TW201345211A - System and method for conflict management in sensor networks - Google Patents

Abstract

In a conflict management system, a control conflict detector determines a time window of a new control according to a sensor type and an actuator type obtained from the new control, and determines that the new control has a control conflict when the actuator is same as at least one actuator of at least one original control and the time window overlaps with at least one time window of the at least one related original control. A conflict resolution module generates a control attribute of the new control, and compares the control attribute of the new control and at least one control attribute of the at least one original control according to at least one priority rule of a current mode to arbitrate the winner between the new control and the at least one original control.

Description

System and method for sensing management conflicts in a network

The present disclosure is directed to a system and method for sensing management conflict management in a network.

In a sensing network that includes multiple sensors and actuators, the sensors are used to detect events and the actuators are used to evoke motion. There may be many control actions in the sensing network, and control and control may also conflict with each other. When a condition detected by a sensor is met, the specified action is performed by the actuator, and a relationship can be established between the sensor and the actuator. This relationship is called a binding, and the act of performing this link is called control. For example, the event detected by the sensor is "press the switch", and when the "press switch" is established, the "light on" action is performed by the actuator. For example, the event detected by the sensor is "temperature exceeds 30 degrees", and when "temperature exceeds 30 degrees" is established, the actuator performs "cold air opening" operation. As sensing networks become more complex, there is an opportunity for conflict between control and control.

Some patent documents disclose techniques for conflict detection and/or conflict resolution. For example, a smart control technology used in home digital networks can detect context conflicts and service conflicts, and use a priority-based approach to resolve these conflicts. One technique is to detect and resolve context conflicts caused by information from different sources. A resource management system for multimedia playback and storage systems automatically selects an appropriate source of information to obtain stored data, and also detects multiple audio and video by overlapping the time windows of each stored action. The resource conflicts that occur when the data is pre-recorded, and the system presets the processing rules or prompts the user to decide how to resolve the conflict. An architecture for multiple core processors can detect and resolve access conflicts caused by common access memory; the collision detection is based on the fact that multiple cores simultaneously access the same memory within a certain time window In the case of a body address, a conflict occurs, and the conflict resolution is based on the order in which the request is issued (FIFO) to determine that the request is not allowed to be executed. One application in the semiconductor manufacturing process to solve the conflicts caused by multiple production controllers is to judge and resolve this control conflict through a supervisory controller.

In general, solving control conflicts can be done manually/semi-manual/automatically. Manually resolving control conflicts means that when a conflict occurs, the user decides how to resolve the conflict. Resolving control conflicts in a semi-manual manner means that when a conflict occurs, the user decides how to resolve the conflict, but has sufficient algorithm or learning ability to reduce user intervention. To automatically resolve the control conflict means that when a conflict occurs, the user does not need to intervene at all, and the system will resolve the conflict by itself.

There are some non-patent literatures that reveal techniques for conflict detection and/or conflict resolution. For example, a paper proposes a context-aware framework for collecting content on home-sensing platforms and users and proposing a weight-based or priority for conflicts between different users and different services. The solution to the type of conflict belongs to the actuator control conflict and the environmental control conflict. A paper proposes a semantic-based conflict management technique that designs a content-aware and conflict-managed architecture to serve multiple users and multiple groups, and to detect and resolve semantics between services. Control conflicts, the type of which is subject to actuator control conflicts. A paper proposes an arithmetic framework called Physicalnet that manages and arranges heterogeneous sensors/actuators, multiple networks, and multi-user environments with techniques that use fine-grained access rights. The way to resolve conflicts, the type of conflict is part of the sensor configuration control conflict.

A paper proposes a conflict resolution technique to deal with control conflicts between energy-efficient applications and comfort applications. This technology separates priority and privilege-average between different users to achieve conflict control, and uses Radio Frequency Identification (RFID) to identify different users and access user preferences ( Preference), the type of conflict is affiliated with the actuator control conflict. The first figure is an example schematic diagram of the application scenario of this conflict resolution technology. The application scenario in the first figure is a smart energy-saving office 100. The office 100 includes a personalized zone 110 and a public zone 120, and is provided with a central control system (central control no artificial). And multiple individual control systems. Each system is equipped with a thermostat, an RFID reader and antenna, and a sensor and actuator.

One paper is a set of conflict resolution algorithms for multi-user pervasive computing environments. This conflict resolution algorithm considers preemptive based, non-preemptive based, role based, priority, time slice based, democratic based, and personal Factors such as personal preference resolve conflicts, and the type of conflict is subject to actuator control conflicts.

Most of the above control conflict management techniques are not directed to the control conflicts in the sensing network environment, and focus on the aspects of controlling conflicts. The way to control conflicts is not limited to the priority or rule system. A very small number of control conflict management techniques have mechanisms that reveal how to detect control conflicts. Control is the most basic behavioral pattern of the sensing network. As sensing networks become more complex, the chances of a control conflict increase. Controlling the occurrence of conflicts can cause inconvenience to the user; in the worst case, it can endanger life. Therefore, it is important to detect control conflicts and resolve control conflicts in a sensing network environment.

