Sensor Network

Abstract

To provide a sensor network capable of reducing an amount of data transmitted from a plurality of sensors while maintaining estimation accuracy of a state based on sensor data output from the plurality of sensors of different kinds.SOLUTION: When first sensor data SD1 output from a proximity sensor S1 owned by a sensor module Smod(i) represents a first specified state, only first composite sensor data Dt1 including the first sensor data SD1 and a type of the proximity sensor S1 or a first flag F1 representing the first specified state is transmitted from the sensor module Smod(i) to a central processing unit Smod(0) via a network. In this case, second composite sensor data Dt2 including second sensor data SD2 and a type of a contact force sensor S2 or a second flag F2 representing a second specific state is not transmitted from the sensor module Smod(i) to the central processing unit Smod(0) via the network.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a sensor network having a plurality of sensor modules as components. [Background technology]

[0002] When multiple sensors are arranged on a target object, the amount of data output from the multiple sensors increases as the number of sensors increases (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-092319 A Summary of the Invention [Problem to be solved by the invention]

[0004] If a composite sensor consisting of a proximity sensor for detecting the proximity of a target object to other objects and a contact sensor for detecting the pressing force of the target object against other objects is distributed over the target object, the amount of data collected from the composite sensor will increase significantly, which may hinder the processing of the data.

[0005] Therefore, an object of the present invention is to provide a sensor network that can reduce the amount of data transmitted from multiple sensors while maintaining the accuracy of state estimation based on sensor data output from the multiple sensors of different types. [Means for solving the problem]

[0006] The sensor network of the present invention comprises: A sensor network comprising a plurality of sensor modules and a central processing unit for processing data transmitted from each of the plurality of sensor modules via a network, Each of the plurality of sensor modules is Multiple sensors with different detection targets; a sensor data processing unit that determines whether each of a plurality of sensor data output from each of the plurality of sensors represents a respective one of a plurality of designated states, and causes a sensor data communication unit to transmit to the central processing unit via the network only composite sensor data that includes a piece of sensor data determined to represent a designated state among the plurality of designated states and a flag for identifying a type of a sensor that has output the piece of sensor data or the designated state; It is equipped with:

[0007] According to a sensor network of this configuration, when a piece of sensor data output from a sensor among a plurality of different types of sensors possessed by a sensor module represents a specified state, i.e., when the piece of sensor data is highly useful from the viewpoint of estimating the state of the location of the sensor module (and / or its surroundings), composite sensor data including the piece of sensor data and a flag representing the type of the sensor or the piece of specified state is transmitted from the sensor module to a central processing unit via the network.

[0008] On the other hand, if other sensor data output from other sensors among the multiple sensors of different types possessed by the sensor module does not represent other specified states, i.e., if the usefulness of the other sensor data is low from the perspective of estimating the state of the location where the sensor module is placed, composite sensor data including the other sensor data and a flag representing the type of the other sensor or the other specified state is not transmitted from the sensor module to the central processing unit via the network.

[0009] Therefore, while maintaining the accuracy of estimating the state of each location where each sensor is installed based on sensor data output from multiple sensors of different types, the amount of data transmitted from the multiple sensors to a central processing unit via a network is reduced.

[0010] In the sensor network of the present invention, The sensor data processing unit determines whether the sensor data represents the specified state depending on whether at least one of the time-series change amounts of the sensor data exceeds a threshold value in addition to or instead of whether the sensor data is included in a specified range. It is preferred.

[0011] According to a sensor network of this configuration, when a piece of sensor data output from one of a plurality of different types of sensors possessed by a sensor module is included in a specified range and / or the amount of change over time is equal to or greater than a threshold, i.e., when the usefulness of the piece of sensor data is high from the viewpoint of estimating the state of the location where the sensor module is installed, composite sensor data including the piece of sensor data is transmitted from the sensor module to a central processing unit via the network.

