JP7207824B2 - Network control device, method and program - Google Patents

Description

The present invention relates to a network control device, method and program.

Ethernet (registered trademark) is used as a communication network to connect various devices such as ECUs (Electronic Control Units) and sensor devices in automobiles, vehicles, ships, aircraft, etc. ) are increasing. For in-vehicle Ethernet, for example, 1000BASE-T1 with a communication speed of 1 Gbps (Giga bit per second) and 100BASE-T1 with a communication speed of 100 Mbps (Mega bit per second) are also used.

Among mobile objects, automobiles tend to increase the amount of data in various sensor devices due to improvements in automatic driving technology and the like.

In addition, in order to prevent damage to passengers due to malfunction of mobile control due to delays or omissions of data flowing on the network, especially important data is multiplexed in the communication path ( redundancy), etc., to improve reliability.

In OSPF (Open Shortest Path First), which is one of routing protocols for IP (Internet Protocol) networks on Ethernet, the optimum route is determined using cost values set for ports of routers and layer 3 switches. Then, when a failure occurs in the optimum route (main route), it switches to the sub route. In addition, in Patent Document 1, in OSPF ECM (Open Shortest Path First Equal Cost Multi path) and MPLS TE (Multi Protocol Label Switching Traffic Engineering), there is a problem that it is difficult to control according to the traffic reception status of each terminal. The point is pointed out that even if an event occurs that causes traffic to be discarded on the main route due to a change in the network including multipath routes of the main route and the sub-route, the entire traffic can be handled using the traffic on the sub-route. A configuration is disclosed that enables rapid restoration and improves the reception performance of traffic at a receiving terminal.

RIP (Routing Information Protocol), which is one of IP network routing protocols, uses route information with a smaller metric (the number of hops) as the optimum route. In RIP, etc., the number of hops passes through a device (router, etc.) having a route selection (routing) function on the network layer (the third layer of seven layers in the OSI (Open System Interconnection) reference model). Refers to the number of times, and devices (repeaters, bridges, layer 2 switches, etc.) that perform transfers and relays at the level of the data link layer and physical layer (the 2nd and 1st layers of the 7 layers in the OSI reference model) , do not count.

As described above, when determining the route of an IP flow, if there are multiple routes from the source or network branch point to the destination node, the number of hops and the line of the link (circuit) connecting the nodes The best route is selected by quantifying the speed (communication speed between ports). Also, when determining redundant routes, after determining the main route, the number of hops and line speed are quantified in the same way as the main route, and the best route is selected as the secondary route. It will be done.

Japanese Patent No. 5418925

An analysis of the related art is given below.

As described in the aforementioned Patent Document 1, OSPF ECM statically generates multipath routes according to link costs. Also, in MPLS TE, multipath routes are dynamically generated according to the network topology, link cost, and traffic volume flowing through the links, but multipaths cannot be generated from nodes in the middle of the network. It is difficult to control according to the traffic reception status of each terminal.

Depending on the state of the mobile object, the environment in which the mobile object is placed, the operation mode of the mobile object (manual, automatic, semi-automatic operation), etc., the characteristics of the data and routes flowing on the network inside the mobile object such as the in-vehicle network may change. . Therefore, simply determining the route of the flow based on the number of hops and the line speed of the link between nodes does not always lead to the optimum route for the flow.

In addition, in Patent Document 1, according to the state of the mobile body (for example, presence or absence of movement, running state, environment in which the mobile body is placed, operation mode of the mobile body, operation content, etc.). There is no recognition of the problem of enabling dynamic routing of communication flows flowing through a network (and no specific example or explanation is given to the extent that a person skilled in the art can recognize that the problem can be solved. ).

Realization of dynamic route selection of communication flows flowing through a network within a mobile body according to the environment and state of the mobile body is desired (knowledge of the present inventors).

SUMMARY OF THE INVENTION The present invention has been devised based on the recognition of the above problems, and its object is to enable dynamic routing of communication flows in a network onboard a mobile depending on the state of the mobile. An object of the present invention is to provide a network control device, method, and program.

According to one aspect of the present invention, a storage unit for pre-storing communication flow information for each state of the mobile body with respect to a network mounted on the mobile body;
a communication flow selection unit that acquires the state of the mobile object, refers to the storage unit, and selects at least one communication flow corresponding to the acquired state;
A network control device is provided, comprising a route determination unit that determines at least one route from among a plurality of routes of the selected communication flow based on a cost calculation result of each route.