Embodiments of the present disclosure can provide a system and method for sensing control conflicts in a network.

One disclosed embodiment is directed to a system for sensing control conflicts in a sensing network. The system can include a Control Conflict Detector and a Control Conflict Resolution Module. The control conflict detector determines a time window of the new control according to the type of a sensor obtained from a new control and the type of the actuator, and when the actuator A collision is determined to occur when at least one target actuator of at least one original control is identical and the time window overlaps with at least one time window of the at least one original control. The control conflict resolution module generates a control attribute of the new control, and compares the new control with at least one control attribute of the at least one original control to arbitrate which of the at least one original control is the winner (winner).

Another embodiment disclosed is directed to a method of managing control conflicts in a sensing network. The method can include: determining a time window of the new control and assigning a control attribute according to a type of sensor obtained from a new control, and a type of the actuator; when the actuator and the at least one original control At least one target actuator is the same, and the time window overlaps with the at least one time window of the at least one original control, determining that a conflict occurs; and when a conflict occurs, comparing according to at least one priority rule The new control prioritizes the at least one control attribute of the at least one original control to arbitrate which of the at least one of the original controls is the winner.

The above and other advantages of the present invention will be described in detail below with reference to the following drawings, detailed description of the embodiments, and claims.

In a sensing network comprising a plurality of sensors and actuators, the actuator performs the specified action when the condition of the event generated by the sensor is established. Such a series of actions is called control. That is, the control in the sensing network consists of a sensor and an actuator. A control has its own time of action. In this disclosure, this action time is defined as the time window of this control. When a control is still in its own time window, it means that this control still works, so if there are other controlled target actuators that are the same as this control, a control conflict will occur. As sensing networks become more complex, the chances of control conflicts increase, so fixed time window lengths are not suitable for all controls.

In an environment of sensing a network, the disclosed embodiments provide a management mechanism based on a time window to control conflicts. The management mechanism of this control conflict mainly consists of two parts, one is to control conflict detection, and the other is to control conflict resolution. Control collision detection automatically determines the time window size according to the sensor and actuator type, and when the time windows overlap, a control conflict occurs. Control conflict resolution assigns different attributes according to different controls, and determines the priority order of control attributes according to the priority rules of different modes.

In a sensing network, when more than two controls are active for the same actuator, that is, when more than two control actions occur simultaneously by the same actuator, a control conflict occurs. For example, the user turns "cool air on" through the remote control; at the same time, the accumulated power consumption detected by the meter exceeds 100 degrees, and the system automatically turns "cool air off". This will create a control conflict.

Control conflicts may also occur if there is a slight time difference between the two controls. For example, in the example of FIG. A, the user turns off the light through the switch (control 1 occurring at time 1), and the environmental control system turns on the light according to insufficient illumination (control 2 occurring at time 2). For the user, it actually feels conflicting. This is because the user has just turned off the light, but it has been turned off. If there is a large time difference between the two controls, a control conflict may occur. For example, in the example of the second B diagram, the control 1 occurring at time 0 is that the user presses the button to turn on the cold air for four hours; the control 2 that occurs at time 3 is that the system meter detects that the power consumption exceeds 5 degrees. , will turn off the air. Since the action time of the control 1 needs to last for four hours, but the control 2 is generated in the third hour, a control conflict occurs.

Embodiments of the present disclosure generate time windows in the form of sensors (events) and actuators (actions) in response to sensing the complex types of control of the network. That is to say, the length of a controlled time window is determined by the event generated by the sensor and the type of action of the actuator. In the present disclosure, the events generated by the sensor can be divided into a one-time trigger and a conditional trigger. A one-time trigger means that the sensor can only detect a single state and trigger an event when the state is established. For example, events generated by sensors such as television remote controls, electric roller door remotes, doorbell buttons, user interface buttons, and the like. Conditional triggering means that the sensor can detect more than two consecutive states and trigger an event when the specified state is established; for example, a thermometer, a smoke detector, a switch, etc. event.

In the present disclosure, the actuator (action) can be divided into no specified execution time and a specified execution time. The meaning of not specifying the execution time is that when the event detected by the sensor is triggered, the actuator continues to perform an action such as a lamp opening, a gas valve closing, and the like. The purpose of specifying the execution time is that when the event detected by the sensor is triggered, the actuator performs an action within a specified time, such as an operation in which the air conditioner is turned on for one hour and the doorbell is ringed for 10 seconds.

In view of the above, a combination of four time window types can be obtained in accordance with a combination of a sensor (event) and an actuator (action). The third figure illustrates the four types of time windows in accordance with an embodiment of the present disclosure. Type 1 is a one-time trigger for the sensor, the actuator does not specify the execution time; type 2 is the sensor is a one-time trigger, the actuator has a specified execution time; type 3 is the sensor is a conditional trigger The actuator is a condition in which no execution time is specified; the fourth type is that the sensor is a conditional trigger, and the actuator has a specified execution time.