[0012] On the other hand, if the sensor data output from another sensor among the multiple sensors of different types possessed by the sensor module is outside the specified range and / or the amount of change over time is less than the threshold, i.e., if the usefulness of the other sensor data is low from the perspective of estimating the state of the location where the sensor module is installed, the composite sensor data including the other sensor data is not transmitted from the sensor module to the central processing unit via the network.

[0013] Therefore, while maintaining the accuracy of estimating the state of each location where each sensor is installed based on sensor data output from multiple sensors of different types, the amount of data transmitted from the multiple sensors to a central processing unit via a network is reduced.

[0014] In the sensor network of the present invention, The sensor data processing unit determines whether or not each of the plurality of sensor data represents each of a plurality of designated states that are not realized simultaneously, based on each of the plurality of sensor data. It is preferred.

[0015] According to the sensor network of this configuration, it is not determined that the location of the sensor module is in one designated state and in another designated state at the same time, based on each of the multiple sensor data output from each of the multiple sensors of different types that the sensor module has. Therefore, when composite sensor data including one sensor data is transmitted from the sensor module to the central processing unit via the network, composite sensor data including the other sensor data is not transmitted from the sensor module to the central processing unit via the network. Therefore, while maintaining the estimation accuracy of the state of each location where each sensor is installed based on the sensor data output from the multiple sensors of different types, the amount of data transmitted from the multiple sensors to the central processing unit via the network is reduced.

[0016] In the sensor network of the present invention, The sensor data processing unit determines whether or not the one piece of sensor data represents the one specified state based on other sensor data output from another sensor among the plurality of sensors in addition to the one piece of sensor data. It is preferred.

[0017] According to the sensor network of this configuration, in each sensor module, in addition to one sensor data output from one sensor, other sensor data output from other sensors are taken into consideration, so that it is possible to determine with high accuracy whether the first sensor data represents the first specified state or not. Therefore, while maintaining the estimation accuracy of the state of each location where each sensor is installed based on the sensor data output from multiple sensors of different types, the amount of data transmitted from the multiple sensors to the central processing unit via the network is reduced.

[0018] In the sensor network of the present invention, Each of the plurality of sensor modules is The one sensor includes a proximity sensor for detecting a proximity state of a target object, The other sensor includes a contact force sensor for detecting the presence or absence of contact with a target object and the strength of the contact force. It is preferred.

[0019] With a sensor network of this configuration, it is possible to maintain the accuracy of estimating the proximity and contact states of an object relative to other objects based on the sensor data output from each of the proximity sensors and contact force sensors possessed by each sensor module, while reducing the amount of data transmitted from the proximity sensors and contact force sensors.

[0020] In the sensor network of the present invention, The sensor data processing unit generates the composite sensor data including a placement factor representing a placement of the sensor modules. It is preferred.

[0021] The sensor network of this configuration maintains the accuracy of estimating the state of each of a plurality of locations on an object according to the arrangement of each sensor module based on the sensor data output from a plurality of different types of sensors, and also reduces the amount of data transmitted from the plurality of sensors as described above.

[0022] In the sensor network of the present invention, The central processing unit: A plurality of arithmetic processing units; a central data processing unit that identifies the type of the sensor or the designated state by the flag and transmits the composite sensor data to a processing unit among the plurality of processing units that corresponds to the type of the sensor or the designated state; It is preferable that the device further comprises:

[0023] According to the sensor network having the above configuration, the composite sensor data is transmitted to a processor for processing the sensor data included in the composite sensor data among the multiple processors in the central processing unit. In other words, composite sensor data that does not include the sensor data to be processed by each processor is not transmitted to the processor. Therefore, while maintaining the accuracy of estimating the state based on the sensor data output from the multiple sensors of different types as described above, the amount of data transmitted from the multiple sensors is reduced, and the computation load of each control unit constituting the central processing unit is also reduced. [Brief description of the drawings]