According to another aspect of the present invention, with respect to a network installed in a mobile body, communication flow information is stored in advance in a storage unit for each state of the mobile body,
obtaining the state of the moving object;
referring to the storage unit to select at least one communication flow corresponding to the acquired state;
A network control method is provided, which determines at least one route from among a plurality of routes of the selected communication flow based on a cost calculation result of each route.

According to yet another aspect of the present invention, a process of acquiring the state of a moving object;
a process of selecting at least one communication flow corresponding to the obtained state by referring to a storage unit pre-stored with information on communication flows for each state of the mobile body with respect to the network mounted on the mobile body;
A program for causing a computer to execute a process of determining at least one route based on a cost calculation result of each route from among the plurality of routes of the selected communication flow is provided.

According to still another aspect of the present invention, a computer-readable recording medium (for example, RAM (Random Access Memory), ROM (Read Only Memory), or EEPROM (Electrically Erasable and Programmable ROM)) storing the program and non-transitory computer readable recording mediums such as HDDs (Hard Disk Drives), CDs (Compact Discs), DVDs (Digital Versatile Discs).

According to the present invention, for example, it is possible to dynamically select a route of a communication flow flowing through a network within a mobile object according to the state of the mobile object.

FIG. 2 illustrates an exemplary embodiment of the invention; FIG. 2 illustrates an exemplary embodiment of the invention; FIG. 2 illustrates an exemplary embodiment of the invention; 1 is a diagram illustrating Example 1 of an exemplary embodiment of the present invention; FIG. FIG. 4 is a diagram illustrating data flow costs and path costs by state according to an exemplary embodiment of the present invention; FIG. 5 is a diagram illustrating main path calculation results of an exemplary embodiment of the present invention; FIG. 5 is a diagram illustrating secondary path calculation results of an exemplary embodiment of the present invention; FIG. 4 is a diagram illustrating Example 2 of the exemplary embodiment of the present invention; FIG. 4 is a diagram illustrating data flow costs and path costs by state according to an exemplary embodiment of the present invention; FIG. 5 is a diagram illustrating main path calculation results of an exemplary embodiment of the present invention; FIG. 5 is a diagram illustrating secondary path calculation results of an exemplary embodiment of the present invention; FIG. 2 illustrates an exemplary embodiment of the invention; FIG. 2 illustrates a computer implementation of an exemplary embodiment of the invention;

In the following, the basic form of the invention is first described, followed by exemplary embodiments. FIG. 12 is a diagram for explaining a network control method of one basic form of the present invention. This network control method may be executed, for example, by a network control device (controller (processor)) connected to a network mounted on a mobile object.

Regarding the network installed in the mobile object, the storage unit 11 stores communication flow information (communication flow information by condition) for each state of the mobile object.

The state of the moving body is detected (step S1).

Next, referring to the storage unit 11, at least one communication flow corresponding to the detected state is selected (step S2). At this time, if there are a plurality of communication flows corresponding to the detected state, the communication flows may be selected in a predetermined order according to the priority set in advance for the communication flows. Alternatively, a communication flow with a cost equal to or higher than a predetermined value may be selected according to a cost or the like set in advance for the communication flow.

Next, at least one route is determined from among the plurality of routes of the selected communication flow based on the cost calculation result of each route (step S3).

In step S1, a change in the state of the mobile object is detected, and in steps S2 and S3, in response to the detection of the state change, at least one communication flow corresponding to the changed state is selected, and the path of the communication flow is selected. may be determined dynamically.

In the storage unit 11, as the state-specific communication flow information, cost information of the communication flow may be set for each state of the mobile body.

For the cost set by state for the communication flow, the same value is
In step S2, it is used as a priority when selecting a communication flow,
In step S3, the cost may be used for cost calculation of each route of the communication flow selected corresponding to the state.

That is, in step S2, the communication flow cost information may be treated as the communication flow priority, and if there are a plurality of communication flows corresponding to the acquired state, the communication flow may be selected in descending order of priority. . For example, the higher the cost set for a communication flow, the higher the priority of that communication flow. In this case, in step S3, the cost information (priority) set corresponding to the state of the selected communication flow is included in the calculation of the path cost of the selected communication flow. good too.

In step S3, the cost of the communication flow whose route has been determined may be added to the cost of the route including the determined route in another communication flow corresponding to the state.

In step S3, a main path may be determined for the selected communication flow, and then a secondary path (redundant path) of the communication flow may be determined.