When a control is type one (sensor (event) is a one-time trigger, actuator (action) is no specified execution time), the time window length of this control is infinite; when the time window is controlled by this time event is triggered Produced and will not end. For example, when the TV remote control presses the switch, the TV turns on. The duration of the control is unlimited and the TV does not turn itself off.

When a control is type two (sensor (event) is a one-time trigger, actuator (action) is a specified execution time), the time window length of this control is the execution time for this; the time window is controlled here. The event is generated when it is triggered and disappears after the execution time. For example, when the doorbell button is pressed, the ringer sounds for 10 seconds. The control action time is 10 seconds, and after 10 seconds the control action is over. In other words, the time window of this control is thus specified by the execution time.

When a control is type three (sensor (event) is conditional trigger, actuator (action) is no specified execution time), the length of time window of this control is related to whether the condition is established or not. The time window is generated when the event condition of this control is established, and disappears when the event condition of this control is not established. For example, when the light switch is turned on, the light is turned on. The duration of the control is that the switch is turned on until the switch is turned off. In other words, the time window of this control disappears when the condition of the event does not hold.

When a control is type four (sensor (event) is conditional trigger, actuator (action) is specified execution time), the length of time window of this control is related to whether the specified execution time and condition are established. . The time window is generated when the condition of the event is established, and disappears when the condition of the event does not hold or the longest disappears after the specified execution time ends. For example, the human motion detector detects a person and turns on the light for 3 minutes. The control action time is up to 3 minutes. If no person is detected within 3 minutes, the control action time will end prematurely. In other words, the time window is thus specified by the execution time or disappears when the condition of the event does not hold.

After the control has a time window, it can be determined whether there is a control conflict between the controls. As mentioned earlier, the control in its own time window represents that this control is still functional. The time windows of different controls can be of different lengths. Therefore, if there are more than two controls whose time windows overlap each other and the controlled target actuators are the same, a control conflict occurs. When different controls act on different target actuators, if there is control in which time windows overlap each other, there is no conflict with each other, because when the time windows overlap each other, it is controlled for different target actuators, so There is no control conflict.

Accordingly, according to an embodiment of the present disclosure, the fourth figure is a schematic diagram of a plurality of controls in time in a sensing network system, wherein each square bar graph represents a control, a pattern of different bar graphs The target actuators representing the control are different. Referring to the fourth figure, at time 0, there are control A, control B, and control C in this system. Since Control A, Control B, and Control C are controlled for different targets, there is no conflict with each other. At time 1, the system has Control A, Control B, and Control C. At this time, the control D is to be executed, but it is the same as the target actuator of the control C in the system, so the control C and the control D have a control conflict. At time 2, there is control A in this system. At this time, control E is to be executed. There is no control in the system that is the same as the target actuator of control E. Therefore, control conflict does not occur in control E.

Therefore, when a control is generated, the event generated by the sensor and the action of the actuator can be controlled from then on, and then automatically classified according to the type of the sensor and the type of the actuator. For example, if the control is "the thermometer detects that the temperature is higher than 30 degrees, the cold air is turned on", then the event after the dismantling is controlled as "the thermometer detects that the temperature is higher than 30 degrees", therefore, the sensor type is "Conditional trigger"; the action after the control is disassembled is "air-conditioning", so the actuator type is "no execution time specified". Therefore, after automatic classification, the time window of this control belongs to "Type 3" in the third figure. For another example, if the control is "the doorbell is ringing for 10 seconds when the doorbell switch is pressed", the event after the control is disassembled is "the doorbell switch is pressed", so the sensor type is "one-time trigger". The action after the control is disassembled is "the doorbell rings for 10 seconds", so the actuator type is "the specified execution time". Therefore, after automatic classification, the time window of this control belongs to "Type 2" in the third figure.

After automatic classification, the time window of control can be determined according to the type of time window obtained. In other words, the event generated by a controlled time window by the sensor is determined by the type of action of the actuator. Whether a new control of the same target actuator conflicts with the original control in the system determines whether the time window of the new control overlaps with the respective time window of the original control.

As described above, as shown in FIG. 5, when a control is generated, the control conflict detection may include automatic classification 510, control time window generation 520, and automatic detection of whether a control conflict occurs 530. When no control conflict is detected, control execution is performed; when a control conflict is detected, the control conflict is resolved. The sixth diagram illustrates the composition of each unit in a control collision detector and operates in a sensing network environment 666, in accordance with an embodiment. The seventh figure illustrates the detailed operation of the control collision detector according to an embodiment.

