Congestion degree estimation system and congestion degree estimation method
The congestion degree estimation system uses existing door system sensors to accurately estimate congestion levels, addressing low accuracy and cost issues in conventional methods, enhancing passenger boarding efficiency.
Patent Information
- Application Number
- JP2021019093
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-02-09
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2041-02-09
Smart Images

Figure 0007585841000001 
Figure 0007585841000002 
Figure 0007585841000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a congestion level estimation system and a congestion level estimation method. [Background technology]
[0002] There is a known technology for estimating the degree of congestion at each door of each car in passenger trains. For example, Patent Document 1 describes a technology for determining the degree of congestion at each door by calculating the number of passengers at each door based on the weight of each door acquired by a vehicle weight sensor. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] JP 2019-89417 A Summary of the Invention [Problem to be solved by the invention]
[0004] However, the conventional technology has problems such as low congestion degree estimation accuracy and the need to introduce new equipment, which incurs huge additional costs. For example, in the above-mentioned Patent Document 1, the number of passengers per door is calculated by dividing the weight of each door by the average weight of Japanese people, and this does not take into account differences in physique of individuals, the presence or absence of luggage, or the posture of the vehicle itself (for example, the inclination of the vehicle), which may result in low congestion degree estimation accuracy. In addition, for example, it is necessary to place a weight sensor at each door, which incurs huge additional costs.
[0005] An embodiment of the present invention has been made in consideration of the above points, and has an object to estimate a congestion degree with high accuracy at low cost. [Means for solving the problem]
[0006] In order to achieve the above-mentioned object, a congestion level estimation system according to one embodiment is a congestion level estimation system that estimates the congestion level at each door of a railway vehicle, and includes an estimation unit that estimates the congestion level at the door using a predetermined congestion level estimation model based on sensor information indicating the sensing results of a sensor used to control the opening and closing of the door, and a notification unit that notifies a notification target of the congestion level at the door. Effect of the Invention
[0007] The degree of congestion can be estimated accurately at low cost. [Brief description of the drawings]
[0008] [Figure 1] 1 is a diagram illustrating an example of an overall configuration of a congestion degree estimation system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a diagram illustrating an example of a hardware configuration of a control device according to the present embodiment. [Diagram 3] FIG. 2 is a diagram illustrating an example of a functional configuration of a control device according to the present embodiment. [Figure 4] 4 is a flowchart illustrating an example of an overall process of the congestion degree estimation system according to the present embodiment. [Diagram 5] FIG. 11 is a diagram illustrating an example of a congestion degree table. [Figure 6] FIG. 11 is a diagram illustrating weight parameters in one embodiment. [Figure 7] FIG. 11 is a diagram for explaining congestion degree estimation in an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0009] An embodiment of the present invention will be described below. In this embodiment, a congestion level estimation system 1 capable of estimating a congestion level with high accuracy at low cost will be described.
[0010] <Overall configuration of crowding estimation system 1> 1, the congestion degree estimation system 1 according to this embodiment includes a plurality of door systems 10 and a notification system 20. Each door system 10 and the notification system 20 are connected to each other so as to be able to communicate with each other via wires, wirelessly, or both.
[0011] The door system 10 is installed at each door of a railway car (hereinafter, simply referred to as a car) and is a device and a group of devices that control the opening and closing of the door. For example, if one car has eight doors, there will be eight door systems 10 in one car.
[0012] In addition, before the train arrives at the next station, the door system 10 estimates the degree of congestion at the door that it controls to open and close (i.e., an index showing the degree of congestion around that door) and transmits the degree of congestion to the notification system 20.
[0013] The notification system 20 notifies a notification target of the congestion degree received from each door system 10 (i.e., the congestion degree for each door). Examples of the notification target include a terminal such as a smartphone of a user who plans to board a train at the next station, and an on-site terminal such as an electronic bulletin board installed in the premises of the next station. However, this is only one example, and the notification target may be, for example, a web server that publishes a railroad company's web page, a terminal such as a smartphone used by a station staff member at the next station, or another vehicle control system.
[0014] Here, the door system 10 includes a control device 100, a sensor group 200, and a door closing device 300. The control device 100 and the sensor group 200 are connected to be able to communicate with each other by wire or wirelessly or both, and the control device 100 and the door closing device 300 are also connected to be able to communicate with each other by wire or wirelessly or both. Note that the sensor group 200 and the door closing device 300 may also be connected to be able to communicate with each other by wire or wirelessly or both.