In step S3, in determining the path of the communication flow corresponding to the state, the cost of the communication flow itself, the number of relay nodes (the number of hops) through which the path of the communication flow passes, the line cost between relay nodes, and , if the path includes an already determined path of another communication flow corresponding to the state, then calculating the cost of the path of the other communication flow based on the cost of the other communication flow, and summing A path with the lowest cost may be determined as the optimum path for the communication flow.

For example, in determining the main route of the communication flow corresponding to the state, the cost of the communication flow itself, the number of relay nodes through which the main route of the communication flow passes, the circuit cost between relay nodes, and the main route includes an already determined main route of another communication flow corresponding to the state, the cost of the main route of the communication flow is calculated based on the cost of the other communication flow. good too.

Further, in determining the secondary route of the communication flow corresponding to the state, the cost of the communication flow itself, the number of relay nodes through which the route of the communication flow passes, the line cost between relay nodes, and the secondary route are , if it includes already determined routes (main route, secondary route) of the one or more communication flows corresponding to the state, the total cost of the one or more communication flows, based on the communication flow may be configured to calculate the cost of the sub-path of .

In step S1, for example,
・Presence or absence of movement of the moving body,
・Surrounding environment of the moving body,
・Movement operation mode,
・Details of operation,
You may make it determine the state of a mobile body based on any one or some combination of.

The cost calculation for each route of the communication flow is completed in advance, and in step S3, when the state of the mobile body changes, the route calculation result for which the cost calculation has been completed is referred to for the communication flow corresponding to the state. to determine the route. In this case, for example, the cost calculation of each route of the communication flow is completed in advance in a server, cloud, etc., and when the state of the mobile body changes, the mobile body accesses the server, cloud, etc. A route calculation result for which calculation has been completed may be acquired to determine the route of the communication flow corresponding to the state.

The communication flow cost information stored in the storage unit may be variably set according to the state of the mobile unit.

In steps S2 and S3, when selecting the communication flow to be used for the sub-path, the sub-path may be determined from the communication flows whose cost is equal to or more than a predetermined threshold or less than or equal to a predetermined threshold.

If the detected state corresponds to a predetermined situation such as a malfunction of the mobile unit, communication of communication flows whose priority is equal to or lower than a predetermined threshold may be restricted.

An exemplary embodiment of the invention will now be described with reference to FIG. Note that FIG. 1 exemplifies a mobile object as an automobile (private car) and Ethernet (in-vehicle Ethernet) as an example of a network mounted on the mobile object, but the mobile object is limited to an automobile. Of course, it may also be a railway vehicle, a ship, an aircraft, or the like.

Referring to FIG. 1, sensor 1 (40A) is a front monitoring camera, sensor 2 (40B) is a left front monitoring camera, sensor 3 (40C) is a rear monitoring camera, and sensor 4 (40D) is a left rear monitoring camera. The relay node 1 (30A), the relay node 2 (30B), and the relay node 3 (30C) are composed of layer 2 equivalent switches or layer 3 equivalent switches (routers) connected to the Ethernet. In the following, for the sake of simplicity of explanation, the relay node will be explained as a layer 2 (data link layer) switch.

Each relay node 1, 2, 3 determines the destination of the frame received from the port based on the MAC (Media Access Control) address in the data link layer, and transfers the frame to the corresponding port.

Sensors 1 and 2 (40A and 40B) and relay nodes 2 and 3 (30B and 30C) are connected to the relay node 1 (30A). The ECU 20 and relay nodes 1 and 3 (30A and 30C) are connected to the relay node 2 (30B). Sensors 3 and 4 (40C and 40D) and relay nodes 1 and 2 (30A and 30B) are connected to the relay node 3 (30C).

Ethernet link 1 (line) between relay nodes 1 and 2 (30A and 30B) has a communication speed of 1 Gbps, Ethernet link 2 and relay nodes 2 and 3 (30B and 30C) between relay nodes 1 and 3 (30A and 30C). The communication speed of the Ethernet link 3 between them is 100 Mbps. Although one ECU is shown in FIG. 1 simply for the sake of simplification of the drawing, there are cases in which 100 or more ECUs are installed in one vehicle depending on the type of vehicle.

The network control device 10 stores communication flow costs in a storage device (not shown) for each state in association with the state of the mobile object (for example, a state requiring switching of the communication flow or the like). The cost of a communication flow between a source and a destination is set based on the communication speed of each link included in the communication flow and the number of relay nodes (hop count) through which the communication flow passes. In the following description, even when the relay node is a layer 2 switch, the number of relay nodes passed through is referred to as the number of hops.