Please refer to the sixth and seventh figures together. The control conflict detector 600 includes a Control Classification Unit 610, a Time Window Determination Unit 620, and a Control Determination Unit 630. When a new control 610a enters the control conflict detector 600, the control classification unit 610 classifies the characteristics of the sensor and the actuator, and obtains a type of sensor and actuator from the new control 610a (step 710). And a sensor type 612 and an actuator type 614 are supplied to the time window decision unit 620. The time window decision unit 620 determines a time window 622 for the new control 610a based on the type of sensor 612 and the type 614 of the actuator (step 720). When the actuator is identical to the at least one target actuator that is originally controlled, the control conflict determination unit 630 checks whether the time window 622 of the new control 610a overlaps with the at least one time window 630a of the at least one original control (step 730). ).

When the time window 622 of the new control 610a overlaps with the at least one original control time window 630a, the control conflict decision unit 630 determines that the new control 610a has a control conflict (step 740), and notifies a control conflict resolution module 640. Conduct a control conflict resolution procedure. When the time window 622 and the time window 630a do not overlap, the control conflict decision unit 630 determines that the new control 610a has no control conflict (step 750), notifies a Control Execution Module 650 to perform the action of the new control 610a until Its time window 622 disappears.

According to the embodiment of the present disclosure, in the sensing network environment 666 of the sixth figure, when the state detected by the sensor 690 is established, a control generation module 670 can be notified to generate a new control 610a. When the time window 622 of the new control 610a does not overlap with the currently active control time window 630a, the control execution module can notify the actuator 692 in a connected relationship with the new control 610a to perform the action of the new control 610a. The control conflict determination unit 630 can obtain at least one original control time window 630a from a control time window list 680 through a control management module 660, or can pass through the control management module 660. The time window 622 of the new control 610a is stored in the control time window form 680.

In the above, the control conflict detection technology used in the sensing network of the embodiment of the present disclosure determines that a control conflict occurs when the action time (time window) of the two control of the same target actuator overlaps, and is a The controlled event (sensor type) and action (actuator type) determine the type of time window for this control. Therefore, the control conflict detection mechanism of the disclosed embodiment can automatically generate a time window of the control according to a controlled event and action; and can also effectively know when the control occurs. This control collision detection technique can automatically detect a control conflict when a conflict occurs.

When a mechanism detects that a new control conflicts with the currently active control, this mechanism needs to control the conflict resolution technique to resolve the conflict. As mentioned before, the management mechanism for controlling conflicts of the disclosed embodiment includes a part for controlling conflict resolution in addition to the part for controlling collision detection. And the control conflict resolution system gives different attributes according to different control, and also judges the priority order of the control attributes through different modes. By controlling the priority of the attributes, the two controls are arbitrated to achieve the resolution of the control conflict. In other words, the part of this control conflict resolution includes control attribute generation, priority rule repository, and control arbitration.

Embodiments of the present disclosure assign each control a corresponding control attribute that can indicate characteristics of the control, such as artificial, safety, security, comfort (comfortableness) ), etc., combined with the binary distinction method to determine the original control or the new control to stay, in order to achieve the control conflict resolution. The binary distinction can dynamically change the order through patterns to conform to various usage scenarios.

According to an embodiment of the present disclosure, when a control conflict occurs, the control attributes of the two controls are compared to determine the priority of the control. Embodiments of the present disclosure impart control attributes in accordance with the type of sensor (event) and actuator (action) that are controlled. A controlled control attribute describes the application category of the control. Control attributes are mainly divided into three types: application domain, user information, and location information. Types of usage scenarios include, for example, a security domain, a safety domain, and a comfortable domain, and an energy saving domain. Types of operator information include, for example, artificial and non-artificial. There are two types of location information, including inside and outside.

When a control's control attribute belongs to the "use context", the event and action of this control must be the same situation. For example, the smoke detector detects smoke (safety) and turns the alarm on (safe). The events and actions of this control are all security types. Therefore, the control attribute of this control is "safe type". Therefore, when the control attribute of the control belongs to the "use context", it can be generated by a combination of the sensor (event) type and the actuator (action) type. When a control attribute belongs to "operator information", the control attribute "operator information" can be determined by the type of sensor (event). For example, if the remote control presses the button (man-made), the socket is turned on (non-human), and the event of this control is artificial, so the control attribute of this control is "man-made". When a control attribute belongs to "Location Information", if the event and action of this control are in the same area, the control attribute of this control is "in-area"; otherwise, it is "out-of-area". For example, the switch state is ON (living room), and the light (living room) is turned on, wherein the event and action of the control are in the same living room, so the control property of the control is "in-area".

From the above, it can be known that the source of a controlled control attribute is the sensor (event) and actuator (action) controlled for this purpose. In order to automatically assign control properties to each control, the disclosed embodiments are described with a variety of sensors and actuators, the description of which is a description of the properties of the sensors and actuators, including usage context, operator information, and Location information. 8A and 8B are respectively some examples of the description of the attributes of the sensor and the actuator according to the embodiment of the present disclosure. In the eighth diagram, for example, the attributes of the On/Off event corresponding to the switch are described as: use context (comfort), operator information (man-made), and location information (living room). In the eighth diagram B, for example, the attributes of the Turn on/off action corresponding to the Light are described as: use context (comfort), operator information (man-made), and location information (living room) .