[0015] The control device 100 transmits an opening / closing control command for controlling the opening / closing of the door to the door closing device 300 in response to a door opening command or closing command or sensor information. The control device 100 also estimates the degree of congestion at the door from an event that occurs when the door is opened or closed (particularly an irregular event that is likely to occur during congestion). The opening and closing commands are transmitted from a higher-level system in response to opening and closing operations by train crew members, station staff, etc.
[0016] The sensor group 200 includes various sensors that sense the state of the door and events that occur at the door, and transmits sensor information including the sensing results to the control device 100. The sensor group 200 includes an opening / closing sensor 201, a door jam detection sensor 202, an external force measurement sensor 203, and an opening / closing speed detection sensor 204. The opening / closing sensor 201 detects whether the door is opening or closing. The door jam detection sensor 202 detects the occurrence of door jamming (i.e., a passenger's body, luggage, etc. being caught when the door is closing). The external force measurement sensor 203 measures an external force on the door (for example, an external force generated by a passenger leaning against the door, etc.). The opening / closing speed detection sensor 204 detects the speed (speed) of the door opening and closing operations.
[0017] The door opening / closing device 300 includes a motor (drive device) for opening and closing the door, and opens and closes the door by passing a current through the motor in response to an opening / closing control command. The opening / closing control command includes information indicating whether an opening operation or a closing operation is to be performed, and information indicating the magnitude of the current to be passed through the motor. Generally, the larger the current passed through the motor, the larger the torque of the motor becomes, and the faster the door opens and closes.
[0018] The motor includes not only an alternating-current (AC) motor, but also various other motors such as a direct-current (DC) motor and a linear motor.
[0019] <Hardware configuration of the control device 100> 2, the control device 100 according to this embodiment includes, as hardware, a processor 101, a memory device 102, and an I / F 103. These pieces of hardware are connected via a bus 104 so as to be able to communicate with each other.
[0020] The processor 101 is, for example, a computing device such as an MCU (Micro Controller Unit) (particularly, a computing device for embedded devices). The memory device 102 is, for example, a storage device such as a semiconductor memory. The I / F 103 is an interface with the sensor group 200, the door closing device 300, and the like.
[0021] The control device 100 according to this embodiment can realize various processes described below by having the hardware configuration shown in Fig. 2. However, the hardware configuration shown in Fig. 2 is only an example, and the control device 100 may have a hardware configuration other than this.
[0022] <Functional configuration of the control device 100> 3, the control device 100 according to this embodiment has, as functional units, a switching control unit 110, a congestion level estimation unit 120, a congestion level transmission unit 130, and a storage unit 140. The switching control unit 110, the congestion level estimation unit 120, and the congestion level transmission unit 130 are realized, for example, by processing executed by the processor 101 of one or more programs stored in the memory device 102. The storage unit 140 is realized by the memory device 102.
[0023] When the opening / closing control unit 110 receives a door open command or a door close command or sensor information, the opening / closing control unit 110 transmits an opening / closing control command based on a known door control method to the door closing device 300 in response to the opening command or the close command or the sensor information. For example, when the opening / closing control unit 110 receives an opening command, it transmits an opening / closing control command for opening the door, and when the opening / closing control unit 110 receives a close command, it transmits an opening / closing control command for closing the door. Also, for example, when the opening / closing control unit 110 receives sensor information indicating that a door is caught during a door closing operation, it transmits an opening / closing control command for opening the door. Similarly, for example, when the opening / closing control unit 110 receives sensor information indicating an external force equal to or greater than a predetermined threshold, it transmits an opening / closing control command for increasing the current flowing through the motor. In this way, the opening / closing control unit 110 controls the opening and closing of the door by a known door control method.
[0024] Furthermore, the opening / closing control unit 110 stores the sensor information received from the sensor group 200 and the opening / closing control command transmitted to the door closing device 300 in the memory unit 140 as a log.
[0025] The congestion degree estimation unit 120 estimates the congestion degree at a corresponding door (i.e., a door whose opening and closing is controlled by the door system 10) at each station according to a congestion degree estimation model using the sensor information and opening and closing control commands stored in the memory unit 140. In addition, the congestion degree estimation unit 120 stores the congestion degree in the memory unit 140 as a log.