The network control device 10 acquires information from various sensors, an ECU, a GPS (Global Positioning System) receiver, or the like. The network control device 10 derives the current state of the vehicle (mobile body) based on information from various sensors, an ECU, a GPS receiver, or the like.

The network control device 10 performs communication of each link included in the communication flow corresponding to the state of the vehicle among the communication flows between the source (for example, sensor) and the destination (for example, ECU) in the in-vehicle network. The cost is calculated based on the speed and the number of relay nodes (the number of hops) passed through, and the route with the lowest total cost is selected as the optimum route.

For example, in FIG. 1, when the communication flow corresponding to the state of the vehicle after the change corresponds to the communication flow between the sensor 1 and the ECU 20, the communication flow includes:
Route 1: sensor 1-relay node 1-relay node 2-ECU 20 (link 1 communication speed 1 Gbps, number of hops=2),
Route 2: includes sensor 1-relay node 1-relay node 3-relay node 2-ECU 20 (communication speed of links 2 and 3: 100 Mbps, number of hops=3).

The network control device 10 selects the route 1 from among the routes 1 and 2 as a result of cost calculation based on the communication speed (line speed) of the link between the relay nodes and the number of relay nodes (hop count).

When the network control device 10 receives an Ethernet frame from the sensor 1 (40A) to the relay nodes 1 and 2 (30A and 30B) on the path 1 of the communication flow between the sensor 1 (40A) and the ECU 20, Each is routed to a 1 Gbps Ethernet interface (a port of a network interface card (NIC)).

Although not particularly limited, the route setting information set in each relay node from the network control device 10 may be configured to include match conditions and actions, for example, like OpenFlow flow entries. For example, the matching condition of the route setting information set in the relay node 1 is source MAC address: the MAC address of the sensor 1 (40A) and destination MAC address: the MAC address of the ECU 20. FIG. The action when this match condition is met is transfer to the port connected to link 1 (1 Gbps Ethernet). Alternatively, the route setting information set in each relay node may conform to a MAC address table that defines the correspondence between MAC addresses and ports (in this case, the relay node automatically uses a layer 2 switch to MAC address learning is not performed). If each relay node is a layer 3 switch that determines the destination by referring to network routing information (routing table), the match condition is, for example, a source IP address and a destination IP address, and an action when the match condition is matched. is the operation of transferring the received packet to the interface (port) connected to the network corresponding to the destination IP address (in this case, the relay node (layer 3 switch) where the route setting information is set from the network control device 10 , regarding the in-vehicle network, it is not necessary to update the routing table autonomously in the layer 3 switch).

FIG. 2 is a diagram schematically illustrating an example of the configuration of the network control device 10. As shown in FIG. As will be described later, the network control device 10 is composed of a processor connected to a memory (which stores programs), and may execute the following processes, for example.

The monitor unit 101 includes an ECU, an ECU connected to an in-vehicle sensor, a steering ECU, a brake ECU, a GPS (Global Positioning System) receiver, a predetermined voltage between terminals (power supply voltage), power supply current, ambient temperature, acceleration , information such as the state (whether or not the vehicle is running and the running state), steering operation, environmental data, current position, etc., may be obtained from a gyro sensor or the like.

Based on the information acquired by the monitor unit 101, the state detection unit 102 detects, for example, a state that triggers route determination. A memory (table) stores in advance the correspondence between a state that triggers route determination (for example, a change in state from a steering angle from a steering sensor to a state change from going straight to turning left, etc.) and information acquired by the monitor unit 101. The state detection unit 102 may search the information acquired by the monitor unit 101 and the memory (table) to detect a state (state identification number) that triggers route determination. The state detection unit 102 detects whether the vehicle is moving,
The state of the vehicle may be determined based on one or more of the surrounding environment of the vehicle, the operation mode (manual, automatic, semi-automatic), and the details of the operation. Alternatively, the state detection unit 102 may determine the state based on (combination of) information acquired by the monitor unit 101, based on a supervised learning model (classification model) or the like.

Communication flow selection unit 103 determines the priority of data from sensors or the like related to the state detected by state detection unit 102 (for example, state change to left turn). By referring to the memory (storage device) 106 that stores the communication flow cost (also referred to as the priority when selecting a communication flow), the detected state (if a change in state is detected, the state). If there are a plurality of communication flows corresponding to the detected state, the communication flow may be selected in descending order of the cost (priority) value set for the communication flow.

For example, when the detected state is left-turn driving (state change from straight driving to left-turn driving), from sensor 2 (left front monitoring sensor) (40B) and sensor 4 (left rear monitoring sensor) (40D) in FIG. data has the highest priority. Therefore, a communication flow having endpoints of sensor 2 (left front monitoring sensor) (40B) and ECU 20 and sensor 4 (left rear monitoring sensor) (40D) and ECU 20 is selected.