[0024] [Figure 1] FIG. 1 is an explanatory diagram relating to the configuration of a sensor network according to an embodiment of the present invention. [Diagram 2] FIG. 2 is an explanatory diagram relating to the configuration of a sensor module and a central processing unit. [Figure 3A] FIG. 13 is an explanatory diagram of two composite sensor data. [Figure 3B] FIG. 11 is an explanatory diagram regarding one of the composite sensor data. [Figure 3C] FIG. 11 is an explanatory diagram regarding the other composite sensor data. [Figure 4A] An illustration of how a finger interacts with a surface. [Figure 4B] 10A and 10B are explanatory diagrams relating to a mode of transmission of composite sensor data from a sensor module placed on a finger [Diagram 5] FIG. 13 is an explanatory diagram illustrating the configuration of a sensor network according to another embodiment of the present invention. [Figure 6] FIG. 13 is an explanatory diagram illustrating the configuration of a sensor network according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] The sensor network according to one embodiment of the present invention shown in FIG. 1 is composed of a plurality of (n) sensor modules Smod(1) to Smod(n) and a central processing unit Smod(0) for processing data transmitted from each of the plurality of sensor modules Smod(1) to Smod(n) via the network.

[0026] As shown in FIG. 1, starting from a central processing unit Smod(0), a first sensor module Smod(1), a second sensor module Smod(2), ... the (i-1)th sensor module Smod(i-1), the i-th sensor module Smod(i), the (i+1)th sensor module Smod(i+1), ... the (n-1)th sensor module Smod(n-1) and the n-th sensor module Smod(n) are connected in sequence, and the n-th sensor module Smod(n) is connected to the central processing unit Smod(0) to form a ring network.

[0027] The i-th sensor module Smod(i) (i=1, 2, . . . n) is, for example, distributed over multiple locations on multiple fingers and the palm of a robot hand that is the object to be controlled.

[0028] As shown in FIG. 2, the central processing unit Smod(0) includes a central data communication unit 40, a central data processing unit 41, a first arithmetic processing unit 421, and a second arithmetic processing element 422. Small At least a portion of the equipment is composed of a storage device (memory such as ROM, RAM, EEPROM, SSD, HDD, etc.) that stores and retains software and / or data, and a calculation processing device (single-core processor, multi-core processor, CPU, etc.) that reads the necessary software and / or data from the storage device and performs calculations on the data.

[0029] The central data communication unit 40 is configured to transmit and receive data to and from the i-th sensor module Smod(i) (i=1, 2, . . . n) via the network. The central data processing unit 41 is configured to identify the type of sensor or the designated state by a flag included in the composite sensor data (described later) received by the central data communication unit 40, and to transmit the composite sensor data to one of the first arithmetic processing unit 421 and the second arithmetic processing unit 422 that corresponds to the type of sensor or the designated state.

[0030] Each of the first arithmetic processing unit 421 and the second arithmetic processing unit 422 transmits a control command signal to one or more actuators 44 of the control target, thereby controlling the operation of the actuators 44. The control target is, for example, a robot hand having a palm portion and multiple fingers extending from the palm portion, or further, a robot arm having the robot hand as an end effector.

[0031] The first arithmetic processing unit 421 sets gripping positions of the object with the fingers based on the first sensor data SD1 included in the first composite sensor data Dt1 and representing the proximity state between each of the multiple points of the robot hand to be controlled and the object, and transmits a first control command signal to the actuator 44 to bring the gripping positions of the fingers into contact with the object. The second arithmetic processing unit 422 sets a force to be applied to the object from the fingers and / or palm based on the second sensor data SD2 included in the second composite sensor data Dt2 and representing the contact state between each of the multiple points of the robot hand to be controlled and the object, and transmits a second control command signal to the actuator 44 to increase or decrease the force to be applied to the object from the fingers and / or palm.