[0026] Here, the congestion level estimation unit 120 estimates (calculates) the congestion level using a congestion level estimation model that is a weighted sum of the number of events (especially irregular events) that occurred when the doors were opened and closed at the station and a weighting parameter that was previously set for each event.
[0027] Specifically, the set of events for which congestion levels are to be estimated is defined as E = {E i} i∈{1,··,I} (I is the total number of events in set E), when the door opens or closes at the station, event E i The number of times that occurs is N i , Event E i The weight parameter set for Wi Then, the congestion level R at the relevant door in the station is R = W1N1 + + W I N I However, for example, only events that occur when the doors are opened or closed may be used as the congestion degree estimation target. Also, past congestion degrees (for example, congestion degrees estimated at the previous station, or a series of congestion degrees estimated at stations prior to the current station, etc.) may be treated as events to be used as the congestion degree estimation target, in which case the congestion degree value itself may be used as the number of occurrences of the event.
[0028] Incidentally, an event that occurs when the door is opened or closed can be identified from the sensor information and the opening / closing control command stored in the storage unit 140. For example, an event that "the door that was about to close is opened once and then closed again" during the closing operation of the door can be identified from the sensor information of the opening / closing sensor 201. Also, for example, an event that "something is caught in the door" during the closing operation of the door can be identified from the sensor information of the door-catch detection sensor 202 (and, in addition, the door-catch detection signal after receiving this sensor information, etc.). In addition, an event that "the current flowing to the motor is increased due to an external force" during the closing operation of the door can be identified from the sensor information of the external force measurement sensor 203 (and, in addition, the opening / closing control command after receiving this sensor information). Similarly, an event that "the current flowing to the motor is increased significantly due to a large external force" during the closing operation of the door can be identified from the sensor information of the external force measurement sensor 203 (and, in addition, the opening / closing control command after receiving this sensor information). When the door is opened or closed, the event that "the door is opened or closed more slowly than usual" can be identified by the opening / closing speed detection sensor 204.
[0029] The congestion degree transmission unit 130 transmits the congestion degree R estimated by the congestion degree estimation unit 120 (i.e., the congestion degree of the door after the door is opened and closed at the station) to the notification system 20. This enables the notification system 20 to notify the notification target of the congestion degree of each door of each car after departure from the station.
[0030] <Overall processing of the congestion level estimation system 1> Next, the flow of the overall process of the congestion degree estimation system 1 according to the present embodiment is shown in Fig. 4. Below, the overall process when a train arrives at a certain station will be described with reference to Fig. 4. Note that the overall process shown in Fig. 4 is repeatedly executed every time a train arrives at a station.
[0031] When a train arrives at a station, the doors are opened by a train crew member or station staff member, and then the doors are closed after passengers have disembarked and boarded the train. These opening and closing operations result in an opening command and a closing command, respectively, being sent to the control device 100. When the opening / closing control unit 110 of the control device 100 receives an opening command or a closing command or sensor information, it sends an opening / closing control command based on a known door control method to the door closing device 300 in response to the opening command or the closing command or the sensor information (step S101). This controls the opening and closing of the doors. After the doors are closed, the train departs for the next station.
[0032] Next, the congestion level estimation unit 120 of the control device 100 estimates the congestion level at the door in the station using the sensor information and the opening / closing control command stored in the storage unit 140 (step S102).
[0033] Next, the congestion level estimation unit 120 of the control device 100 stores the congestion level estimated in the above step S102 in the storage unit 140 as a log (step S103).
[0034] Then, the congestion degree transmitting unit 130 of the control device 100 transmits the congestion degree estimated in the above step S102 to the notification system 20 (step S104).
[0035] When the notification system 20 receives the congestion degree from each door system 10, it transmits the congestion degree for each door of each car to the notification target (step S105). This allows, for example, passengers who plan to board at the next station and station staff at the next station to know the congestion degree for each door of each car. Therefore, for example, it is possible to encourage passengers to board in a dispersed manner so as to avoid crowded doors, which is expected to alleviate congestion in the entire car.
[0036] Instead of directly transmitting the congestion level itself in step S104 above, the congestion level may be converted into a category that is easy for people to understand, and then this category may be transmitted to the notification system 20. This conversion can be performed, for example, by using a congestion level table stored in the storage unit 140.