Therefore, knowing the type of sensor and the type of actuator that is controlled, and referring to the attribute descriptions of the sensors and actuators provided in advance, the control attributes of this control can be known. The ninth figure is an example of a control attribute for automatically giving control according to an embodiment of the present disclosure. Referring to the ninth figure, the events of the switch are ON and OFF, respectively. The attribute descriptions of the opening and closing events are: use context (comfort), operator information (man-made), and location information (living room). The action of the light lamp has Turn on and Turn off actions respectively; the properties of the Turn on action are described as: use context (comfort), operator information (man-made), and location information (living room). The properties of the Turn off action are described as: use context (comfort), operator information (man-made), and location information (living room). Therefore, the control A can be combined: when the switch detects the open event, the trigger light performs a Turn on action; then, the disclosed embodiment is based on the attribute description of the detected On event, and is turned on (Turn On the attribute description of the action, the control attributes automatically assigned to control A are comfortableness, artificial, and inside.

Similarly, control B can be combined: when the switch detects off (Off), the triggering light performs a Turn off action, and then, the disclosed embodiment is based on the attribute description of the Off event, and is turned off (Turn The attribute description of the action of off) automatically assigns the control attributes of control B to comfort, man-made, and intra-area.

In different situations, the resolution of control conflicts requires different priority arbitrations. After obtaining the control attributes of the control, according to the disclosed embodiment, the priority of the control can be determined by comparing the control attributes. For example, the priority order of the control attributes is judged according to different modes. These modes are manual mode, safety mode, comfortable energy saving mode, outdoor mode, and the like. The manual mode has the highest priority with the "man-made" control attribute. The security mode has the highest priority with the control attribute of the "preservation type". The comfort and energy saving situation mode has the highest priority with the control attribute of the "comfort type". The outdoor mode has the highest priority with the control attribute of "outside the area".

The control conflict resolution mechanism of the present disclosure loads the corresponding priority rule description file according to the current mode, and determines the priority of the control and the solution of the same priority according to the priority rule in the description file to solve the control conflict. According to an embodiment of the present disclosure, the description file of the priority rule is as shown in the example of the tenth figure. In the example of the tenth figure, the description content of the profile of a priority rule may include a priority name, a first priority attribute, a second priority attribute, ..., a last priority attribute, and a worst case. The priority name describes the context in which the priority rule is used, such as manual mode, security mode, and the like. The control conflict arbiter that controls the conflict resolution mechanism describes these attributes in accordance with the order of precedence in the description file, the higher the attribute priority described earlier. Worst case is the solution when the two controls have the same priority, such as First In First Out (FIFO), or Last In First Out (LIFO), or both. It is decided by the way.

An example of the description of the priority rule description file generated by the three different modes is as follows in response to different usage scenarios. Manual mode: Man-made > Security > Security, and FIFO; Security mode: Security > Security > Human > Comfort, and FIFO; Comfort Situation Mode: Comfort > Security > Security, and FIFO. According to the disclosed embodiment, the control conflict resolution mechanism may determine the priority of the two controls according to the above rules. These priority rules can be dynamically increased or decreased depending on the usage context, and the same priority solution can also be dynamically adjusted based on the usage context.

According to an embodiment of the present disclosure, the priority comparison of the control is whether the control attribute exists in the sequential comparison control. By comparing the priorities of the two controls, the arbitration for controlling the conflict is explained below. First, comparing the highest priority control attribute, if the two controls only have the control attribute, the control is superior; if the two controls have or do not have the control attribute at the same time, then continue to compare the next priority control attribute, Sequentially compared to the end. If both parties have the same priority, they can be determined by first-in, first-out (FIFO), or last-in first-out (LIFO), or simultaneous. The result of the first in first out (FIFO) is that the new control wins; the result of the LIFO result is that the original control wins; at the same time, both parties fail and regain.

The control conflict arbitration mechanism disclosed in the present disclosure may have multiple conflict management rules to achieve control conflict resolution. Among them, each conflict management rule is based on the binary distinction method to judge the priority of control. Among the various conflict management rules, one rule distinguishes between "man-made" and "non-human", and the priority of "man-made" is greater than "non-human"; there is a rule that distinguishes between "safe" and "non-safe" and "safe" The priority is greater than "non-secure". There is a rule that distinguishes between "same area" and "different area", and the priority of "same area" is greater than "different area".