[0032] As shown in Figure 2, the i-th sensor module Smod(i) includes a proximity sensor S1 (first sensor) for detecting a proximity state to a target object, a contact force sensor S2 (second sensor) for detecting the presence or absence of contact with the target object and pressure (reaction force received from the object), a sensor data processing unit 21, and a sensor data communication unit 22.

[0033] The sensor data processing unit 21 shown in FIG. 2 is composed of a storage device (memory such as ROM, RAM, EEPROM, SSD, HDD, etc.) that stores and retains software and / or data, and an arithmetic processing device (single-core processor, multi-core processor, CPU, etc.) that reads the necessary software and / or data from the storage device and performs arithmetic processing on the data.

[0034] Each of the proximity sensor S1 and the contact force sensor S2 shown in FIG. 2 is, for example, a capacitance type sensor, and the multiple states are determined by the sensor data processing unit 21 based on the capacitance C1 measured by the proximity sensor S1 and the capacitance C2 measured by the contact force sensor S2.

[0035] Table 1 shows a summary of the state determination results according to the measurement results of the capacitance C1 of the proximity sensor S1 and the measurement results of the capacitance C2 of the contact force sensor S2.

[0036] [Table 1]

[0037] The sensor data processing unit 21 receives the first sensor data output from the proximity sensor S1. SD1 Based on the second sensor data output from the contact force sensor S2, the sensor data processing unit 21 determines the states (1) to (3) as a proximity state (first specified state) in which the arrangement position of the i-th sensor module Smod(i) is in proximity to the target object. SD2Based on this, the states (4) to (5) are determined as contact states (second specified states) in which the arrangement location of the i-th sensor module Smod(i) is in contact with or pressing against the target object.

[0038] As shown in FIG. 3A, the sensor data processing unit 21 generates first composite sensor data SD1 including the first sensor data SD1 output from the proximity sensor S1 and a first flag F1 indicating the type of the proximity sensor S1. Dt1 and second composite sensor data including second sensor data SD2 output from the contact force sensor S2 and a second flag F2 indicating the type of the contact force sensor S2. Dt2 The scalar has the function of generating each of the above.

[0039] However, when the sensor data processing unit 21 determines that the arrangement location of the i-th sensor module Smod(i) is in a proximity state (first designated state) where the arrangement location is in proximity to the target object, the sensor data processing unit 21 causes the sensor data communication unit 22 to transmit only the first composite sensor data Dt1 including the first sensor data SD1 and a first flag F1 for identifying the type of the proximity sensor S1 that output the first sensor data SD1 or the first designated state, to the central processing unit Smod(0) via the network, as shown in Fig. 3B. At this time, the generation of the second composite sensor data Dt2 may be omitted.

[0040] Furthermore, when the sensor data processing unit 21 determines that the arrangement location of the i-th sensor module Smod(i) is in a contact state (second designated state) in which the arrangement location is in contact with or pressing against the target object, the sensor data processing unit 21 causes the sensor data communication unit 22 to transmit only the second composite sensor data Dt2 including the second sensor data SD2 and a second flag F2 for identifying the type of the contact force sensor S2 that output the second sensor data SD2 or the second designated state, to the central processing unit Smod(0) via the network, as shown in Fig. 3C. At this time, the generation of the first composite sensor data Dt1 may be omitted.

[0041] For example, as shown in FIG. 4A, a case will be considered in which a finger of a robot hand, on whose surface a sensor module Smod(i) is arranged, is moved along an uneven object surface. Specifically, as shown in FIG. 4A, when the finger is in contact with the object surface (at time t=t i-2 ), when the finger is close to the object surface (time t = t i-1 ), the finger is in contact with the surface of the object (time t = t i and t i+1 ), when the finger is close to the object surface (time t = t i+2 ), and when the finger is removed from the surface (time t = t i+3 ) state transitions.

