, acceleration information indicating acceleration detected by an acceleration sensor (not shown) mounted on the vehicle. The type of information that the difficulty level calculation information acquisition unit 12 acquires as difficulty level calculation information is determined in advance by an administrator or the like depending on how the judgment difficulty calculation unit 13 calculates the judgment difficulty. It is sufficient that the difficulty level calculation information acquisition unit 12 acquires, as difficulty level calculation information, information used by the judgment difficulty calculation unit 13 to calculate the judgment difficulty. The difficulty level calculation information acquisition unit 12 outputs the acquired difficulty level calculation information to the judgment difficulty calculation unit 13.

[0025] The judgment difficulty calculation unit 13 calculates the judgment difficulty based on the difficulty calculation information acquired by the difficulty calculation information acquisition unit 12. The judgment difficulty calculation unit 13 may calculate the judgment difficulty as, for example, a binary value of "high" or "low," or may calculate it in multiple stages. Here, as shown in FIG. 1 , the judgment difficulty calculation unit 13 includes an out-of-space detection unit 131, a vibration detection unit 132, an interference detection unit 133, a nearby object detection unit 134, and a moving object size detection unit 135. The out-of-space detection unit 131 detects whether a moving object may be present outside the vehicle based on the difficulty calculation information. The vibration detection unit 132 detects vibration within the vehicle cabin based on the difficulty calculation information. The interference detection unit 133 detects whether there is interference with the detection of the moving object by the sensor 2 based on the difficulty calculation information. The nearby object detection unit 134 detects an obstruction based on the difficulty calculation information. In the first embodiment, the obstructing object detected by the nearby object detection unit 134 is assumed to be an object or structure that is close to the sensor 2, in other words, that is located within a preset threshold (hereinafter referred to as the "obstructing object determination threshold") from the sensor 2. The obstructing object determination threshold is set in advance by an administrator or the like. The moving object size detection unit 135 determines the size of the moving object based on the difficulty level calculation information, and detects the amount of change in the determined size of the moving object.

[0026] The judgment difficulty calculation unit 13 calculates the judgment difficulty based on the detection result of whether or not there is a possibility of a moving object being present outside the vehicle detected by the outside-space detection unit 131, the detection result of shaking within the vehicle cabin detected by the shaking detection unit 132, the detection result of whether or not there is interference detected by the interference detection unit 133, the detection result of an obstruction detected by the nearby object detection unit 134, or the detection result of the change in size of the moving object detected by the moving object size detection unit 135.

[0027] Here, the details of the calculation of the judgment difficulty by the judgment difficulty calculation unit 13 based on various detection results by the out-of-space detection unit 131, the vibration detection unit 132, the interference detection unit 133, the nearby object detection unit 134, or the moving object size detection unit 135 will be explained using several examples, along with details of the difficulty calculation information and various detection methods used by the out-of-space detection unit 131, the vibration detection unit 132, the interference detection unit 133, the nearby object detection unit 134, or the moving object size detection unit 135 to perform various detections.

[0028] First, we will explain, using an example, how the outside-space detection unit 131 detects whether there is a possibility that a moving object is outside the vehicle, and how the judgment difficulty calculation unit 13 calculates the judgment difficulty based on the detection result of whether there is a possibility that a moving object is outside the vehicle detected by the outside-space detection unit 131.

[0029] <Example (1) of Determination Difficulty Calculation> For example, the out-of-space detection unit 131 can detect whether there is a possibility that a moving object is outside the vehicle based on the distance profile. In this case, the difficulty level calculation information is a distance profile. The difficulty level calculation information acquisition unit 12 acquires the distance profile as the difficulty level calculation information. Note that the difficulty level calculation information acquisition unit 12 can acquire the distance profile from the moving object detection unit 11. Based on the distance profile, the out-of-space detection unit 131 detects whether there is a possibility that a moving object is outside the vehicle depending on whether a peak in reflection intensity appears within a distance range that can be considered to be outside the vehicle.

[0030] FIG. 2 is a diagram illustrating the concept of the distance profile in the first embodiment. As shown in FIG. 2, the distance profile shows a power distribution in the distance direction. Specifically, the distance profile shows a waveform representing the relationship between the reception intensity of the reflected component of the signal reflected by a moving object and the distance. Note that in FIG. 2, the distance profile is indicated by "P." In FIG. 2, the interior of the vehicle corresponding to the distance in the distance profile, as viewed from the left side in the direction of vehicle travel, is shown above the distance profile in FIG. 2, with the distance from sensor 2 coinciding with the distance on the distance profile. Here, as an example, it is assumed that sensor 2 is mounted on an overhead console (not shown) in the vehicle interior and transmits radio waves from the overhead console toward the interior of the vehicle. FIG. 2 shows a situation in which there is no person inside the vehicle interior, but there are people outside the vehicle around the vehicle.

[0031] In the distance profile, the distance indicated by the peak position coincides with the distance of a straight line connecting the sensor 2 to the moving object. Note that in the first embodiment, "coincide" is not limited to an exact match, but also includes an approximate match. In other words, it is estimated that the moving object is located at the distance indicated by the peak position in the distance profile.

[0032] The outside-space detection unit 131 detects that a moving object may be outside the vehicle if the distance indicated by the peak position in the distance profile corresponds to the distance outside the vehicle cabin, in other words, if a waveform peak appears in the distance profile at a distance corresponding to the distance outside the vehicle cabin. The distance that corresponds to the distance outside the vehicle cabin is determined in advance by an administrator or the like based on the installation location of the sensor 2, the output power of the radio waves from the sensor 2, the characteristics of the antenna, etc. As described above, the radio waves transmitted from the sensor 2 toward the vehicle cabin may also reach the surrounding area outside the vehicle. The distance profile may also indicate the reflection intensity of the reflected signal from the moving object outside the vehicle.

[0033] For example, if the judgment difficulty is calculated as "high" or "low," the judgment difficulty calculation unit 13 determines that the judgment difficulty is "high" if the outside-space detection unit 131 detects that a moving object may be outside the vehicle.

[0034] Here, it is difficult to determine whether a moving object present at a certain distance is inside or outside the vehicle cabin based on distance alone. For example, if a vehicle occupant is present in the vehicle cabin at a distance from sensor 2 equal to the distance from sensor 2 of the person shown in FIG. 2 , a peak may appear in the distance profile at the distance shown in FIG. 2 . Note that in the first embodiment, "same" does not necessarily mean "exactly the same" but includes "approximately the same." If, in reality, there is only a person outside the vehicle, but it is determined that there is an occupant in the vehicle cabin at the same distance from sensor 2 as the person outside the vehicle, and the occupant is an infant, it is determined that an infant has been left behind, and an alarm is output, this would result in an unnecessary alarm being output to people around the vehicle. Such an alarm is essentially unnecessary. For example, if an alarm is output by sound, light, horn, or flashing lights, unnecessary sound, light, horn, or flashing lights would be output around the vehicle. On the other hand, there is a possibility that a child inside the vehicle may be determined to be a person outside the vehicle, and even if a child has been left behind, this may not be determined, and an alarm that should be issued may not be issued.

[0035] In this way, when a waveform peak appears in the distance profile at a distance corresponding to the outside of the vehicle, it is difficult to determine whether the peak indicates a moving object outside the vehicle or a moving object inside the vehicle cabin, and as a result, it becomes difficult to determine whether an event requiring a warning, in this case, an infant being left behind, has occurred. The determination of whether an infant has been left behind is made by the comprehensive determination unit 14. Details of the comprehensive determination unit 14 will be described later. Therefore, when the outside-space detection unit 131 detects that there is a possibility that a moving object is outside the vehicle, the determination difficulty calculation unit 13 determines that the determination difficulty is "high."

[0036] For example, the determination difficulty calculation unit 13 can calculate the determination difficulty in stages based on the detection result of the out-of-space detection unit 131 as to whether or not a moving object is likely to be outside the vehicle. For example, the out-of-space detection unit 131 detects whether or not a moving object is likely to be outside the vehicle based on the distance profile, from the number of reflection intensity peaks that appear within a distance range that can be considered outside the vehicle, or the strength of reflection indicated by the reflection intensity corresponding to the distance that can be considered outside the vehicle. The determination difficulty calculation unit 13 increases the determination difficulty as the number of peaks detected by the out-of-space detection unit 131 increases. It is estimated that the greater the number of peaks, the more difficult it is to determine whether or not a child has been left behind in the vehicle. Furthermore, the determination difficulty calculation unit 13 may increase the determination difficulty as the strength of reflection indicated by the reflection intensity detected by the out-of-space detection unit 131 increases. The stronger the reflection strength indicated by the reflection intensity, the more likely it is that even if there is a moving object inside the vehicle, the reflected signal from the moving object inside the vehicle will be drowned out by the reflected signal from a moving object outside the vehicle, making it more difficult to determine whether or not a child has been left behind inside the vehicle.

[0037] The judgment difficulty calculation unit 13 outputs information indicating the calculated judgment difficulty (hereinafter referred to as “judgment difficulty information”) to the comprehensive judgment unit 14 .

[0038] <Determination Difficulty Calculation Example (2)> For example, the outside-space detection unit 131 may detect whether there is a possibility that a moving object is present outside the vehicle based on the distance profile using a method different from the above-mentioned "Determination Difficulty Calculation Example (1)". Specifically, the outside-space detection unit 131 may detect whether there is a possibility that a moving object is present outside the vehicle based on the distance profile from the gradient of attenuation of reflection intensity with respect to the distance direction in the distance profile.

[0039] 3A and 3B are diagrams illustrating the different characteristics represented by the distance profile when a person is present inside the vehicle cabin and when a person is present outside the vehicle. In FIGS. 3A and 3B, the distance profile is indicated by "P." In FIGS. 3A and 3B, the interior of the vehicle corresponding to the distance in the distance profile, as viewed from the left side in the direction of vehicle travel, is illustrated on the distance profile, with the distance from sensor 2 being consistent with the distance on the distance profile. Here, as an example, sensor 2 is mounted on an overhead console (not shown) in the vehicle cabin, and radio waves are transmitted from the overhead console toward the interior of the vehicle cabin. In FIG. 3A, a moving object (a person in this case) is shown inside the vehicle cabin, while in FIG. 3B, a moving object (a person in this case) is shown outside the vehicle.