[0037] Here, an example of the congestion level table is shown in FIG. 5. The congestion level table shown in FIG. 5 is a table for converting the congestion level R into a congestion level category and a boarding rate. In the congestion level table shown in FIG. 5, when the congestion level R satisfies 0 ≦ R ≦ 1, it is converted into "low congestion level (boarding possible)" and "boarding rate less than 100%"; when 1 < R ≦ 2, it is converted into "medium congestion level (boarding possible)" and "boarding rate 100% or more and less than 180%". Similarly, when 2 < R ≦ 3, it is converted into "high congestion level (difficult to board)" and "boarding rate 100% or more and less than 180%"; when 3 < R, it is converted into "extremely high congestion level (impossible to board)" and "boarding rate 250% or more". Note that the relationship between the congestion level category and the boarding rate in the congestion level table shown in FIG. 5 refers to the congestion level standard presented by the Ministry of Land, Infrastructure, Transport and Tourism (https: / / www.mlit.go.jp / tetudo / toshitetu / 03_04.html), but this is just an example, and it goes without saying that it can be defined in various ways.
[0038] However, the above conversion may be performed by the notification system 20 instead of the door system 10. That is, in step S105 above, after converting the congestion level into a category that is easy for people to understand (for example, the congestion level category and the boarding rate), the converted category may be notified to the notification target.
[0039] <Example> Next, an example will be described. In this example, five events, namely, "door reopening / closing", "large current", "extremely large current", "low opening / closing speed", and "congestion level of the previous station", are targeted as events for which the congestion level is estimated, and the congestion level of a certain door is estimated.
[0040] "Door reopening" refers to the event when "a door that was about to close opens once and then closes again" during the door closing operation. "High current" refers to the event when "an external force increases the current flowing through the motor" during the door closing operation, and "extra large current" refers to the event when "a large external force increases the current flowing through the motor very much" during the door closing operation. Furthermore, "slow opening / closing speed" refers to the event when "the door opens and closes more slowly than usual" during the door opening and closing operation. Furthermore, "previous station congestion level" refers to the estimated congestion level at the previous station.
[0041] In addition, the weighting parameters for each event were set as shown in Fig. 6. That is, for "door reopening", the weighting parameter W A = 0.2, the weighting parameter W for "large current" B1 = 0.45, and the weighting parameter W B2 = 1.2, and the weight parameter W for "low opening / closing speed" C = 0.3, and the weight parameter W RN-1 =0.6. In this way, when congestion occurs to the extent that boarding is difficult, the external force on the door becomes large and the current flowing through the motor becomes large, which is a notable characteristic, so the weighting parameter values for "large current" and "extra large current" are made relatively large. In addition, since the congestion level at the previous station directly affects the congestion level at the current station, the weighting parameter value for "previous station congestion level" is also made relatively large. On the other hand, door reopening and slowing down of opening and closing speed often occur due to sudden events other than when the train is crowded (e.g., door jamming, rushing to board, etc.), so the weighting parameter values for "door reopening" and "slow opening and closing speed" are made relatively small. However, since congestion is expected when "door reopening" and "slow opening and closing speed" occur multiple times or simultaneously, the values are not too small.
[0042] At this time, a case where the congestion degree is estimated from the starting station to the next station to the station after that to the station after that will be described with reference to FIG.
[0043] First, assume that the door reopening occurs only once when the door is opened and closed at the starting station. In this case, the congestion degree R1 of the door at the time of departure from the starting station is R1 = W A ×1=0.2×1=0.2. When this is converted into a congestion level classification using the congestion level table shown in FIG.
[0044] Next, assume that when the door is opened at the next station, "extra large current" occurs once and "slow opening / closing speed" occurs once. In this case, the congestion degree R2 of the door at the time of departure from the next station is R2 = W B2 ×1+W C ×1+W RN-1 × R1 = 1.2 × 1 + 0.3 × 1 + 0.6 × 0.2 = 1.62. When this is converted into a congestion level classification using the congestion level table shown in FIG.
[0045] Next, assume that when the door is opened and closed at the station after the next, "door reopening" occurs once and "extra-high current" occurs once. In this case, the congestion degree R3 of the door at the time of departure from the station after the next is R3 = W A ×1+W B2 ×1+W RN-1 × R2 = 0.2 × 1 + 1.2 × 1 + 0.6 × 1.62 = 2.37. When this is converted into a congestion level classification using the congestion level table shown in FIG.