[0042] In this case, as shown in FIG. 4B, at time t=t i-2 ~t i+2 According to each state of i-2 ), the first composite sensor data Dt1(t i-1 ), the second composite sensor data Dt2(t i ), the second composite sensor data Dt2(t i+1 ) and the first composite sensor data Dt1(t i-2 ) are sequentially transmitted from the sensor module Smod(i) to the central processor Smod(0). i+3 In response to this state, neither the first composite sensor data Dt1 nor the second composite sensor data Dt2 is transmitted from the sensor module Smod(i) to the central processing unit Smod(0).

[0043] The sensor data communication unit 22 constituting the i-th sensor module Smod(i) transmits and receives sensor data between the data communication unit 22 constituting the adjacent i-1th sensor module Smod(i-1) and / or the i+1th sensor module Smod(i+1) via the network.

[0044] The sensor data transmitted from the ith sensor module Smod(i) is transferred by the i-1th sensor module Smod(i-1) to the i-2nd sensor module Smod(i-2). Then, the sensor data is transferred to the central processing unit Smod(0) in the order of the ith sensor module Smod(i) → the i-1st sensor module Smod(i-1) → → the i-2nd sensor module Smod(i-2) → ... → the second sensor module Smod(2) → the first sensor module Smod(1) → the central processing unit Smod(0).

[0045] Conversely, the sensor data transmitted from the i-th sensor module Smod(i) may be transferred to the i;2 sensor module Smod(i+2) by the i+1-th sensor module Smod(i+1). Then, the sensor data is transferred from the i-th sensor module Smod(i) to the i+1-th sensor module Smod(i+1). → The signal may be transferred to the central processing unit Smod(0) in the order of the i+2 sensor module Smod(i+2) → . . . → the n-1th sensor module Smod(n-1) → the nth sensor module Smod(n) → the central processing unit Smod(0).

[0046] (Action and effect) According to the sensor network having the above configuration, when the first sensor data SD1 output from the proximity sensor S1 of the sensor module Smod(i) indicates the first specified state, i.e., when the first sensor data SD1 is highly useful from the viewpoint of estimating the state of the location (and / or the surroundings) of the sensor module Smod(i), only the first composite sensor data Dt1 including the first sensor data SD1 and the first flag F1 indicating the type of the proximity sensor S1 or the first specified state is transmitted from the sensor module Smod(i) to the central processing unit Smod(0) via the network (see FIGS. 3B and 4B (at time t=t i-1 , t i+2 )reference).

[0047] In this embodiment, this corresponds to a case where the second sensor data SD2 output from the contact force sensor S2 does not represent the second specified state, so the second composite sensor data Dt2 including the second sensor data SD2 and the second flag F2 representing the type of the contact force sensor S2 or the second specified state is not transmitted from the sensor module Smod(i) to the central processing unit Smod(0) via the network (see FIGS. 3B and 4B (at time t=t i-1 , t i+2 )reference).

[0048] On the other hand, when the second sensor data SD2 output from the contact force sensor S2 of the sensor module Smod(i) indicates the second specified state, that is, when the second sensor data SD2 is highly useful from the viewpoint of estimating the state of the arrangement location (and / or its surroundings) of the sensor module Smod(i), only the second composite sensor data Dt2 including the second sensor data SD2 and the second flag F2 indicating the type of the contact force sensor S2 or the second specified state is transmitted from the sensor module Smod(i) to the central processing unit Smod(0) via the network (see FIGS. 3C and 4B (at time t=t i-2 , t i , t i+1 )reference).

[0049] In this embodiment, this corresponds to a case where the first sensor data SD1 output from the proximity sensor S1 does not represent the first designated state, so the first composite sensor data Dt1 including the first sensor data SD1 and the first flag F1 representing the type of the proximity sensor S1 or the first designated state is not transmitted from the sensor module Smod(i) to the central processing unit Smod(0) via the network (see FIGS. 3C and 4B (at time t=t i-2 , t i , t i+1 )reference).