[0040] For example, as shown in the upper part of FIG. 3A , when radio waves transmitted by sensor 2 are reflected by a moving object present in the vehicle cabin, the distance profile, as shown in the lower part of FIG. 3A , exhibits a peak at a position corresponding to the linear distance from sensor 2 to the moving object, after which the reflection intensity gradually attenuates in the distance direction. In other words, in this case, the gradient of attenuation of the reflection intensity in the distance direction becomes smaller. This is because the radio waves transmitted by sensor 2 are repeatedly reflected by various objects in the vehicle cabin, including the moving object, resulting in multipath propagation, which makes them more likely to be received by sensor 2 via multiple propagation paths. As the propagation distance increases, the multipath propagation position becomes further behind the peak position, resulting in lower reception intensity. Furthermore, the more times a multipath propagation is reflected, the longer the propagation distance.

[0041] On the other hand, for example, as shown in the upper part of Fig. 3B, when the radio waves transmitted by sensor 2 are reflected by a moving object outside the vehicle, the distance profile has a peak at a position corresponding to the linear distance from sensor 2 to the moving object, and thereafter, the reflection intensity in the distance direction exhibits a characteristic of rapidly attenuating, as shown in the lower part of Fig. 3B. That is, in this case, the gradient of attenuation of the reflection intensity in the distance direction becomes large. This is because there is no influence of multipath as described above.

[0042] Therefore, in this case, the out-of-space detection unit 131 can detect the presence of a moving object inside the vehicle cabin when the gradient is gentle (for example, smaller than a predetermined value), and can detect the presence of a moving object outside the vehicle when the gradient is steep (for example, larger than a predetermined value). Note that the threshold value for determining whether the gradient is small or large can be set appropriately.

[0043] The outside-space detection unit 131 detects whether a moving object is present inside or outside the vehicle by, for example, comparing the slope of attenuation of reflection intensity with respect to the distance direction in the distance profile with a preset condition (hereinafter referred to as a "gradient determination condition"). The gradient determination condition is set in advance by an administrator or the like, and information indicating the determination condition is stored in an internal buffer or the like of the outside-space detection unit 131. The gradient determination condition may, for example, be set as follows: "When the slope of attenuation of reflection intensity is smaller than a predetermined value, a moving object is detected to be present inside the vehicle, and when the slope is larger than the predetermined value, a moving object is detected to be present outside the vehicle." Furthermore, the outside-space detection unit 131 may, for example, determine the feature amount of the shape of the distance profile using a known machine learning determination technique to detect whether a moving object is present inside or outside the vehicle. The outside-space detection unit 131 may be capable of detecting the possibility of a moving object being present outside the vehicle based on the different characteristics of the distance profile between when a moving object is present inside the vehicle cabin and when a moving object is present outside the vehicle, as described with reference to FIGS. 3A and 3B . Regarding moving objects outside the vehicle, for example, surrounding structures may be detected as moving objects outside the vehicle due to the vehicle's movement while parked. An example of such a case would be when the vehicle moves up and down after being parked in a mechanical parking lot. The moving objects outside the vehicle detected by the outside-space detection unit 131 may also include structures around the vehicle that may be detected as moving objects due to the vehicle's movement. Furthermore, in the above specific example, a person was cited as an example of a moving object outside the vehicle, but this is merely an example. Moving objects outside the vehicle that are extracted based on the received signal obtained by sensor 2 may include, in addition to a person outside the vehicle, other vehicles, plants, automatic car washes, etc.

[0044] For example, if the determination difficulty is calculated as "high" or "low," the determination difficulty calculation unit 13 determines the determination difficulty to be "high" when the out-of-space detection unit 131 detects that a moving object may be outside the vehicle. More specifically, as shown in FIG. 3B , when the gradient of attenuation of reflection intensity in the distance direction in the distance profile is large, the determination difficulty calculation unit 13 determines the determination difficulty to be "high." If a moving object is present outside the vehicle, the reflected signal from the moving object becomes a disturbance, increasing the likelihood that the moving object outside the vehicle will be mistakenly detected as a moving object inside the vehicle cabin, making it difficult to determine whether or not an infant has been left behind inside the vehicle. Furthermore, even if a moving object is present inside the vehicle cabin, the reflected signal from the moving object inside the vehicle cabin is likely to be obscured by the reflected signal from the moving object outside the vehicle, making it difficult to determine whether or not an infant has been left behind inside the vehicle.

[0045] 3A and 3B, for convenience, the magnitude of the gradient of the attenuation of reflection intensity with respect to the distance direction in the distance profile has been described using an example in which a moving object is present either inside or outside the vehicle. However, this is merely an example. For example, depending on the situation inside or outside the vehicle, the gradient of the attenuation of reflection intensity with respect to the distance direction in the distance profile may be intermediate between the gradient shown in FIG. 3A and the gradient shown in FIG. 3B. In such a case, the judgment difficulty calculation unit 13 may determine the judgment difficulty to be "high." In this case, for example, the administrator or the like may determine in advance the intermediate magnitude of the gradient that determines the judgment difficulty to be "high," and the administrator or the like may set the gradient judgment condition to indicate that the moving object cannot be determined to be present either inside or outside the vehicle. If the outside-space detection unit 131 detects that it cannot be determined that the moving object is present either inside or outside the vehicle based on the gradient judgment condition, the judgment difficulty calculation unit 13 may determine the judgment difficulty to be "high." Furthermore, for example, if the out-of-space detection unit 131 determines the shape feature of the distance profile using a machine learning determination technique, the out-of-space detection unit 131 may learn the shape feature of the distance profile that indicates that it is not possible to determine whether a moving object is present inside or outside the vehicle. Furthermore, as described above, the determination difficulty calculation unit 13 may calculate the determination difficulty in stages. In this case, for example, the determination difficulty calculation unit 13 may determine the determination difficulty so that the determination difficulty increases as the gradient of the attenuation of the reflection intensity in the distance direction on the distance profile approaches the middle between the gradient when a moving object is detected to be present outside the vehicle (e.g., see FIG. 3A ) and the gradient when a moving object is detected to be present inside the vehicle (e.g., see FIG. 3B ).

[0046] The judgment difficulty calculation unit 13 outputs judgment difficulty information indicating the calculated judgment difficulty to the comprehensive judgment unit 14 .

[0047] <Determination Difficulty Calculation Example (3)> For example, the outside-space detection unit 131 may detect whether there is a possibility that a moving object is present outside the vehicle based on the distance profile using a method different from the above-described "Determination Difficulty Calculation Example (1)" and "Determination Difficulty Calculation Example (2)." For example, the difficulty level calculation information may be a distance profile and moving object information. The difficulty level calculation information acquisition unit 12 acquires the distance profile and the moving object information as the difficulty level calculation information. Note that the difficulty level calculation information acquisition unit 12 may acquire the distance profile and the moving object information from the moving object detection unit 11. Based on the moving object information, the outside-space detection unit 131 can determine at what distance and from what angle the moving object was detected based on the reflected signal obtained. In other words, the outside-space detection unit 131 can determine whether the moving object was detected at a distance and angle corresponding to the inside of the vehicle cabin or at a distance and angle corresponding to the outside of the vehicle. For example, when the out-of-space detection unit 131 determines that a moving object is outside the vehicle from the peak position of the distance profile and determines from the moving object information that the moving object has been detected at a distance and angle corresponding to the outside of the vehicle, the out-of-space detection unit 131 detects that the moving object may be outside the vehicle. This enables the out-of-space detection unit 131 to achieve out-of-space detection with higher accuracy.

[0048] For example, if the determination difficulty is calculated as "high" or "low," the determination difficulty calculation unit 13 determines the determination difficulty to be "high" when the outside-space detection unit 131 detects that a moving object may be outside the vehicle. If a moving object is present outside the vehicle, the reflected signal from the moving object will act as a disturbance, increasing the likelihood that the moving object outside the vehicle will be mistakenly detected as a moving object inside the vehicle, making it difficult to determine whether or not an infant has been left behind inside the vehicle. Furthermore, even if a moving object is present inside the vehicle, the reflected signal from the moving object inside the vehicle will likely be drowned out by the reflected signal from the moving object outside the vehicle, making it difficult to determine whether or not an infant has been left behind inside the vehicle.

[0049] The combination of the result of the determination based on the distance profile as to whether a moving object is located outside the vehicle and the result of the determination based on the moving object information as to whether a moving object is detected at a distance and angle corresponding to the outside of the vehicle, and the final determination of whether a moving object may be located outside the vehicle, may be appropriately determined depending on the application to be realized, i.e., how to determine whether an alarm event, in this case, an infant left behind in the vehicle cabin, has occurred. For example, to reduce the frequency of erroneously determining that a moving object is located inside the vehicle cabin when there is no infant inside, the outside-space detection unit 131 may detect that a moving object may be located outside the vehicle when either the determination based on the distance profile or the determination based on the moving object information determines that the moving object is located outside the vehicle. Furthermore, to reduce the frequency of erroneously determining that there is no infant inside the vehicle cabin when there is an infant inside, the outside-space detection unit 131 may detect that a moving object may be located outside the vehicle only when both the determination based on the distance profile and the determination based on the moving object information determine that the moving object is located outside the vehicle.