The route determining unit 104 refers to the network configuration information, the route cost information, etc. stored in the memory 107 for each communication flow corresponding to the state, and determines the cost of the communication flow, the cost corresponding to the number of relay nodes to pass through, The total cost is calculated based on the cost corresponding to the communication speed of the links (circuits between relay nodes) included in each route, and the route with the lowest total cost is determined as the optimum route. The network configuration information includes, for example, connection information (network topology) of the network (Ethernet) shown in FIG. You can also try to

The route setting unit 105 sets route setting information corresponding to the route determined by the route determining unit 104 to each relay node on the corresponding route.

After that, each relay node switches the received frame based on the route setting information set by the network control device 10 . That is, each relay node transfers the frame received from the port to the transfer destination port set in the route setting information.

The network control device 10 in FIG. 2 may be implemented as an SDN (Software Defined Network) controller. However, when changing the flow entry in the OpenFlow protocol, which is one of SDN, there is no flow entry that matches the packet header information in the OpenFlow switch that received the packet, and the packet (first packet) is sent to the OpenFlow controller , and the OpenFlow controller generates a flow entry based on the flow calculation result and sets it in the switch with a Flow Modify message (a flow entry is generated for each new flow). On the other hand, in the present embodiment, the network control device 10 determines the route of the communication flow and sets the route setting information in the relay node, for example, using the state (state change) of the vehicle as a trigger.

For example, as shown in FIG. 3, the network control device 10 of the vehicle 1 is connected to a server 241 of a cloud (data center) 240 via a base station 200, a core network 210, and a WAN (Wide Area Network) 230 such as the Internet. The server 241 may variably set the cost (priority) information of the communication flow for each state. Alternatively, the network control device 10 of the vehicle 1 is connected to a server 241 of a cloud (data center) 240 via a WLAN (Wireless Local Area Network) access point (AP) 221 via a WLAN 220 via a WAN 230. good too. The server 241 of the cloud (data center) 240 may be configured to store network configuration information of the vehicle 1 and the like. Further, on the server 241 side, the cost calculation of the communication flow route for the network of the vehicle 1 is completed in advance, and when the state of the vehicle 1 changes, the route determination unit 104 of the network control device 10 of the vehicle 1 can calculate the server 241, route information for which cost calculation has been completed may be obtained to determine the route with the lowest cost.

Note that the route determined by the above procedure may be set as the main route, and the redundant route may be similarly determined as described in the following embodiments.

According to the embodiment of the present invention, the state of a vehicle (moving body) is quantified for each communication flow (data flow) and used as a parameter for determining communication flow paths and redundancy in an in-vehicle network. When the state of the vehicle, the environment in which it is placed, or the content of the input operation changes, the priority of the data is determined according to the state, environment, or operation content, and the communication route and redundant route are determined using the priority. decide. If the vehicle's condition or the environment in which it is placed is judged to be urgent, low-priority data will be temporarily blocked, and performance limits such as the upper limit bandwidth of high-priority data and relay nodes that transmit and receive such data will be restricted. It may be loosened.

In the following, an example of communication flow selection and route change in an in-vehicle network that accompanies a change in the state of the vehicle from forward running to left-turning will be described with reference to FIG. 4 and the like. FIG. 4 is a diagram schematically exemplifying an example of an in-vehicle Ethernet network when the vehicle is running straight (referred to as "state 1") superimposed on the vehicle. Note that the network control device 10 of FIG. 1 is not shown in FIG. 4 for simplification. Moreover, reference numerals are not given to each element. Note that the network control device 10 may be provided in any ECU.

As described above, the paths used in the communication flow are determined based on the cost of each path. The cost is calculated based on the communication speed (line speed) of the links on the route, the number of relay nodes that pass through, and the like. A cost defined for each communication flow is added to a line used as a route for a certain communication flow, and is referred to when calculating the route for the next communication flow.

In FIG. 4, sensors x (x=1 to 8) are transmission nodes that transmit data, and control 1 (ECU1) is a data destination node.

A combination of sensor x, which is a source node, and control 1, which is a destination node, is assumed to be a communication flow x. A communication flow is also called a “data flow”.

Each communication flow has its cost defined in advance according to the state of the vehicle (1 to N).