[0050] The sensor data processing unit 21 calculates the first sensor data SD1 based on the second sensor data SD2. SD1indicates the first designated state (proximity state) (see states (2) and (4) in Table 1). This makes it possible to more accurately determine whether the first sensor data SD1 indicates the first designated state.

[0051] Therefore, while maintaining the accuracy of estimating the state of the location of the sensor module Smod(i) having each sensor S1, S2 based on the sensor data output from multiple sensors S1, S2 of different types, the amount of data transmitted from the multiple sensors S1, S2 to the central processing unit Smod(0) via the network is reduced.

[0052] Among the multiple arithmetic processing units of the central processing unit Smod(0), the composite sensor data Dt1, Dt2 is transmitted to the first arithmetic processing unit 421 and the second arithmetic processing unit 422 for processing the sensor data SD1, SD2 included in the composite sensor data Dt1, Dt2. That is, composite sensor data that does not include sensor data to be processed by each of the arithmetic processing units 421, 422 is not transmitted to the corresponding arithmetic processing unit. Therefore, while maintaining the state estimation accuracy based on the sensor data SD1, SD2 output from the multiple sensors S1, S2 of different types as described above, the amount of data transmitted from the multiple sensor modules Smod(i) is reduced, and further the arithmetic processing load of each of the arithmetic processing units 421, 422 constituting the central processing unit Smod(0) is reduced.

[0053] (Another embodiment of the present invention) In the above embodiment, the sensor network is configured by the ring network shown in FIG. 1, but in other embodiments, the sensor network may be configured by various other ring networks different from this.

[0054] For example, as shown in FIG. 5, a sensor network may be formed by a sub-network extending in a branch shape from an i-th sensor module Smod(i) that forms a ring network, with multiple sub-sensor modules Smod(i_1), Smod(i_2), ... connected to the ends of the branches.

[0055] 6, a sensor network may be configured by a sub-network extending in a ring shape with a plurality of sub-sensor modules Smod(i_1), Smod(i_2), ... connected in sequence and an i-th sensor module Smod(i) constituting a ring network as a base point. In this case, the i-th sensor module Smod(i) may have the function of a central processing unit for the plurality of sub-sensor modules Smod(i_1), Smod(i_2), ....

[0056] In the above embodiment, the sensor module Smod(i) is equipped with a proximity sensor S1 and a contact force sensor S2, but additionally or alternatively, the sensor module may be equipped with a plurality of different types of sensors for detecting different physical quantities and different specified states, such as a visual sensor, a radiation dose sensor, a magnetic sensor, an olfactory sensor and / or a taste sensor.

[0057] In the above embodiment, the first specified state (see states (1) to (3) in Table 1) and the second specified state (see states (4) to (5) in Table 1) are defined as states in which they are not realized simultaneously, but in other embodiments, the first specified state and the second specified state may be defined as states in which they are realized simultaneously (see state (3) in Table 1). As described above, when the sensor module Smod(i) is equipped with multiple sensors of different types for detecting different physical quantities and different specified states, the multiple specified states determined based on each of the multiple sensor data output from each of the multiple sensors may be defined as states in which they are realized simultaneously.

[0058] In the above-described embodiment, in addition to the first sensor data SD1, it is determined whether or not the first sensor data SD1 represents a first specified state based on the second sensor data SD2 (see states (2) and (4) in Table 1). As described above, when the sensor module Smod(i) includes a plurality of sensors of different types for detecting different physical quantities and different specified states, each specified state can be determined based on each sensor data output from each of the plurality of sensors.

[0059] In the above-described embodiment, the sensor data processing unit 21 determines whether or not the first sensor data SD1 (capacitance C1) represents a first specified state (proximity state) according to whether or not the first sensor data SD1 is included in a specified range (Cp_min ≦ C1 < Cp) (see Table 1). Instead of or in addition to this, it may be determined whether or not the first sensor data SD1 represents a first specified state (proximity state or a state where the amount of change in the distance from an object is large) according to whether or not the amount of temporal change in the first sensor data SD1 exceeds a threshold value.