[0046] Next, assume that a "high current" occurs only once when the door is opened and closed at the station two stations after the next. In this case, the congestion level R4 of the door at the time of departure from the station two stations after the next is R4 = W B1 ×1+W RN-1 × R3 = 0.45 × 1 + 0.6 × 2.37 = 1.87. When this is converted into a congestion level classification using the congestion level table shown in FIG.
[0047] <Summary> As described above, the congestion degree estimation system 1 according to this embodiment can estimate the congestion degree of a door from the number of events that occur when the door is opened and closed (especially irregular events such as "a door that was about to close is opened once and then closed again" or "an external force increases the current flowing through the motor"). These events can be identified from the sensor information of the sensors (i.e., the sensor group 200) included in the existing door system, so the congestion degree estimation system 1 according to this embodiment can easily estimate the congestion degree of each door without introducing new equipment.
[0048] In addition, since the congestion level estimation system 1 according to this embodiment uses a weighted sum of the number of occurrences of an event as a congestion level estimation model, it is possible to estimate the congestion level with high accuracy by setting the weighting parameters to appropriate values.
[0049] The weight parameters must be set to appropriate values in advance, but may be set using various methods other than the setting based on insight or empirical rules as described in the above embodiment. For example, the optimal weight parameter value for the congestion degree estimation model may be determined by analyzing or learning the relationship between the sensor information or the opening / closing control command and the congestion degree using methods such as data analysis or machine learning. In this case, even if the weight parameter value is for the same event, it may be different for each door or each vehicle. Moreover, the weight parameter value does not necessarily have to be positive, and may be a negative value.
[0050] Furthermore, other models may be used as the congestion level estimation model. For example, a model using an algorithm such as a regression tree may be used as the congestion level estimation model.
[0051] The present invention is not limited to the above-described embodiments specifically disclosed, and various modifications, changes, combinations with known technologies, etc. are possible without departing from the scope of the claims. [Explanation of symbols]
[0052] 1. Congestion estimation system 10 Door System 20 Notification System 100 Control device 101 Processor 102 Memory device 103 Interface 104 Bus 110 Opening and closing control section 120 Congestion degree estimation unit 130 Congestion level transmitter 140 Storage section 200 Sensors 201 Open / close sensor 202 Door jamming detection sensor 203 External Force Measurement Sensor 204 Opening and closing speed detection sensor 300 Door closing device
Claims
1. A congestion degree estimation system that estimates a congestion degree for each door of a railway car, an estimation unit that estimates a congestion degree at the door using a predetermined congestion degree estimation model based on sensor information indicating a sensing result of a sensor used for opening and closing control of the door; A notification unit that notifies a notification target of the congestion degree at the door; A congestion estimation system having the above configuration.
2. the congestion degree estimation model is represented by a weighted sum of a number of predetermined events that have occurred during the opening and closing control of the door and a weight parameter that is predetermined for the event, The estimation unit is 2. The congestion degree estimation system according to claim 1, further comprising: a controller for detecting a congestion degree of the vehicle; a controller for detecting a congestion degree of the vehicle; and a controller for detecting a congestion degree of the vehicle.
3. The estimation unit is The congestion level estimation system according to claim 2 , further comprising: a step of identifying a predetermined event that has occurred during control of opening and closing of the door, based on a control command for controlling opening and closing of the door.
4. The congestion degree estimation system according to claim 3 , wherein the value of the weighting parameter is determined by analyzing or learning a relationship between at least one of the sensor information and the control command and the congestion degree at the door through data analysis or machine learning.
5. 5. The congestion level estimation system according to claim 2, wherein the specified event includes at least one of the following events: a door being reopened or closed, an object being trapped in the door, an increase in current flowing to a drive device that opens and closes the door due to an increase in external force on the door, and a decrease in the opening and closing speed of the door.
6. A congestion estimation system that estimates the degree of congestion at each door of a railway car has been developed. an estimation step of estimating a congestion degree at the door using a predetermined congestion degree estimation model based on sensor information indicating a sensing result of a sensor used for opening and closing control of the door; a notification step of notifying a notification target of the congestion degree at the door; A congestion estimation method that performs the above.
Citation Information
Patent Citations
Vehicle congestion information reading system
JP2002193102A
Door drive control device
JP2005145240A
Apparatus for controlling electrically-driven door
JP2010053529A
Information service system for congestion in train
JP2012101640A
Electronic equipment
JP2017205531A