FIG. 5(A) is a table showing an example of a combination of communication flows by transmission/destination nodes and costs in each state. In communication flow 1 (sensor 1->control ECU 1), the cost in state 1 is 20 and the cost in state 2 is 10. In the example of FIG. 5A, in communication flow 2 (sensor 2->control ECU 1), the cost in state 1 is 10, and the cost in state 2 is 20.

The relay nodes 1 to 5 relay the transmission data to the destination node by transferring the transmission data to the next stage.

A cost is added each time a relay node is passed through.

There are multiple routes (via relay nodes) for the communication flow x. In the example of FIG. 4, when a relay node i (i=1 to 5) is represented by an integer i, there are the following seven routes as routes of communication flow 1 (see FIG. 6).

1→5,
1→2→5,
1→3→5,
1→3→4→5,
1→3→4→2→5,
1→2→4→5,
1→2→4→3→5

The relay nodes are connected by an Ethernet link (circuit) with a communication speed of 1 Gbps or 100 Mbps.

One cost is added when passing through a 1 Gbps Ethernet link (circuit).

A cost of 10 is added when passing through a 100 Mbps Ethernet link (line).

FIG. 5B shows an example of route costs. The cost of the line (1 Gbps) between relay nodes 1→2 is 1, the cost of the line (100 Mbps) between relay nodes 1→3 is 10, and the cost of the line (1 Gbps) between relay nodes 1→5 is 1.

The state of the mobile object is uniquely determined from the details of the operation input to the mobile object, sensor information provided by the mobile object, the operation mode of the mobile object, the position/environment where the mobile object exists, and the like.

An example of the operation of the network control device 10 of the embodiment will be described with reference to FIG. 4 and the like. In the following, when it is obvious that the network control device 10 is the main actor, the main actor will be omitted as appropriate.

Identify the current mobile state and view the cost of each communication flow. Route calculation is performed starting from the communication flow with the highest cost (priority).

State 1 in FIG. 4: when the vehicle is traveling forward will be described below as an example.

First, determine the main route of each communication flow. FIG. 6 is a diagram for explaining calculation of the main route.

From the state-specific communication flow costs in FIG. 5A, first, among the communication flows corresponding to state 1, the main route of communication flow 1 (cost/priority: 20) with the highest cost (priority) is calculated. . Based on the number of hops, the line speed, and the cost of the communication flow 1 itself (the cost (priority) in FIG. 5A), the route with the lowest cost among the routes that the communication flow 1 can take: "1 →5” is adopted as the main route of communication flow 1 (“1→5” surrounded by a dashed line in communication flow 1 in FIG. 6).

The cost (total cost) of the path of communication flow 1 in FIG. 6: 1→5 is obtained by multiplying the cost of communication flow 1 itself (=20) by the number of hops (=1), and multiplying this by the total path cost ( = 1) is added (= 21). Path of communication flow 1: 1→2→5 cost (total cost) is the result (=20×2) of multiplying the cost of communication flow 1 itself (=20) by the number of hops (=2). , a value (=42) obtained by adding the route total cost (=2).

At this point (when determining the main route of communication flow 1: 1→5), there is no selected communication flow (a communication flow whose main route has been decided). cost” (value is 0), equivalently not added to the total cost of each path of communication flow 1.

When route 1→5 is determined as the main route of communication flow 1, the cost of communication flow 1: 20 is added to each line of route 1→5 (relay node 1→5). That is, in FIG. 6, for example, 20 is set in the column of "selected communication flow cost" for the route of communication flow 2: 1→5. Therefore, the cost of the route of communication flow 2: 1→5 is the cost of communication flow 2 itself (=10)×(number of hops (=1))+total route cost (=1)+selected communication flow cost ( =20)=31. Also, the cost of communication flow 1: 20 is set in the "selected communication flow cost" of the route of communication flow 3: 2→1→5. Since route 1→5 is shared with communication flow 1 and communication flow 2 (or 3), the link (circuit) of route 1→5 (between relay node 1→5) It can be said that it reflects the decrease in the bandwidth per flow (a decrease in the bandwidth of a link corresponds to an increase in the cost of the link). Therefore, the total cost of the route of communication flow 3: 2→1→5 is the cost of communication flow 3 itself (=20)×(number of hops (=2))+total route cost (=2)+selected communication Flow cost (=20)=62.

A path that minimizes the total cost as an objective function is determined as the optimal path. The cost (priority) of the selected communication flow added to the total cost can correspond to the penalty function (coefficient) of the route optimization objective function (total cost).