[0060] In the above-described embodiment, the sensor data processing unit 21 determines whether or not the second sensor data SD2 (capacitance C2) represents a second specified state (contact state) according to whether or not the second sensor data SD2 is equal to or greater than a reference value Cf (whether or not it is included in a specified range (Cf < C2)). Instead of or in addition to this, it may be determined whether or not the second sensor data SD2 represents a second specified state (contact state or a state where the change in the acting force received from an object is large) according to whether or not the amount of temporal change in the second sensor data SD2 exceeds a threshold value.

[0061] According to the sensor network of this configuration, in each sensor module, in addition to one sensor data output from one sensor, other sensor data output from other sensors are taken into consideration, so that it is possible to determine with high accuracy whether the first sensor data represents the first specified state or not. Therefore, while maintaining the estimation accuracy of the state of each location where each sensor is installed based on the sensor data output from multiple sensors of different types, the amount of data transmitted from the multiple sensors to the central processing unit via the network is reduced.

[0062] The sensor data processing unit 21 may generate the composite sensor data Dt including a placement factor representing the placement of the sensor modules Smod(i). In this case, the estimation accuracy of the state at each of a plurality of locations on an object according to the placement of each sensor module Smod(i) is maintained based on the sensor data SD output from a plurality of different types of sensors S1, S2. [Explanation of symbols]

[0063] 10: substrate, 21: sensor data processing unit, 22: sensor data communication unit, 40: central data communication unit, 41: central data processing unit, 421: first arithmetic processing unit, 422: second arithmetic processing unit, S1: proximity sensor (first sensor), S2: contact force sensor (second sensor), Smod(0): central processing unit, Smod(i): sensor module.

Claims

1. A sensor network comprising a plurality of sensor modules and a central processing unit for processing data transmitted from each of the plurality of sensor modules via a network, Each of the plurality of sensor modules is Multiple sensors with different detection targets; a sensor data processing unit that determines whether each of a plurality of sensor data output from each of the plurality of sensors represents a respective one of a plurality of designated states, and causes a sensor data communication unit to transmit to the central processing unit via the network only composite sensor data that includes a piece of sensor data determined to represent a designated state among the plurality of designated states and a flag for identifying a type of a sensor that has output the piece of sensor data or the designated state; A sensor network comprising:

2. 2. The sensor network according to claim 1, (i) the sensor data processing unit determines whether or not the sensor data represents the specified state depending on whether or not the sensor data is included in a specified range, or (ii) the sensor data processing unit determines whether or not the sensor data represents the specified state depending on whether at least one of the time-series changes in the sensor data exceeds a threshold, in addition to or instead of whether or not the sensor data is included in a specified range.

3. 3. The sensor network according to claim 1, The sensor data processing unit determines whether or not each of the plurality of sensor data represents each of a plurality of designated states that are not realized simultaneously, based on each of the plurality of sensor data. Sensor network.

4. In the sensor network according to any one of claims 1 to 3, The sensor data processing unit determines whether or not the one piece of sensor data represents the one specified state based on other sensor data output from another sensor among the plurality of sensors in addition to the one piece of sensor data. Sensor network.

5. The sensor network according to claim 4, Each of the plurality of sensor modules is The one sensor includes a proximity sensor for detecting a proximity state of a target object, The other sensor includes a contact force sensor for detecting the presence or absence of contact with a target object and the strength of the contact force. Sensor network.

6. In the sensor network according to any one of claims 1 to 5, The sensor data processing unit generates the composite sensor data including a placement factor representing a placement of the sensor modules. Sensor network.

7. In the sensor network according to any one of claims 1 to 6, The central processing unit: A plurality of arithmetic processing units; a central data processing unit that identifies the type of the sensor or the designated state by the flag and transmits the composite sensor data to a processing unit among the plurality of processing units that corresponds to the type of the sensor or the designated state; A sensor network comprising:

Images