[0050] Furthermore, as described above, the determination difficulty calculation unit 13 may calculate the determination difficulty in stages. When calculating the determination difficulty in stages, the determination difficulty calculation unit 13 may calculate a determination difficulty level based on, for example, the reliability of the detection result that the moving object detected by the outside-space detection unit 131 from the distance profile is likely to be outside the vehicle, and the reliability of the detection result that the moving object detected by the outside-space detection unit 131 from the moving object information is detected at a distance and angle corresponding to being outside the vehicle. In this case, for example, the outside-space detection unit 131 calculates the reliability of the detection result that the moving object detected from the distance profile is likely to be outside the vehicle based on whether the distance profile used for the detection is closer to a pre-learned distance profile indicating that a moving object is outside the vehicle or a pre-learned distance profile indicating that a moving object is inside the vehicle cabin. The closer the distance profile used for the detection is to the pre-learned distance profile indicating that a moving object is outside the vehicle, the higher the reliability of the detection result. Furthermore, for example, for a detection result that a moving object detected from the moving object information has been detected at a distance and angle corresponding to outside the vehicle, the outside-space detection unit 131 calculates the reliability of the detection result that the moving object has been detected at a distance and angle corresponding to outside the vehicle based on the range outside the vehicle indicated by the distance and angle based on the moving object information. The outside-space detection unit 131 increases the reliability of the detection result as the distance and angle based on the moving object information move further away from the boundary toward outside the vehicle. The determination difficulty calculation unit 13 calculates the determination difficulty so that the higher the reliability of the detection result by the outside-space detection unit 131 that a moving object has been detected outside the vehicle, i.e., the higher the possibility that the moving object is outside the vehicle.

[0051] The judgment difficulty calculation unit 13 outputs judgment difficulty information indicating the calculated judgment difficulty to the comprehensive judgment unit 14 .

[0052] Next, an example will be given to explain how the vibration detection unit 132 detects vibrations in the vehicle cabin, and how the judgment difficulty calculation unit 13 calculates the judgment difficulty based on the detection results of vibrations in the vehicle cabin detected by the vibration detection unit 132.

[0053] <Example of Determination Difficulty Calculation (4)> For example, the shaking detection unit 132 can detect whether shaking is occurring in the vehicle cabin based on shaking information related to shaking occurring in the vehicle cabin. In this case, the difficulty level calculation information is shaking information. The difficulty level calculation information acquisition unit 12 acquires the shaking information as the difficulty level calculation information. For example, the difficulty level calculation information acquisition unit 12 may acquire acceleration information indicating acceleration from an acceleration sensor provided in the vehicle cabin as the shaking information. For example, based on the acceleration information, the shaking detection unit 132 may determine whether or not acceleration other than gravity is at a certain level (for example, 1 m / s ２ For example, in the case where the vehicle moves up and down after being parked in a mechanical parking lot as described above, the difficulty-calculation information acquisition unit 12 may acquire acceleration information from an acceleration sensor as shaking information, and the shaking detection unit 132 may detect whether shaking is occurring in the vehicle cabin based on the shaking information.

[0054] Furthermore, the vibration detection unit 132 may detect whether vibration is occurring inside the vehicle cabin from, for example, the frequency components of the received signal detected by the sensor 2. In this case, the difficulty level calculation information acquisition unit 12 acquires, as vibration information, information indicating the frequency components of the received signal from the sensor 2. As described above, the sensor 2 detects changes in the frequency components of the received signal when calculating the speed of a moving object. The difficulty level calculation information acquisition unit 12 simply acquires, from the sensor 2, information indicating the frequency components of the received signal. In other words, the frequency components of the received signal detected by the sensor 2 are the frequency components of vibration. The vibration detection unit 132 can detect whether vibration is occurring inside the vehicle cabin by analyzing the frequency components of the received signal.

[0055] By analyzing the frequency components of the received signal, the vibration detection unit 132 can determine whether the vibration occurring in the vehicle cabin is due to the movement of a living body, such as body movement associated with a person's breathing, or due to the movement of an object. FIG. 4 is a diagram illustrating an example of the difference in the motion components (Doppler spectrum) caused by the movement of a living body and the vibration of an object. FIG. 4A shows an example of the motion component caused by the body movement associated with a person's breathing, and FIG. 4B shows an example of the motion component caused by the vibration of a mobile phone. As shown in FIGS. 4A and 4B, low-frequency components are observed in the motion component caused by the body movement associated with a person's breathing, whereas relatively high-frequency components are observed in the motion component caused by the vibration of a mobile phone compared to the motion component caused by the body movement associated with a person's breathing. Note that while FIG. 4B shows an example of the motion component caused by the vibration of a mobile phone, the motion components caused by the movement of an object, such as the swaying of a hanging toy or the swaying of water in a plastic container, also exhibit relatively high-frequency components compared to the motion component caused by the body movement associated with a person's breathing, similar to the motion component caused by the vibration of a mobile phone. The vibration detection unit 132 extracts the difference in such movement components and can determine whether the vibration occurring in the vehicle cabin is due to the movement of a living body or the vibration of an object.

[0056] Furthermore, in the case of object sway, the amount of sway tends to decrease over time. Therefore, the sway detection unit 132 may, for example, determine whether the sway occurring in the vehicle cabin is due to biological movement or object sway by observing changes in the sway amplitude over time. Furthermore, human movement has various sway components in each part of the body in addition to breathing. Therefore, in the motion component due to biological movement, in addition to a single frequency component, many wide-ranging frequency components may be observed. On the other hand, because object sway is caused by an object vibrating at a predetermined period, a specific frequency component is often strongly observed in the motion component due to object sway. The sway detection unit 132 may detect the sway occurring in the vehicle cabin by analyzing the difference between the motion component due to biological movement and the motion component due to object sway, as described above.

[0057] For example, if the judgment difficulty is calculated as "high" or "low," the judgment difficulty calculation unit 13 determines that the judgment difficulty is "high" if the shaking detection unit 132 detects shaking occurring in the vehicle cabin.

[0058] For example, if the vehicle is shaking, the moving object detection unit 11 may detect the presence of a moving object even when there is no one in the vehicle cabin. Furthermore, for example, if only an object placed in the vehicle cabin is shaking, the moving object detection unit 11 may detect the presence of a moving object. In such cases, it may be difficult to determine whether the shaking of the vehicle or the shaking of an object placed in the vehicle cabin indicates the movement of a moving object in the vehicle cabin. As a result, it may be difficult to determine whether an alarm-requiring event, in this case, an infant being left behind, has occurred. If the shaking of the vehicle or the shaking of an object placed in the vehicle cabin is mistakenly determined to be an infant being left behind, a false alarm may be output due to the mistaken determination, which may result in, for example, an annoying alarm being issued to those around the vehicle. Therefore, for example, when the shaking detection unit 132 detects shaking occurring in the vehicle cabin, the judgment difficulty calculation unit 13 determines the judgment difficulty to be “high.” For example, if an object placed inside the vehicle cabin is shaking, there is a high possibility that the reflected signal from a moving object inside the vehicle cabin will be drowned out by the reflected signal caused by the shaking of the object, making it difficult to determine whether or not a child has been left behind inside the vehicle cabin.

[0059] For example, the judgment difficulty calculation unit 13 may calculate the judgment difficulty in stages based on the detection results of tremors occurring in the vehicle cabin by the tremor detection unit 132. In this case, for example, the tremor detection unit 132 executes tremor detection processes to determine whether tremors are occurring in the vehicle cabin using multiple tremor detection methods such as those described above. The judgment difficulty calculation unit 13 calculates the judgment difficulty in stages according to the number of tremor detection processes that the tremor detection unit 132 has executed that detect tremors occurring in the vehicle cabin. Furthermore, for example, when the tremor detection unit 132 detects tremors occurring in the vehicle cabin, the judgment difficulty calculation unit 13 may calculate the judgment difficulty in stages according to the reflection intensity of the tremors. Note that the tremor detection unit 132 can determine the reflection intensity of the tremors from, for example, moving object information. In this case, the judgment difficulty calculation unit 13 increases the judgment difficulty, for example, as the reflection intensity of the detected tremors increases. It is estimated that the stronger the reflection intensity of the object sway, the more likely it is that if there is an occupant in the vehicle, the occupant will be buried in the object sway and will be difficult to detect.

[0060] Furthermore, for example, when the sway detection unit 132 detects swaying of an object occurring in the vehicle cabin, the judgment difficulty calculation unit 13 may increase the judgment difficulty as the frequency components of the object swaying become more similar to the frequency components of human body swaying. Note that the judgment difficulty calculation unit 13 may determine whether the frequency components of object swaying are similar to the frequency components of human body swaying using a known method. For example, the judgment difficulty calculation unit 13 may determine whether the frequency components of object swaying are similar to the frequency components of human body swaying using a known method using an SVM (Support Vector Machine), or may determine whether the frequency components of object swaying are similar to the frequency components of human body swaying based on a correlation coefficient. When determining whether the frequency components of object swaying are similar to the frequency components of human body swaying based on the correlation coefficient, the determination difficulty calculation unit 13 calculates the correlation coefficient of the frequency components observed this time by the sway detection unit 132 with frequency components such as breathing of the human bodies that have been measured in advance for multiple human bodies, and determines that they are similar if the correlation coefficient is above a certain level (e.g., 0.7 or above).

[0061] The judgment difficulty calculation unit 13 outputs judgment difficulty information indicating the calculated judgment difficulty to the comprehensive judgment unit 14 .

[0062] Next, an example will be described of a method for detecting the presence or absence of interference by the interference detection unit 133 and a calculation of the determination difficulty level by the determination difficulty calculation unit 13 based on the detection result of the presence or absence of interference detected by the interference detection unit 133. Note that in the first embodiment, the interference detected by the interference detection unit 133 with the detection of a moving object by the sensor 2 is assumed to be interference from, for example, an external device (not shown). For example, in the alarm control system 4, in this case, an infant abandoned vehicle detection system, if the sensor 2 is a radio wave sensor in the 60 GHz band, examples of interference with the detection of a moving object by the sensor 2 include a radio wave sensor installed in another vehicle that uses the same frequency band, or wireless communication used in an information device such as a smartphone.