In view of the above, the 11th and 11th B diagrams are based on a manual mode priority rule description file, and the priority of the two controls is arbitrated by a binary method. Referring to Figure 11A, the description of the priority rule description file of this manual mode is: manual mode, man-made > security > zone, and first in, first out. In the eleventh B-picture, control A is the original control and control B is the new control. Therefore, when it is detected that the control B and the control A collide, according to the priority order of the manual mode, the binary distinction first compares whether the attributes of the control A and the control B are "artificial". When the attribute of the control A is "man-made" and the attribute of the control B is "non-human", it is determined that the control A is superior; when the attribute of the control A is "non-human" and the attribute of the control B is "man-made", then It is determined that control B is superior.

When the attributes of Control A and Control B are both "man-made" or "non-human", does this binary method compare the attributes of Control A and Control B to "safe"? Similarly, when only one control attribute is "safe" in control A and control B, then the control that determines that the attribute is "safe" is superior. When the attributes of Control A and Control B are both "safe" or "non-secure", does this binary method compare the attributes of Control A and Control B to "in-area"? Similarly, when only one control attribute in the control A and the control B is "in-area", it is determined that the attribute is "in-area" and the control is superior. When the control A and the control B have the same priority attribute, the new control B is determined to be superior according to the first in first out.

In the above description, the twelfth figure illustrates the composition of each unit in a Control Conflict Resolution Module and operates in a sensing network environment, according to an embodiment. A thirteenth diagram illustrates the detailed operation of the control conflict resolution module in accordance with an embodiment. Please refer to the twelfth and thirteenth drawings together. In the twelfth figure, the control conflict resolution module 640 includes a control attribute generation unit 1210, a priority rule base 1220, and a control conflict arbiter 1230. When the control conflict detector 600 notifies the control conflict resolution module 1200 that a new control conflict occurs, the control attribute generating unit 1210 generates the control attribute of the new control and supplies it to the control conflict arbiter 1230. For example, the control attribute generation unit 1210 can assign the control attribute of the new control by the sensor type and the actuator type thus newly controlled, as shown in step 1310 of the thirteenth diagram. The control conflict arbiter 1230 may retrieve the corresponding priority rule from the priority rule base 1220 according to a current mode, as shown in step 1320 of the thirteenth figure; and compare the new control with at least one according to the priority rule. The control attribute of the original control is as shown in step 1330 of the thirteenth figure; to arbitrate the new control and which of the at least one original control is the winner.

For example, the control conflict arbiter 1230 can check whether the new control has a higher priority, as shown in step 1340 of the thirteenth diagram; when the new control has a higher priority than the at least one original control, the new control is determined to be The winner is shown in step 1350 of the thirteenth diagram; and the control execution module 650 can be notified to perform the action of the new control. When the new control is lower than the priority of the at least one original control, it is determined that the at least one original control is the winner, as shown in step 1360 of the thirteenth figure. The control conflict arbiter 1230 can update the most recent valid control to a control element storage through the control management module 660.

Therefore, the control conflict resolution module of the disclosed embodiment can automatically give control attributes according to the sensors and actuators; can automatically give different prioritized arbitration according to different situations; and can automatically resolve control conflicts.

As described above, the mechanism for managing control conflicts in the sensing network of the present disclosure mainly includes controlling conflict detection and controlling conflict resolution. In view of the above, FIG. 14 is a diagram illustrating a system for managing control conflicts in a sensing network in accordance with an embodiment of the present disclosure. In the embodiment of the fourteenth figure, The system 1400 that manages control conflicts can include a control conflict detector 600 and a control conflict resolution module 640. The control conflict detector 600 determines the time window 622 of the new control 610a in accordance with the type of sensor taken from the new control 610a and the type of the actuator, and when the actuator is associated with at least one target of the original control When the actuators are the same and the time window 622 overlaps with the at least one time window of the at least one original control, it is determined that the conflict 1400b occurs and the control conflict resolution module 640 is notified. The control conflict resolution module 640 generates a control attribute of the new control 610a, and compares the new control with the control attribute of the at least one original control according to one of the plurality of different modes to arbitrate the new control and the at least one original control Which is the winner.

As described above, the control conflict resolution module may obtain a corresponding at least one priority rule from a priority rule base according to one of a plurality of different modes, and compare the new control with the at least one priority rule according to the at least one priority rule. The control attribute of an original control. When the new control has a lower priority than the at least one original control, the system 1400 that manages the control conflict can remove the new control 610a from a library of control elements by the control management module 660. When the new control has a higher priority than the at least one original control, the new control 610a can be written to the control element library and the control execution module 650 can be notified to perform the action of the new control 610a.

In the above, the fifteenth figure illustrates a method for managing control conflicts in a sensing network according to an embodiment of the present disclosure. In the fifteenth figure, the method of managing the control conflict determines a time window of the new control and assigns a control attribute according to the type of a sensor obtained from a new control and the type of the actuator (step 1510); and, when the actuator is identical to the at least one target actuator that is originally controlled, and the time window overlaps with the at least one time window of the at least one original control, it is determined that a conflict occurs (step 1520). When a conflict occurs, the method 1500 compares the priority of the control attribute with the at least one control attribute of the at least one original control according to at least one priority rule to arbitrate which of the at least one original control is the winner ( Step 1530).