Next, the main route of communication flow 3 (cost/priority: 20) is calculated. From the number of hops, the line speed, the cost of the own communication flow, and the cost of the selected communication flow, the route with the lowest cost among the possible routes for the communication flow 3: 2 → 5 is the main route of the communication flow 3. adopted as

The cost 20 of communication flow 3 is added to each line of route 2→5 (line between relay nodes 2→5). In FIG. 6, 20 is added to the "selected communication flow cost" column for the path of communication flow 4: 2→5. After that, the main routes of the remaining communication flows are similarly determined.

Next, a sub-path (redundant path) for each communication flow is determined. FIG. 7 is a diagram for explaining the calculation of subpaths. In FIG. 7, the route of communication flows 1 and 2: 1→5, the route of communication flows 3 and 4: 2→5, etc. are determined as main routes for communication flows 1, 2, 3, 4, etc. The route column is shown with a solid gray background.

The criteria for communication flows that use secondary routes are:
・If the cost (priority) of the communication flow itself is higher than a certain threshold α (predetermined value), or
- The calculated total cost of the route is lower than a certain threshold value β (predetermined value).

Although not particularly limited, in the present embodiment, the threshold α is set to 20, and communication flows having a cost (priority) of 20 or higher are targeted.

As a communication flow whose cost (priority) exceeds the threshold α (=20), first, the secondary route of the communication flow 1 is calculated.

Based on the number of relay nodes to pass through, the line speed between relay nodes, the cost of the communication flow itself, and the cost of the selected communication flow, the route with the lowest total cost: 1→3→5 is the sub-route of communication flow 1. adopt. In FIG. 7, the "selected communication flow cost" of the route: 1→2→5 of the communication flow 1 is the cost of the communication flow 3 itself in which the route: 2→5 is selected as the main route: 30 and the communication flow 4 Own cost: The sum of 10: 20 + 10 = 30, and the total cost of the route is the communication flow 1 own cost (= 20) x (number of hops (= 2)) + total route cost (= 2) + selected communication flow cost (=30)=72. In addition, the "selected communication flow cost" of the route selected as the main route in communication flow 1: 1→5 is the sum of the cost of communication flow 1 itself: 20 and the cost of communication flow 2 itself: 10: 20 + 10. =30. In FIG. 7, no corresponding example is shown, but when calculating the sub-path, the "selected communication flow cost" includes the cost of the communication flow itself for which the path including the sub-path has already been determined as the sub-path. Additional costs are added.

Next, the secondary route of the communication flow 3 whose cost (priority) of the communication flow exceeds the threshold α (=20) is calculated.

Based on the number of relay nodes to pass through, the line speed between relay nodes, the cost of the own communication flow, and the "selected communication flow cost", the route with the lowest cost among the possible routes for the communication flow 3: 2→4→5 is adopted as the secondary route of the communication flow 3. The cost of the communication flow 3 itself: 20 is added to each line of the route 2→4→5.

With the above procedure, the route of each communication flow in state 1: forward running is determined. In FIG. 4, the calculation results of the main path and the sub-circuit in state 1 are indicated by thick solid lines and broken lines.

Next, the case where the state of the moving object changes from state 1 in FIG. 4 to state 2 in FIG. 8: left turn running will be described. The network control device 10 determines the main route and sub route of the communication flow in state 2 in the same manner as when the vehicle is traveling forward described with reference to FIGS. 5 to 7 . In the following, if it is obvious that the network control device 10 is the main actor, the main actor will be omitted.

When the vehicle is traveling forward, it is desirable that the sensors 1 and 3 provided in the front of the vehicle body send data with high priority from the viewpoint of course confirmation and collision prevention. Sensors 2 and 4 that are present have a high priority. The sensors 5, 6, 7, and 8 provided at the rear of the vehicle body need to transmit data for rear monitoring, but they do not require the same amount of data and reliability as the front sensors.

However, when the vehicle is turning left, it is desirable that the sensors 2 and 6 provided on the left side of the moving body transmit data with high priority from the viewpoints of checking the course on the left side and preventing collisions/entrainment. At the next point, the sensors 1 and 5 provided on the front and rear left sides of the vehicle body have high priority. On the other hand, the sensors 3, 4, 7, and 8 provided on the right side of the vehicle body need to transmit data for monitoring the right side, but the amount of data and reliability are not required as much as the sensors on the left side.

First, the main route of the communication flow 2 between the sensor 2 and the ECU in FIG. 8 is determined. The main route of communication flow 2 (cost/priority: 20) (see FIG. 9A) with high priority in state 2 is calculated. Note that FIG. 9B shows the route cost, which is the same as FIG. 5B.