[0063] <Example of Determination Difficulty Calculation (5)> For example, the interference detection unit 133 can detect the presence or absence of interference based on a temporal change in the reflection intensity of the received signal obtained by the sensor 2. In this case, the difficulty calculation information is moving object information. The difficulty calculation information acquisition unit 12 acquires the moving object information as the difficulty calculation information. Note that the difficulty calculation information acquisition unit 12 can acquire the moving object information from the moving object detection unit 11. For example, if interference occurs, spike-like noise is superimposed on the received signal due to the interference. The interference detection unit 133 detects the presence of interference based on a temporal change in the reflection intensity of the received signal, for example, if there is a sudden change in the reflection intensity. Furthermore, for example, if there is interference, the reflection intensity of the received signal increases in a distance range where there would normally not be a strong reflection. The interference detection unit 133 may detect the presence of interference based on the moving object information if the reflection intensity of the received signal increases in a distance range where there would normally not be a strong reflection. Note that the distance range where there would normally not be a strong reflection is set in advance by an administrator or the like.

[0064] For example, the interference detection unit 133 can also detect the presence or absence of interference as described above from the distance profile. For example, the interference detection unit 133 detects the presence of interference when, for example, the waveform in the time-series distance profile changes suddenly. Furthermore, the interference detection unit 133 detects the presence of interference when, in the distance profile, the reflection intensity of the received signal increases in a distance range where there would normally be no strong reflection. In this case, the difficulty level calculation information is the distance profile.

[0065] For example, if the judgment difficulty is calculated as "high" or "low," the judgment difficulty calculation unit 13 determines that the judgment difficulty is "high" if the interference detection unit 133 detects that there is interference.

[0066] For example, if a different external device using the same frequency band as sensor 2 is present in the vicinity of sensor 2, the influence of interference from the external device will be superimposed on the sensing result of sensor 2. This may make it difficult to determine a moving object in the vehicle cabin, and as a result, it may become difficult to determine whether an event requiring a warning, in this case, an abandoned child, has occurred. Therefore, for example, when the interference detection unit 133 detects the presence of interference, the determination difficulty calculation unit 13 determines that the determination difficulty is "high."

[0067] For example, the determination difficulty calculation unit 13 may calculate the determination difficulty in stages based on the detection result of the interference detection unit 133, which indicates whether or not there is interference. For example, if the interference detection unit 133 detects interference, it determines the influence of the detected interference. The interference detection unit 133 may determine the magnitude of the influence of the interference based on, for example, the magnitude and frequency of spike-like noise superimposed on the received signal, the amount of increase in noise in the signal after FFT of the received signal, etc. Note that this is merely an example, and the interference detection unit 133 may determine the influence of the detected interference using any known method for determining the influence of interference. The determination difficulty calculation unit 13 calculates the determination difficulty in stages based on the influence of the interference determined by the interference detection unit 133. For example, the determination difficulty calculation unit 13 calculates the determination difficulty so that the smaller the influence of the interference determined by the interference detection unit 133, the lower the determination difficulty. The determination difficulty calculation unit 13 calculates the determination difficulty so that the greater the influence of the interference determined by the interference detection unit 133, the higher the determination difficulty.

[0068] Possible factors for determining that the influence of interference is small include, for example, the external device being located sufficiently far from the sensor 2, the external device having frequency characteristics different from the frequency characteristics of the sensor 2, and the external device not frequently causing radio wave interference. Possible factors for determining that the influence of interference is large include, for example, the external device being located close to the sensor 2, the external device having frequency characteristics similar to the frequency characteristics of the sensor 2, and the external device frequently causing radio wave interference. Note that, although the sensor 2 is assumed to be a radio wave sensor here, even if the sensor 2 is an ultrasonic sensor or a near-infrared sensor, the interference detection unit 133 can detect interference using a known interference detection method.

[0069] The judgment difficulty calculation unit 13 outputs judgment difficulty information indicating the calculated judgment difficulty to the comprehensive judgment unit 14 .

[0070] Next, an example will be given of a method for detecting the presence or absence of an obstruction by the nearby object detection unit 134, and a method for calculating the judgment difficulty by the judgment difficulty calculation unit 13 based on the detection result of the presence or absence of an obstruction detected by the nearby object detection unit 134.

[0071] <Example of Determination Difficulty Calculation (6)> For example, the nearby object detection unit 134 detects the presence of an obstruction when a reflected signal is detected within a set distance range. The set distance range is set in advance by an administrator or the like. For example, the administrator or the like sets the set distance range to a distance range in which the distance from the sensor 2 is closer than the distance at which an occupant could be present in the vehicle cabin. In this case, the difficulty level calculation information is, for example, moving object information. The difficulty level calculation information acquisition unit 12 acquires the moving object information as the difficulty level calculation information. Note that the difficulty level calculation information acquisition unit 12 can acquire the moving object information from the moving object detection unit 11.

[0072] For example, the nearby object detection unit 134 detects the presence of an obstructing object if it detects, based on the moving object information, that the reflection intensity of the received signal is equal to or greater than a preset threshold (hereinafter referred to as the "proximate object determination threshold") within a set distance range. Note that the nearby object determination threshold is set in advance by an administrator or the like. Note that this is merely an example, and the nearby object detection unit 134 may also detect the presence or absence of an obstructing object using any other known method as long as it is a method that can detect the presence or absence of an obstructing object.

[0073] For example, the nearby object detection unit 134 can also detect the presence or absence of an obstruction as described above from a distance profile. The nearby object detection unit 134 detects the presence of an obstruction when the reflection intensity of the received signal is detected to be equal to or greater than the nearby object determination threshold within a set distance range in the distance profile. In this case, the difficulty level calculation information is the distance profile.

[0074] For example, if the judgment difficulty is calculated as "high" or "low," the judgment difficulty calculation unit 13 determines that the judgment difficulty is "high" if the nearby object detection unit 134 detects that there is an obstruction.

[0075] For example, if some kind of structure is installed or attached in front of the sensor 2, or if an object is floating in the air due to wind or the like, the radio waves transmitted and received by the sensor 2 are blocked, and the radio waves are reflected (including multiple reflections) by the structure. In this case, the moving object detection unit 11 may erroneously detect a moving object due to the radio waves being reflected by the structure, or may be unable to detect the moving object because it is blocked by the structure. Therefore, the determination difficulty calculation unit 13 may determine the determination difficulty to be "high" if, for example, the nearby object detection unit 134 detects the presence of an obstruction. Furthermore, for example, a stationary structure in front of the sensor 2 may act as an obstruction, making it impossible to detect a moving object that is farther away from the sensor 2. The nearby object detection unit 134 may detect the presence of an obstruction when such a stationary structure is present, and the determination difficulty calculation unit 13 may determine the determination difficulty to be "high." In addition, for example, when sensor 2 performs the above-mentioned MTI processing, if a reflection component with a speed of 0 exists within a set distance, the sensor 2 extracts this reflection component from the stationary structure, and the nearby object detection unit 134 detects the stationary structure as an obstruction from the moving object information.

[0076] For example, the determination difficulty calculation unit 13 may calculate the determination difficulty in stages based on the detection result of the nearby object detection unit 134 regarding the presence or absence of an obstructing object. For example, when the nearby object detection unit 134 detects the presence of an obstructing object, the nearby object detection unit 134 estimates the extent of the area obstructed by the obstructing object (hereinafter referred to as the "obstructed area") and determines the size of the estimated area. The determination difficulty calculation unit 13 calculates the determination difficulty so that the determination difficulty increases as the obstructed area determined by the nearby object detection unit 134 becomes larger. The nearby object detection unit 134 may detect the size of the obstructed area based on, for example, the magnitude of the reflection intensity of the obstructing object, or may detect the size of the obstructed area based on the rate at which the reflection intensity of a stationary structure such as a seat or ceiling in the vehicle cabin is reduced due to obstruction. In any case, it is sufficient that the nearby object detection unit 134 can determine the effect of the obstructing object on the sensing performance of the sensor 2—for example, the size of the obstructed area in the above example—and the determination difficulty calculation unit 13 can calculate the determination difficulty in stages based on this.

[0077] The judgment difficulty calculation unit 13 outputs judgment difficulty information indicating the calculated judgment difficulty to the comprehensive judgment unit 14 .

[0078] Next, we will explain an example of how the moving body size detection unit 135 detects the change in size of a moving body, and how the judgment difficulty calculation unit 13 calculates the judgment difficulty based on the change in size of a moving body detected by the moving body size detection unit 135.

[0079] <Example of Determination Difficulty Calculation (7)> ​​For example, the moving object size detection unit 135 detects the amount of change in the size of a moving object determined based on a time-series moving object map. In this case, the difficulty level calculation information is the moving object map. The difficulty level calculation information acquisition unit 12 acquires the moving object map as the difficulty level calculation information. Note that the difficulty level calculation information acquisition unit 12 can acquire the moving object map from the moving object detection unit 11.

[0080] Here, Figures 5A and 5B are diagrams illustrating an example of a moving object map generated by the moving object detection unit 11 in embodiment 1. Figure 5A illustrates an example of a top view of the moving object map, and Figure 5B illustrates an example of a side view of the moving object map. The moving object map illustrated using Figures 5A and 5B is a moving object map generated when an adult is properly seated in the left rear seat in the vehicle interior. Note that, in embodiment 1, a properly seated state refers to a state in which the occupant's posture is not compromised and the occupant is seated deep in the seat along the backrest. In the moving object map, for example, each grid is assigned a numerical value corresponding to the speed of an object present at the grid's position. In particular, the faster the speed of an object present at a given grid, the higher the numerical value assigned to that grid. In other words, in the moving object map, a grid included in a range in which a moving object exhibiting slight movement is present is assigned a higher numerical value than a grid in which no moving object is present. In the moving object map, each grid is assigned a numerical value corresponding to at least one of the object's speed, signal strength, and moving object's position, and all of these numerical values ​​may be assigned. In the moving object map, the grid with a larger assigned numerical value is shown darker. In other words, in the moving object maps shown in Figures 5A and 5B, the grid that is included in the range where a moving object exists is shown darker. In the moving object map, stationary objects, in other words, reflected signal components with a speed of 0, do not appear.