In summary, the disclosed embodiments provide a time window based control conflict management machine system and method. In the aspect of controlling collision detection, different time windows generated according to different types of sensors (events) and actuators (actions) are controlled to improve the accuracy of control collision detection. In the aspect of controlling conflict resolution, a binary distinction of mode selection is revealed to resolve conflicts. In the sensing network, the disclosed embodiment can automatically detect conflicts; when a control conflict occurs, the correct arbitration can be automatically made according to the currently set mode.

The above is only the embodiment of the disclosure, and the scope of the disclosure is not limited thereto. That is, the equivalent changes and modifications made by the scope of the present invention should remain within the scope of the present invention.

100. . . Smart Energy Saving Office

110. . . Exclusive area

120. . . Public area

510. . . Automatic classification

520. . . Control time window generation

530. . . Automatically detect if control conflicts occur

600. . . Control conflict detector

610. . . Control classification unit

620. . . Time window decision unit

630. . . Control conflict decision unit

640. . . Control conflict resolution module

650. . . Control execution module

660. . . Control management module

670. . . Control generation module

666. . . Sensing network environment

680. . . Control time window form

610a. . . New control

612. . . Sensor type

614. . . Actuator type

622, 630a. . . Time Window

690. . . Sensor

692. . . Actuator

710. . . Get a type of sensor and actuator from the new control

720. . . Determine a time window for new control based on sensor type and actuator type

730. . . Check if the time window of the new control overlaps with the time window of at least one of the original controls

740. . . Determine that the new control has a control conflict

750. . . Determine that the new control has no control conflicts

1200. . . Control conflict resolution module

1210. . . Control attribute generation unit

1220. . . Priority rule base

1230. . . Control conflict arbiter

1310. . . Controlling properties for new controls with newly controlled sensor types and actuator types

1320. . . According to a current mode, the corresponding priority rule is obtained from the priority rule base

1330. . . Compare this new control with at least one of the original control's control properties

1340. . . Check if the new control has a higher priority

1350. . . Determine that the new control wins

1360. . . Determining that at least one original control wins

1400. . . Manage systems that control conflicts

1510. . . Determining a time window of the new control and assigning a control attribute according to the type of a sensor obtained from a new control and the type of the actuator

1520. . . Determining a conflict when the actuator is identical to at least one of the originally controlled at least one target actuators, and the time window overlaps with the at least one time window of the at least one original control

1530. . . When a conflict occurs, comparing the control attribute with the at least one control attribute of the at least one control attribute according to at least one priority rule to arbitrate the new control and the at least one of the original controls as the winner

The first figure is an example schematic diagram of an application scenario of a conflict resolution technique.

The second A is a schematic diagram of an example according to an embodiment of the present disclosure. When a slight time difference between the two controls occurs, a control conflict may also occur.

The second B is a schematic diagram of an example according to an embodiment of the present disclosure. When a large time difference occurs between the two controls, a control conflict may also occur.

The third figure illustrates four types of time windows in accordance with an embodiment of the present disclosure.

The fourth figure is a schematic diagram illustrating the temporal control of a plurality of controls in a sensing network system in accordance with an embodiment of the present disclosure.

The fifth figure illustrates the portion included in the control collision detection according to an embodiment of the present disclosure.

The sixth figure illustrates the composition of each unit in the control collision detector according to an embodiment of the disclosure, and operates in a sensing network environment.

The seventh figure illustrates the detailed operation of controlling the collision detector according to an embodiment of the present disclosure.

Eighth A is an illustration of an attribute description of a sensor in accordance with an embodiment of the present disclosure.

FIG. 8B is an illustration of an example of an attribute description of an actuator in accordance with an embodiment of the present disclosure.

The ninth figure is an example of a control attribute for automatically giving control according to an embodiment of the present disclosure.

The tenth figure is an example of a description file of a priority rule according to an embodiment of the present disclosure.

11A and 11B are diagrams for arbitrating the priority of two controls in a binary distinction according to a manual mode priority rule description file according to an embodiment of the present disclosure.

The twelfth figure illustrates the composition of each unit in a control conflict resolution module according to an embodiment of the present disclosure, and operates in a sensing network environment.

The thirteenth figure illustrates the detailed operation of the control conflict resolution module according to an embodiment of the present disclosure.

Figure 14 is a diagram illustrating a system for managing control conflicts in a sensing network in accordance with an embodiment of the present disclosure.

The fifteenth figure illustrates a method of managing control conflicts in a sensing network in accordance with an embodiment of the present disclosure.