[0081] The moving object size detection unit 135 can detect the approximate size of a moving object present in the vehicle cabin based on the moving object map as shown in Figures 5A and 5B. The moving object size detection unit 135 can then detect the amount of change in the size of the moving object based on the time-series moving object map. For example, the moving object size detection unit 135 may store information about the size of the moving object determined based on the moving object map acquired by the difficulty level calculation information acquisition unit 12 in an internal buffer in time series, and detect the amount of change in the size of the moving object from the stored size of the moving object.

[0082] For example, if the judgment difficulty level is calculated as "high" or "low," the judgment difficulty level calculation unit 13 determines that the judgment difficulty level is "high" when the moving object size detection unit 135 detects that the amount of change in the movement of the moving object is large (for example, equal to or greater than a set threshold (hereinafter referred to as the "threshold for determining the amount of change"). The threshold for determining the amount of change is set in advance by an administrator or the like.

[0083] For example, suppose a child is kicking their legs inside the vehicle. In this case, the moving object detection unit 11 detects an excessively large movement due to the child's kicking. As a result, the moving object may be erroneously determined to be an adult based on the magnitude of the movement. In other words, it may be erroneously determined that a child has been left behind in the vehicle, and a necessary alarm may not be output. When a moving object exhibits abnormal movement, it becomes difficult to determine whether an alarm-requiring event, in this case, the child's abandonment, has occurred. For example, when a child is kicking their legs, the range and size of the darkened grid changes on the time-series moving object map in accordance with the child's kicking. Therefore, when the moving object size detection unit 135 detects a large change in the movement of the moving object (e.g., equal to or greater than the change amount determination threshold), the determination difficulty calculation unit 13 determines the determination difficulty to be "high."

[0084] For example, the determination difficulty calculation unit 13 may calculate the determination difficulty in stages based on the detection result of the amount of change in the movement of the moving object by the moving object size detection unit 135. For example, the determination difficulty calculation unit 13 calculates the determination difficulty so that the determination difficulty increases as the amount of change in the size of the moving object detected by the moving object size detection unit 135 increases.

[0085] The judgment difficulty calculation unit 13 outputs judgment difficulty information indicating the calculated judgment difficulty to the comprehensive judgment unit 14 .

[0086] In this way, the judgment difficulty calculation unit 13 calculates the judgment difficulty based on the detection result of whether or not there is a possibility of a moving object being present outside the vehicle detected by the outside-space detection unit 131, the detection result of shaking within the vehicle cabin detected by the shaking detection unit 132, the detection result of whether or not there is interference detected by the interference detection unit 133, the detection result of an obstruction detected by the nearby object detection unit 134, or the detection result of the change in size of the moving object detected by the moving object size detection unit 135, and outputs the calculated judgment difficulty to the overall judgment unit 14.

[0087] The comprehensive judgment unit 14 determines whether to execute an event occurrence determination to determine whether an event requiring an alarm has occurred or to postpone execution of the event occurrence determination, depending on the judgment difficulty calculated by the judgment difficulty calculation unit 13. If it is decided to execute an event occurrence determination, the comprehensive judgment unit 14 executes an event occurrence determination based on the moving object detection result output from the moving object detection unit 11, and controls the output of an alarm based on the result of executing the event occurrence determination.

[0088] More specifically, in the comprehensive judgment unit 14, first, when the judgment necessity judgment unit 141 judges that a moving object has been detected by the moving object detection unit 11, in other words, when judgment difficulty information is output from the judgment difficulty calculation unit 13, it decides whether to perform an event occurrence judgment to determine whether an event requiring an alarm has occurred or to postpone the execution of the event occurrence judgment, depending on the judgment difficulty calculated by the judgment difficulty calculation unit 13.

[0089] The determination necessity determination unit 141 determines whether to execute or postpone the event occurrence determination by, for example, comparing the determination difficulty level with conditions for determining whether to execute an event occurrence determination (hereinafter referred to as "difficulty level determination conditions"). The difficulty level determination conditions are set in advance by an administrator or the like, and information indicating the difficulty level determination conditions (hereinafter referred to as "difficulty level determination condition information") is stored in an internal buffer or the like of the determination necessity determination unit 141. The difficulty level determination conditions include, for example, the following conditions: "If the determination difficulty level is 'low,' execute the event occurrence determination; if the determination difficulty level is 'high,' postpone the event occurrence determination. If the determination difficulty level is expressed in stages, the determination difficulty level is set to 'high' if the determination difficulty level is equal to or greater than a set difficulty level determination threshold, and 'low' if the determination difficulty level is less than the difficulty level determination threshold." Note that the difficulty level determination threshold is set in advance by an administrator or the like together with the difficulty level determination conditions. The determination necessity determination unit 141 compares the determination difficulty with the above-described difficulty determination conditions, and if the determination difficulty is "low," determines to execute the event occurrence determination. On the other hand, if the determination difficulty is "high," the determination necessity determination unit 141 determines to suspend the event occurrence determination.

[0090] Here, a plurality of judgment difficulty levels may be output from the judgment difficulty calculation unit 13. For example, the judgment difficulty calculation unit 13 may output a judgment difficulty level calculated by the judgment difficulty calculation unit 13 based on the detection result of whether or not there is a possibility of a moving object being present outside the vehicle, which is detected by the outside-space detection unit 131, and a judgment difficulty level calculated based on the detection result of shaking in the vehicle cabin, which is detected by the shaking detection unit 132. For example, when a plurality of judgment difficulty levels are output from the judgment difficulty calculation unit 13, if the judgment necessity determination unit 141 determines that even one of the judgment difficulty levels is "high," it decides to suspend the determination of whether or not an event has occurred. Note that this is merely one example, and for example, the judgment difficulty calculation unit 13 may set a priority for determining the judgment difficulty based on which detection result from the out-of-space detection unit 131, the vibration detection unit 132, the interference detection unit 133, the nearby object detection unit 134, or the moving object size detection unit 135 to prioritize, and when it determines that the judgment difficulty is high based on a detection result with a high priority, it may not determine the judgment difficulty based on a detection result with a lower priority, and may output judgment difficulty information indicating that the judgment difficulty is high to the overall judgment unit 14. In this way, the amount of calculation for the judgment difficulty determination by the judgment difficulty calculation unit 13 is reduced, and the judgment necessity determination unit 141 can more quickly determine whether to execute or postpone the event occurrence determination.

[0091] The judgment necessity determining unit 141 outputs information indicating whether it has decided to execute or postpone the event occurrence determination (hereinafter referred to as the “judgment necessity determination result”) to the judgment executing unit 142 .

[0092] Next, in the comprehensive determination unit 14, when the determination necessity determination unit 141 determines that an event occurrence determination should be performed, the determination execution unit 142 executes an event occurrence determination based on the moving object detection result output from the moving object detection unit 11. More specifically, in the comprehensive determination unit 14, when the determination necessity determination unit 141 determines that an event occurrence determination should be performed, the determination execution unit 142 executes an event occurrence determination in accordance with an execution condition based on the moving object detection result output from the moving object detection unit 11. The execution condition is a condition that defines the timing or period for executing the event occurrence determination depending on the application, i.e., the type of alarm-relevant event to be determined, and is set in advance by an administrator or the like. When the administrator or the like sets the execution condition, information indicating the execution condition (hereinafter referred to as "execution condition information") is stored in a buffer or the like internal to the comprehensive determination unit 14. The determination execution unit 142 determines the timing or period for executing the event occurrence determination by referring to the execution condition information. For example, with regard to infant abandonment, an execution condition is set such that, once a decision is made to execute an event occurrence determination, the event occurrence determination is executed immediately. The type of moving object detection result that the determination execution unit 142 uses to execute the event occurrence determination is determined in advance by an administrator or the like. It is sufficient that the moving object detection unit 11 outputs, as a moving object detection result, at least information used by the determination execution unit 142 to execute an event occurrence determination. The determination execution unit 142 determines whether an event occurrence determination, in this case, infant abandonment, has occurred, for example, based on a moving object map. Based on the moving object map, the silhouette of the moving object, i.e., the size of the moving object, can be determined. For example, if the size of the moving object is equal to or smaller than a set threshold (hereinafter referred to as a "physique determination threshold"), the determination execution unit 142 determines that the moving object is an infant. The physique determination threshold is set in advance by an administrator or the like and stored in an internal buffer or the like of the determination execution unit 142. When the determination execution unit 142 determines based on the moving object map that only an infant is present in the vehicle compartment, it determines that an infant has been left behind.

[0093] The determination execution unit 142 outputs the determination result of whether or not an infant has been left behind (hereinafter referred to as the "event occurrence determination result") to the alarm output control unit 143. Note that, when the determination necessity determination unit 141 determines to suspend the determination of whether or not an event has occurred, the determination execution unit 142 suspends the determination of whether or not an event has occurred. In other words, the determination execution unit 142 does not execute the determination of whether or not an event has occurred. An upper limit may be set for the suspension time for which the determination execution unit 142 suspends the determination of whether or not an event has occurred. In this case, for example, when the determination necessity determination unit 141 determines to suspend the execution of the event occurrence determination and the duration of the suspension after the execution of the event occurrence determination has been suspended exceeds the set maximum suspension time, the determination execution unit 142 executes the determination of whether or not an event has occurred. Note that this is merely an example. For example, if the determination necessity determination unit 141 determines to suspend the execution of the event occurrence determination, and the duration of the suspension after the suspension has exceeded a set maximum suspension time, the determination execution unit 142 may output information indicating that an alarm must be output immediately (hereinafter referred to as "alarm output requirement information") to the alarm output control unit 143. In this case, the alarm output control unit 143 immediately outputs an alarm from the alarm device 3. Details of the alarm output control unit 143 will be described later. The maximum suspension time is set in advance, for example, by an administrator or the like. For example, when the determination execution unit 142 starts to suspend the execution of the event occurrence determination, the determination execution unit 142 starts counting the duration of the suspension using a timer stored in an internal buffer or the like. If the counted duration exceeds the maximum suspension time, the determination execution unit 142 executes the event occurrence determination.

[0094] For example, if the determination necessity determination unit 141 determines to suspend the execution of the determination of whether or not an event has occurred, and the duration of the suspension after the execution of the determination of whether or not an event has occurred exceeds a set maximum suspension time, it is possible that the calculation of the determination difficulty level is incorrect, or that even if there is no error in the calculation of the determination difficulty level, the duration of the suspension has been high for a long time, making it impossible to properly determine the occurrence of an event that requires an alarm. By causing the determination execution unit 142 to immediately output alarm output requirement information to the alarm output control unit 143 after the determination necessity determination unit 141 determines to suspend the execution of the determination of whether or not an event has occurred, and the duration of the suspension after the execution of the determination of whether or not an event has occurred exceeds the set maximum suspension time, the alarm control device 1 can reduce the risk of falling into a situation where an alarm is not output even though an event that requires an alarm has occurred. For example, taking the example of vibration detection by the vibration detection unit 132, since it is unlikely that vibration will continue for an extremely long period of time, by setting such a limit, the alarm control device 1 can reduce the risk of falling into a situation where an alarm is not output even though a child has actually been left behind, even if vibration is mistakenly detected when there is no vibration and the judgment difficulty level is incorrectly determined. Alternatively, if the legal regulations or guidelines of each country specify the time period for outputting an alarm, such as requiring an alarm to be issued within 10 seconds after the doors are locked, by setting an upper limit on the hold time accordingly, the alarm control device 1 can comply with the regulations while improving the accuracy of the alarm output.

[0095] The warning output control unit 143 of the comprehensive determination unit 14 performs warning output control when it is determined that an event requiring a warning has occurred as a result of the event occurrence determination performed by the determination execution unit 142. The warning output control unit 143 can determine whether or not it has been determined that an event requiring a warning has occurred based on the event occurrence determination result.

[0096] Specifically, when it is determined that an event requiring an alarm has occurred, the alarm output control unit 143 outputs information (hereinafter referred to as "alarm output control information") to the alarm device 3 to cause it to output an alarm. For example, if the alarm device 3 is an audio output device, the alarm output control unit 143 outputs alarm output control information to cause the alarm device 3 to output an alarm sound or an audio warning such as "Someone has been left behind." Also, for example, if the alarm device 3 is a display device, the alarm output control unit 143 outputs alarm output control information to cause the alarm device 3 to display a warning message such as "Someone has been left behind." Also, for example, if the alarm device 3 is a flashing device, the alarm output control unit 143 outputs alarm output control information to cause the alarm device 3 to perform a predetermined flashing. Note that these are merely examples, and it is sufficient for the alarm output control unit 143 to output alarm output control information to cause the alarm device 3 to output some kind of information notifying that someone has been left behind.

[0097] For example, the alarm output control unit 143 may determine an alarm output mode according to the determination difficulty level, and output alarm output control information to the alarm device 3 to cause the alarm to be output in the determined mode. The alarm output mode is, for example, the intensity of the alarm. In the first embodiment, the intensity of the alarm is indicated, for example, by the duration of the alarm output, the volume of the alarm if it is output as audio, the content of the display if it is displayed, or the alarm output method. The alarm output control unit 143 may adjust at least one of these to set the alarm output mode according to the determination difficulty level. For example, the determination difficulty level is calculated on a five-level scale from "1" to "5," and the difficulty level determination condition includes the above-described condition: "If the determination difficulty level is "low," execute the event occurrence determination; if the determination difficulty level is "high," suspend the event occurrence determination." When the determination difficulty level is indicated by a level, the determination difficulty level is set to "high" if the determination difficulty level is equal to or greater than a set difficulty level determination threshold, and set to "low" if it is less than the difficulty level determination threshold." Let us assume that the difficulty level determination threshold is "4." In this case, if the difficulty level calculated by the judgment difficulty calculation unit 13 is between "1" and "3," the judgment necessity determination unit 141 determines that the judgment difficulty level is "low" and determines to execute an event occurrence determination. Then, the determination execution unit 142 executes the event occurrence determination. For example, suppose that the execution of the event occurrence determination determines that an infant has been abandoned. In such a case, the alarm output control unit 143 determines, for example, to output an alarm with a lower intensity as the judgment difficulty level increases. In the above-described example, it may be determined that an infant has been abandoned when the judgment difficulty level is "1," "2," or "3." For example, the alarm output control unit 143 determines to output an alarm with a lower intensity when the judgment difficulty level is "3" than when the judgment difficulty level is "2," and to output an alarm with a lower intensity when the judgment difficulty level is "2" than when the judgment difficulty level is "1." For example, the warning output control unit 143 may determine to output a warning with a lower intensity when the judgment difficulty level is "3" than when the judgment difficulty level is "1" or "2."

[0098] 6 is a diagram illustrating an example of the mode of the alarm to be output, determined by the alarm output control unit 143 according to the determination difficulty level in embodiment 1. In FIG. 6, as an example, the alarm output control unit 143 determines the mode of the alarm to be output, i.e., the intensity of the alarm, in two stages. For example, in the example described above, the alarm output control unit 143 determines the mode of the alarm to be output so that the intensity of the alarm is lower for the determination difficulty level "3" than for the determination difficulty levels "1" and "2."

[0099] In this way, when the determination necessity determination unit 141 determines to execute an event occurrence determination and the determination execution unit 142 determines that an event requiring an alarm has occurred in the event occurrence determination, the alarm output control unit 143 may determine the type of alarm to be output according to the determination difficulty calculated by the determination difficulty calculation unit 13. More specifically, for example, the alarm output control unit 143 may determine to output an alarm with a lower intensity as the determination difficulty increases. This allows the alarm control device 1 to notify people in the vicinity of the location where the alarm was output, such as the driver, people in the vicinity of the vehicle, or people holding a mobile device, that the alarm was output because it was determined that an infant had been abandoned in a situation where it was difficult to determine whether an infant had been abandoned, thereby reducing the annoyance caused by a false alarm. Furthermore, for example, when it is determined that an infant had been abandoned in a situation where it was not difficult to determine whether an infant had been abandoned, the alarm control device 1 can output an alarm to more reliably notify people in the vicinity of the location where the alarm was output that an infant had been abandoned, thereby more emphasizing the high need for action. When the warning output necessary information is output from the determination execution unit 142, the warning output control unit 143 immediately causes the warning device 3 to output a warning.

[0100] The following describes the operation of the alarm control device 1 according to embodiment 1. Fig. 7 is a flowchart for explaining the operation of the alarm control device 1 according to embodiment 1.

[0101] For example, the alarm control device 1 starts the operation shown in the flowchart of Fig. 7 when the state inside the vehicle cabin satisfies the space determination condition. The space determination condition defines an appropriate target space state as the timing for the alarm control device 1 to start the operation, and the space determination condition is set in advance by an administrator or the like. Here, the space determination condition is defined as the change of the vehicle cabin from an open state to a closed state. A control unit (not shown) of the alarm control device 1 determines, for example, the timing when the vehicle door is closed or the vehicle key is locked as the timing when the vehicle cabin changes from an open state to a closed state, i.e., the timing to start the operation shown in the flowchart of Fig. 7.

[0102] The determination difficulty level calculated by the determination difficulty calculation unit 13 is preferably calculated starting from the timing when the vehicle interior changes from an open state to a closed state. For example, when the vehicle door is open, there is a possibility that a person is present inside or outside the vehicle, such as while getting in or out of the vehicle, or that a person's movement around the sensor 2 may block the transmission and reception of radio waves by the sensor 2. In such a case, the state inside the vehicle interior detected by the sensor 2 may change frequently, making it difficult to obtain stable detection results for moving objects. The alarm control device 1 can calculate the determination difficulty level after the state inside the vehicle interior changes from an open state to a closed state, such as when the door is closed or the key is locked, and determine whether an event has occurred after the state inside the vehicle interior has stabilized. This enables more accurate determination of the determination difficulty level, and as a result, more accurate determination of whether an event has occurred. Therefore, for example, in the operation of the alarm control device 1 shown in the flowchart of FIG. 7 , it is preferable that at least the determination difficulty level calculation process by the determination difficulty calculation unit 13 is performed when the vehicle door is closed or the key is locked.

[0103] In addition, in the alarm control device 1, if the judgment difficulty calculation unit 13 calculates the judgment difficulty using multiple methods from multiple detection results and it takes time to calculate any of the judgment difficulty levels, the control unit will prevent the overall judgment unit 14 from determining whether or not an event has occurred until the judgment difficulty calculation unit 13 has completed calculating all of the judgment difficulty levels.

[0104] The alarm control device 1 repeats the operation shown in the flowchart of FIG. 7 at predetermined intervals (for example, one second).

[0105] The moving object detection unit 11 extracts a moving object based on the received signal obtained by the sensor 2 (step ST1). The moving object detection unit 11 outputs the detection result of the moving object to the difficulty level calculation information acquisition unit 12 and the comprehensive determination unit 14.

[0106] The difficulty level calculation information acquisition unit 12 acquires a moving object detection result from the moving object detection unit 11, and if it determines that the moving object detection unit 11 has detected the presence of a moving object based on the moving object detection result acquired from the moving object detection unit 11, it acquires difficulty level calculation information (step ST2). The difficulty level calculation information acquisition unit 12 outputs the acquired difficulty level calculation information to the determination difficulty level calculation unit 13. Note that if the difficulty level calculation information acquisition unit 12 determines that the moving object detection unit 11 has not detected the presence of a moving object, for example, the operation of the alarm control device 1 skips the processes of steps ST3 and ST4 and returns to the process of step ST1.

[0107] The judgment difficulty calculation unit 13 performs a judgment difficulty calculation process (step ST3) to calculate the judgment difficulty based on the difficulty calculation information acquired by the difficulty calculation information acquisition unit 12 in step ST2. The judgment difficulty calculation unit 13 outputs the calculated judgment difficulty to the overall judgment unit 14.