1400. . . Manage systems that control conflicts

600. . . Control conflict detector

640. . . Control conflict resolution module

650. . . Control execution module

660. . . Control management module

610a. . . New control

690. . . Sensor

692. . . Actuator

670. . . Control generation module

Claims (24)

A system for sensing control conflicts in a network, comprising: a control conflict detector, determining a time of the new control according to a type of sensor obtained from a new control and a type of the actuator a window, and when the actuator is identical to at least one of the originally controlled at least one target actuators and the time window overlaps with the at least one time window of the at least one original control, determining that the new control has a control conflict; and Controlling the conflict resolution module, generating a control attribute of the new control, and comparing the control attribute with the at least one control attribute of the at least one original control to arbitrate which of the at least one original control is the winner. The system for managing conflicts of control according to claim 1, wherein the control conflict detector obtains the at least one time window of the at least one original control from a control time window form through a control management module. The system for managing conflicts of control according to claim 1, wherein the control conflict resolution module obtains at least one corresponding from a priority rule base according to one of a plurality of different modes via a control conflict arbiter a priority rule, and comparing the new control with the at least one control attribute of the at least one original control according to the at least one priority rule. A system for managing conflicts of control as described in claim 1, wherein each of the plurality of controls in the sensing network has a respective duration of time, the duration of time being defined as a time window of the control. A system for managing conflicts of control as described in claim 1, wherein the system utilizes a control attribute generation unit for each control of the plurality of controls in the sensing network, based on its sensor type and actuator Type, given a control attribute. A system for managing conflicts of control as described in claim 4, wherein the length of the time window of each of the plurality of controls is determined by the type of the sensor and the actuator being controlled. The system for managing conflicts of management as described in claim 3, wherein the at least one priority rule is dynamically increased or decreased according to different usage scenarios. A system for managing conflicts of control as described in claim 3, wherein when the new control has a lower priority than the at least one original control, the system removes the control from a control element library when the When the new control has a higher priority than the at least one original control, the new control is written into the control element library. The system for managing conflicts of control as described in claim 3, wherein the control conflict resolution module further comprises a control conflict arbiter having a plurality of conflict management rules. A method for sensing a control conflict in a network includes: determining a time window of the new control and assigning a control attribute according to a type of a sensor obtained from a new control and a type of the actuator; Determining that the new control has a control conflict when the actuator is identical to at least one of the originally controlled at least one target actuators, and the time window overlaps with the at least one time window of the at least one original control; and when a conflict occurs And comparing, according to the at least one priority rule, the priority of the new control and the control attribute of the at least one original control to arbitrate which of the at least one original control is the winner. The method for managing conflicts of management as described in claim 10, wherein the method assigns different control attributes according to different controls, and determines the priority order of the control attributes through different modes. The method for managing conflicts of management as described in claim 10, wherein the method has multiple conflict management rules to achieve control conflict resolution, and each conflict management rule of the plurality of conflict management rules is a type II The meta-differentiation method is used to judge the priority of control. For example, the method for managing conflicts of management described in claim 12, wherein one of the plurality of conflict management rules distinguishes between artificial and non-human, and the artificial priority is greater than non-human. There is a rule that distinguishes. The method for managing conflicts of management as described in claim 12, wherein one of the plurality of conflict management rules distinguishes between security and non-security, and the priority of security is greater than non-security. The method for managing conflicts of management as described in claim 12, wherein one of the plurality of conflict management rules distinguishes between the same area and the different area, and the priority of the same area is greater than the different area. The method for managing conflicts of management as described in claim 10, wherein the method loads a corresponding priority rule description file according to a current mode, and determines according to the at least one priority rule in the description file. Control prioritization and resolution of the same priority. A method for managing conflicts of control as described in claim 10, wherein the method divides the type of the sensor into a one-time trigger and a conditional trigger, and divides the type of the actuator into no specified execution time, And have a specified execution time. The method for managing conflicts of control as described in claim 17 wherein the one-time trigger means that the sensor can only detect a single state and trigger an event when the state is established. The method for managing control conflicts as described in claim 17 wherein the conditional triggering refers to a function of the sensor detecting more than two consecutive states, and triggering an event when at least one specified state is established. A method of managing a control conflict as described in claim 10, wherein the new control is the winner when the new control has a higher priority than the at least one original control. A method for managing conflicts of control as described in claim 17, wherein when the type of a controlled sensor is a one-time trigger and the type of the actuator is a specified execution time, the time window of the control The length is unlimited. A method for managing conflicts of control as described in claim 17, wherein when the type of the controlled sensor is a one-time trigger and the type of the actuator is a specified execution time, the time window of the control The length is the specified execution time. The method for managing conflicts of control as described in claim 17, wherein when the type of the controlled sensor is conditional triggering and the type of the actuator is no specified execution time, the time window of the control is The event condition of the control is generated when it is established, and disappears when the event condition of the control is not established. The method for managing conflicts of control as described in claim 17, wherein when the type of the controlled sensor is conditional triggering and the type of the actuator is a specified execution time, the time window of the control is The event condition of the control is generated when it is established, and disappears when the condition of the event is not established or the longest disappears after the specified execution time ends.