[0108] The comprehensive judgment unit 14 determines whether to perform an event occurrence determination to determine whether an event requiring an alert has occurred or to postpone the execution of the event occurrence determination, depending on the judgment difficulty calculated by the judgment difficulty calculation unit 13 in step ST3.If it decides to perform the event occurrence determination, it performs an event occurrence determination based on the moving object detection result output from the moving object detection unit 11, and performs a comprehensive judgment process to control the alert output based on the result of the event occurrence determination (step ST4).

[0109] Fig. 8 is a flowchart illustrating the details of the processing performed by the comprehensive judgment unit 14 in the comprehensive judgment processing of step ST4 in Fig. 7. It is assumed here that the difficulty level judgment conditions include the following conditions: "If the judgment difficulty level is 'low', execute a judgment on whether an event has occurred, and if the judgment difficulty level is 'high', postpone the judgment on whether an event has occurred. If the judgment difficulty level is indicated in stages, the judgment difficulty level is 'high' if the judgment difficulty level is equal to or greater than the set difficulty level judgment threshold, and 'low' if the judgment difficulty level is less than the difficulty level judgment threshold."

[0110] The judgment necessity determination unit 141 determines whether the judgment difficulty is determined to be "high" by comparing the judgment difficulty with the difficulty determination conditions (step ST101). If the judgment difficulty is determined to be "high" ("YES" in step ST101), the judgment necessity determination unit 141 decides to suspend the determination of whether an event has occurred (step ST102). The judgment necessity determination unit 141 outputs the judgment result of whether or not a judgment is necessary to the judgment execution unit 142. The judgment execution unit 142 suspends the determination of whether or not an event has occurred. Then, the operation of the alarm control device 1 returns to the processing of step ST1 in FIG. 7.

[0111] On the other hand, if the judgment difficulty level is judged to be "low" ("NO" in step ST101), the judgment necessity judgment unit 141 decides to execute the event occurrence judgment (step ST103). The judgment necessity judgment unit 141 outputs the judgment result of whether or not the judgment is necessary to the judgment execution unit 142.

[0112] The determination execution unit 142 determines whether an event has occurred, in this case, whether an infant has been left behind, based on the detection result of the moving object output from the moving object detection unit 11 in step ST1 of Fig. 7 (step ST104). The determination execution unit 142 outputs the result of the determination of whether an event has occurred to the alarm output control unit 143.

[0113] When it is determined that an event requiring an alarm has occurred as a result of the event occurrence determination performed by the determination execution unit 142 in step ST104 (in the case of "YES" in step ST105), the alarm output control unit 143 performs alarm output control. At this time, for example, the alarm output control unit 143 may determine the alarm output mode to be a mode corresponding to the determination difficulty level, and output alarm output control information to the alarm device 3 to cause the alarm to be output in the determined mode.

[0114] On the other hand, if the judgment execution unit 142 executes an event occurrence judgment in step ST104 and it is judged that an event requiring an alarm has not occurred (if the answer is "NO" in step ST105), the operation of the alarm control device 1 returns to the processing of step ST1 in Figure 7.

[0115] If an upper limit is set for the hold time, in step ST102, the determination execution unit 142 determines whether the duration of the hold has exceeded the set maximum hold time. If it is determined that the maximum hold time has passed, the determination execution unit 142 executes a determination as to whether an event has occurred. The comprehensive determination unit 14 executes the processes of steps ST104 to ST106.

[0116] As explained using the flowcharts in Figures 7 and 8, if the calculated determination difficulty is high, the warning control device 1 does not execute the event occurrence determination. As a result, no warning is output. As described above, the warning control device 1 repeats the operations shown in the flowcharts in Figures 7 and 8 at a predetermined cycle. Therefore, over time, for example, if the situation in the vehicle cabin or around the vehicle that was considered to be high in determination difficulty eases and the determination difficulty becomes low, the warning control device 1 executes the event occurrence determination, and if it is determined that an event requiring a warning has occurred, a warning will be output.

[0117] In this way, the alarm control device 1 calculates a determination difficulty level, which is the degree of difficulty of determining whether an event requiring an alarm has occurred, based on the difficulty level calculation information, and determines whether to execute an event occurrence determination or postpone execution of the event occurrence determination based on the determination difficulty level. When it is decided to execute an event occurrence determination, the alarm control device 1 executes an event occurrence determination based on the moving object detection result, and performs alarm output control based on the result of the event occurrence determination. As a result, the alarm control device 1 can perform more accurate alarm output control than conventional technology in alarm output control that outputs an alarm when it is determined that an event requiring an alarm has occurred in the target space based on the moving object detection result obtained from a reflected signal of a signal transmitted by the sensor 2 toward the target space and reflected by an object in the target space.

[0118] In the first embodiment described above, the moving object detection unit 11 is provided in the alarm control device 1, but this is merely an example. For example, the moving object detection unit 11 may be provided in the sensor 2. In this case, the alarm control device 1 may be configured not to include the moving object detection unit 11, and the processing of step ST1 may be omitted in the operation shown in the flowchart of FIG. 7.

[0119] In the first embodiment described above, the warning control device 1 is an on-board device mounted on a vehicle, but this is merely an example. For example, the moving object detection unit 11, the difficulty level calculation information acquisition unit 12, the judgment difficulty calculation unit 13, or the overall judgment unit 14 may each be implemented in a separate device. Alternatively, for example, the moving object detection unit 11, the difficulty level calculation information acquisition unit 12, the judgment difficulty calculation unit 13, or part of the overall judgment unit 14 may be provided in a server (not shown) connected to the on-board device via a communication line, and the on-board device and the server may form a system. Alternatively, for example, the moving object detection unit 11, the difficulty level calculation information acquisition unit 12, the judgment difficulty calculation unit 13, and the overall judgment unit 14 may all be provided in the server.

[0120] In the first embodiment, the target space is the interior of a vehicle, but this is merely an example. For example, the target space may be the interior of a bus, airplane, ship, or train, or a living room in a house, office, or factory, or a toilet, bathroom, or elevator car, or may be a room other than a vehicle. The target space can be any space in which a moving object, including a living body, can exist.

[0121] In the first embodiment, the alarm-requiring event is the abandonment of a child in the vehicle cabin. However, this is merely an example. For example, the alarm-requiring event may be any of a variety of events that require an alarm to be issued when they occur, such as the intrusion of a suspicious person or the like into the target space, or the occurrence of a person who has collapsed due to illness or the like. As described above, the execution conditions that the determination execution unit 142 in the comprehensive determination unit 14 follows when executing the event occurrence determination are set depending on the type of alarm-requiring event to be determined. For example, if the target space is the vehicle cabin and the alarm-requiring event is the intrusion of a suspicious person or the like into the vehicle cabin, the execution conditions include the execution of the event occurrence determination for a period of time after the vehicle cabin key is locked, during which the determination difficulty is determined to be low and the key lock is not normally released. The determination execution unit 142 determines whether a suspicious person or the like has intruded into the vehicle cabin during this period. Furthermore, for example, if the target space is a bathroom and the event to be alerted is the occurrence of a person collapsing in the bathroom due to poor health or the like, a condition is set such that the event occurrence determination is to be performed for a period determined to have a low determination difficulty, such as one hour, during which a person is normally considered to be in the bathroom. The determination execution unit 142 determines whether a person has collapsed in the bathroom during this period. For example, the determination execution unit 142 may determine that a person has collapsed if a person has continuously been present in the bathroom for a set period or longer during this period.

[0122] As described above, the target space can be any space in which a moving object, including a living body, may be present, and the alarm-requiring event can be any event requiring an alarm to be issued upon its occurrence. Therefore, various timings are possible for the alarm control device 1 to consider as the optimal timing for calculating the determination difficulty, depending on the target space or the content of the alarm-requiring event. For example, in the first embodiment, the optimal timing for calculating the determination difficulty is the timing when the vehicle interior changes from an open state to a closed state. However, for example, in a monitoring system in which the target space is an open space and the alarm-requiring event is the presence or absence of a suspicious person, even an open state of the target space can be a optimal timing for calculating the determination difficulty. The space determination conditions for determining the timing for the alarm control device 1 to start operation may be set to target space state conditions that can determine the optimal timing for calculating the determination difficulty, depending on the content of the target space or the alarm-requiring event.

[0123] 9A and 9B are diagrams illustrating an example of the hardware configuration of the alarm control device 1 according to the first embodiment. In the first embodiment, the functions of the moving object detection unit 11, the difficulty level calculation information acquisition unit 12, the determination difficulty level calculation unit 13, the overall determination unit 14, and a control unit (not shown) are realized by a processing circuit 101. That is, the alarm control device 1 includes the processing circuit 101 for performing alarm control according to the determination difficulty level in alarm control that outputs an alarm when it is determined that an alarm-requiring event requiring alarm output has occurred in the target space based on the moving object detection result obtained from the reflected signal of a signal transmitted by the sensor 2 toward the target space and reflected by an object in the target space. The processing circuit 101 may be dedicated hardware as shown in FIG. 9A or a processor 104 that executes a program stored in memory as shown in FIG. 9B.

[0124] When processing circuitry 101 is dedicated hardware, processing circuitry 101 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof.

[0125] When the processing circuit is a processor 104, the functions of the moving object detection unit 11, the difficulty level calculation information acquisition unit 12, the judgment difficulty calculation unit 13, the overall judgment unit 14, and a control unit (not shown) are realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in memory 105. The processor 104 reads and executes the program stored in memory 105 to perform the functions of the moving object detection unit 11, the difficulty level calculation information acquisition unit 12, the judgment difficulty calculation unit 13, the overall judgment unit 14, and a control unit (not shown). In other words, the alarm control device 1 includes memory 105 for storing a program that, when executed by the processor 104, results in the processing of steps ST1 to ST4 of FIG. 7 described above. The program stored in memory 105 can also be said to cause a computer to execute a procedure or method for the processing of the moving object detection unit 11, the difficulty level calculation information acquisition unit 12, the judgment difficulty calculation unit 13, the overall judgment unit 14, and a control unit (not shown). Here, the memory 105 may be, for example, a non-volatile or volatile semiconductor memory such as a RAM, a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), or an EEPROM (Electrically Erasable Programmable Read-Only Memory), a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a DVD (Digital Versatile Disc).

[0126] Note that the functions of the moving object detection unit 11, the difficulty level calculation information acquisition unit 12, the judgment difficulty calculation unit 13, the overall judgment unit 14, and a control unit (not shown) may be partially implemented by dedicated hardware and partially implemented by software or firmware. For example, the functions of the moving object detection unit 11 and the difficulty level calculation information acquisition unit 12 may be implemented by a processing circuit 101 as dedicated hardware, and the functions of the judgment difficulty calculation unit 13, the overall judgment unit 14, and a control unit (not shown) may be implemented by a processor 104 reading and executing programs stored in a memory 105. The alarm control device 1 also includes an input interface device 102 and an output interface device 103 that communicate with devices such as the sensor 2 or the alarm device 3 via wired or wireless communication.

[0127] As described above, according to the first embodiment, the alarm control device 1 outputs an alarm when it determines that an alarm-required event requiring an alarm has occurred in the target space based on a moving object detection result obtained from a reflected signal obtained when a signal transmitted by the sensor 2 toward the target space is reflected by an object in the target space. The alarm control device 1 is configured to include a determination difficulty calculation unit 13 that calculates the determination difficulty based on difficulty calculation information for calculating the determination difficulty, which is the degree of difficulty of determining whether an alarm-required event has occurred, and a comprehensive determination unit 14 that determines whether to perform an event occurrence determination to determine whether an alarm-required event has occurred or to postpone the execution of the event occurrence determination according to the determination difficulty calculated by the determination difficulty calculation unit 13, and if it is decided to perform the event occurrence determination, performs the event occurrence determination based on the moving object detection result, and controls the output of the alarm based on the result of the event occurrence determination. Therefore, the alarm control device 1 can perform more accurate alarm output control than conventional technology in outputting an alarm when it determines that an alarm-required event has occurred in the target space based on a moving object detection result obtained from a reflected signal obtained when a signal transmitted by the sensor 2 toward the target space is reflected by an object in the target space.

[0128] Any of the components of the embodiments may be modified or omitted.

[0129] The alarm control device of the present disclosure can perform more accurate alarm output control in outputting an alarm when it is determined that an alarm-requiring event has occurred within a target space based on the detection result of a moving object obtained from a reflected signal obtained when a signal transmitted by a sensor toward the target space is reflected by an object within the target space.

[0130] 1 Alarm control device, 11 Moving object detection unit, 12 Difficulty level calculation information acquisition unit, 13 Judgment difficulty calculation unit, 131 Out-of-space detection unit, 132 Vibration detection unit, 133 Interference detection unit, 134 Proximate object detection unit, 135 Moving object size detection unit, 14 Overall judgment unit, 141 Judgment necessity judgment unit, 142 Judgment execution unit, 143 Alarm output control unit, 2 Sensor, 3 Alarm device, 101 Processing circuit, 102 Input interface device, 103 Output interface device, 104 Processor, 105 Memory.

Claims

An alarm control device that outputs an alarm when it is determined that an alarm event requiring an alarm has occurred in a target space based on a detection result of a moving object obtained from a reflected signal of a signal transmitted by a sensor toward the target space and reflected by an object in the target space, a determination difficulty calculation unit that calculates a determination difficulty level based on determination difficulty level calculation information for calculating a degree of difficulty in determining whether or not the alarm-required event has occurred; a comprehensive determination unit that determines whether to execute an event occurrence determination as to whether the alarm target event has occurred or to postpone execution of the event occurrence determination, in accordance with the determination difficulty calculated by the determination difficulty calculation unit, and when it is determined to execute the event occurrence determination, executes the event occurrence determination based on the detection result of the moving object, and controls output of the alarm based on the execution result of the event occurrence determination. An alarm control device comprising: 　 The comprehensive judgment unit When the judgment difficulty level calculated by the judgment difficulty level calculation unit is determined to be high, it is decided to suspend the judgment of whether or not an event has occurred, and when the judgment difficulty level calculated by the judgment difficulty level calculation unit is determined to be low, it is decided to execute the judgment of whether or not an event has occurred.

2. The alarm control device according to claim 1. 　 the difficulty level calculation information includes a distance profile that is information indicating a relationship between a distance from the sensor and a reflection intensity of the reflected signal corresponding to the distance, The determination difficulty calculation unit an outside-space detection unit that detects whether the moving object exists outside the target space based on the distance profile, and calculates the determination difficulty level based on a detection result of whether the moving object exists outside the target space by the outside-space detection unit; 2. The alarm control device according to claim 1. 　 The outside-space detection unit Whether or not the moving object is likely to exist outside the target space is detected based on whether or not a peak of the reflection intensity appears in the distance profile within a distance range corresponding to the periphery of the target space that can be considered to be outside the target space.

4. The alarm control device according to claim 3. 　 The outside-space detection unit Detecting whether or not the moving object is likely to exist outside the target space based on the gradient of attenuation of the reflection intensity in the distance direction in the distance profile.

4. The alarm control device according to claim 3. 　 the difficulty level calculation information includes the distance profile and moving object information indicating a distance from the sensor to the moving object obtained by signal processing the reflected signal and an arrival angle of the reflected signal at that distance, The outside-space detection unit Detecting whether the moving object exists outside the target space based on the distance corresponding to the peak of the reflection intensity in the distance profile and the angle of arrival of the reflection signal at that distance.

4. The alarm control device according to claim 3. 　 the difficulty level calculation information includes vibration information regarding vibrations occurring in the target space, The determination difficulty calculation unit a vibration detection unit that detects vibrations in the target space based on the vibration information, and calculates the determination difficulty level based on the vibration detection result by the vibration detection unit; 2. The alarm control device according to claim 1. 　 The sway information is Acceleration information detected by an acceleration sensor installed in the target space, or information indicating frequency components of vibrations in the target space obtained by signal processing of the reflected signal, is included.

8. The alarm control device according to claim 7. 　 the difficulty level calculation information includes information indicating a distance to the moving object obtained from the reflected signal or information indicating a reflection intensity of the reflected signal, The determination difficulty calculation unit an interference detection unit that detects whether or not there is interference with the detection of the moving object by the sensor based on information indicating the distance to the moving object or the reflection intensity of the reflected signal, and calculates the determination difficulty level based on the detection result of the presence or absence of interference by the interference detection unit; 2. The alarm control device according to claim 1. 　 The interference detection unit The presence or absence of the interference is detected based on the change in the reflection intensity over time or the reflection intensity corresponding to a range of distances equal to or greater than a set distance.

10. The alarm control device according to claim 9. 　 the difficulty level calculation information includes a distance profile that is information indicating a relationship between a distance from the sensor and a reflection intensity of the reflected signal corresponding to the distance, The determination difficulty calculation unit a proximity object detection unit that detects the presence or absence of an object blocking the sensor based on the distance profile, and calculates the determination difficulty level based on a detection result of the presence or absence of the blocking object by the proximity object detection unit; 2. The alarm control device according to claim 1. 　 The proximity object detection unit The presence or absence of the obstruction is detected based on the reflection intensity detected within a set distance range.

12. The alarm control device according to claim 11. 　 the difficulty level calculation information includes a moving object map that represents the moving objects present in the target space as a three-dimensional spatial distribution, the moving object map being generated by signal processing of the reflected signals of the signals transmitted by the sensors toward the target space and reflected by the objects in the target space; The determination difficulty calculation unit a moving object size detection unit that detects a change in size of the moving object determined based on the time-series moving object map, and calculates the determination difficulty level based on the change detected by the moving object size detection unit; 2. The alarm control device according to claim 1. 　 The determination difficulty calculation unit starts calculating the determination difficulty at a timing when the state of the target space becomes a state that satisfies a space determination condition.

2. The alarm control device according to claim 1. 　 The target space is a place separated from its surroundings, The determination difficulty calculation unit determines a timing when the target space changes from an open state to a closed state as a timing when the target space satisfies the space determination condition.

15. The alarm control device according to claim 14. 　 The target space is a room, The determination difficulty calculation unit determines that the target space satisfies the space determination condition when a door to the room is closed or when a key to the room is locked.

16. The alarm control device according to claim 15. 　 The comprehensive judgment unit After deciding to suspend the execution of the event occurrence determination, if the duration of the suspension has exceeded a set maximum suspension time, it is decided to execute the event occurrence determination.

2. The alarm control device according to claim 1. 　 The comprehensive judgment unit When it is determined that the event occurrence determination is to be performed and when it is determined that the alarm target event has occurred in the event occurrence determination, the determination difficulty calculation unit determines the mode of the alarm to be output in accordance with the determination difficulty calculated by the determination difficulty calculation unit.

2. The alarm control device according to claim 1. 　 the aspect of the warning is the intensity of the warning, The comprehensive judgment unit The determination unit determines that the warning is to be output with a lower intensity as the determination difficulty calculated by the determination difficulty calculation unit increases.

19. The alarm control device according to claim 18. 　 The intensity is indicated by the duration for which the alarm is output, the volume of the alarm when the alarm is output by voice, the content of the display when the alarm is output by display, or the output method of the alarm.

20. The alarm control device according to claim 19. 　 An alarm control method for outputting an alarm when it is determined that an alarm event requiring output of an alarm has occurred in a target space based on a detection result of a moving object obtained from a reflected signal of a signal transmitted by a sensor toward the target space and reflected by an object in the target space, a determination difficulty level calculation unit calculating the determination difficulty level based on difficulty level calculation information for calculating the determination difficulty level, which is a degree of difficulty in determining whether or not the alarm-required event has occurred; a step in which a comprehensive determination unit determines whether to execute an event occurrence determination as to whether the alarm target event has occurred or to suspend execution of the event occurrence determination, in accordance with the determination difficulty level calculated by the determination difficulty calculation unit, and when it is determined to execute the event occurrence determination, executes the event occurrence determination based on a detection result of the moving object, and controls output of the alarm based on a result of executing the event occurrence determination. An alarm control method comprising:

Images