FIG. 10 is a diagram for explaining calculation of the main route. Based on the number of relay nodes (the number of hops) passed through, the line speed, and the cost of the communication flow 2 itself, among the routes of the communication flow 2, the route with the lowest cost among the routes that the communication flow 2 can take: 1 → 5 is adopted as the main route of the communication flow 2. At this point, since there is already selected communication flow 1, the cost for each communication flow is added to the column of "selected communication flow cost". A cost of 20 for communication flow 1 is added to each line of route: 1→5. After that, the main routes of the remaining communication flows are similarly determined.

Next, the secondary route of communication flow 2 (cost/priority: 20) is calculated. FIG. 11 is a diagram for explaining calculation of the main route. 11, communication flows 1 and 2: 1→5, communication flow 5: 2→5, communication flow 6: 3→5, etc., communication flows 1, 2, 5, 6, etc. The background of the column of the route determined as each main route is filled with gray. From the number of hops, the line speed, the cost of the own communication flow, and the cost of the selected communication flow, the route with the lowest cost among the possible routes of the communication flow 2: 1→3→5 Adopted as a secondary route. Cost 20 of communication flow 2 is added to each line of route 1→5. Sub-routes for the remaining communication flows are determined in the same manner thereafter. In FIG. 8, the main path calculation result in state 2 is indicated by a thick solid line.

In this way, the priority of the communication flow changes according to the state of the mobile object (positioned state, operation details, etc.). According to this embodiment, the priority (cost) of each communication flow in each state is defined. It is possible to flexibly cope with changes in communication required for control. As a result, control reliability can be improved.

4 and 8 is a connected vehicle that manages an in-vehicle network and connects the vehicle and the cloud (240 in FIG. 3) via a mobile network equipped with a wireless communication function, a WAN (Wide Area Network), etc. A gateway is fine. In this case, the state-specific cost (priority) of each communication flow may be changed remotely via the cloud.

Alternatively, on the cloud (240 in FIG. 3) side, all (or part) of the route calculation for each state of the mobile unit 1 is completed, and when the state of the mobile unit 1 changes, the cloud (240 in FIG. 3 of 240) may refer to the completed route calculation to determine the route corresponding to the changed state.

In the mobile object 1 (vehicle) shown in FIGS. 4 and 8, when an emergency occurs (for example, there is a risk of collision, or the mobile object breaks down), a communication flow with a priority lower than a predetermined threshold value γ may be stopped, and resources may be concentrated on communication of communication flows having a priority higher than the threshold γ.

In this case, for example, for a communication flow with a priority higher than a predetermined threshold γ, for example,
・Utilize the surplus bandwidth and power generated by stopping low-priority communication flows to increase the network bandwidth.
・Increase the clock frequency of the transmission and reception nodes of the communication flow to improve the accuracy of the control interval.
・Increase data transfer bandwidth by using additional vacant routes and increasing redundant routes to improve transfer frequency and reliability.
You may make it control for avoiding an emergency by performing at least any one of.

FIG. 13 is a diagram for explaining an example in which the network control device 10 is configured by a computer 110. As shown in FIG. Processor (CPU (Central Processing Unit), data processing device) 111, semiconductor memory (for example, RAM (Random Access Memory), ROM (Read Only Memory), EEPROM (Electrically Erasable and Programmable ROM), etc.), HDD (Hard (Disk Drive), CD (Compact Disc), DVD (Digital Versatile Disc), etc., and a network interface card (NIC) 113 . A program that implements the functions of the network control device 10 described in the above embodiment is stored in the memory 112, and the processor 111 reads out and executes the program, for example, the monitor unit 101 in FIG. Each process of the detection unit 102, the communication flow selection unit 103, the route determination unit 104, and the route setting unit 105 may be executed.

The disclosure of Patent Document 1 above is incorporated herein by reference. Within the framework of the full disclosure (including claims) of the present invention, modifications and adjustments of the embodiments and examples are possible based on the basic technical idea thereof. Also, various combinations and selections of various disclosure elements (including each element of each claim, each element of each embodiment, each element of each drawing, etc.) are possible within the scope of the claims of the present invention. . That is, the present invention naturally includes various variations and modifications that can be made by those skilled in the art according to the entire disclosure including claims and technical ideas.

1 vehicle (moving object)
10 network control device 11 storage unit 20 ECU
30A to 30C relay node 1 to relay node 3
40A to 40D Sensor 1 to Sensor 4
101 monitor
102 State detection unit 103 Communication flow selection unit 104 Route determination unit 105 Route setting units 106, 107 Memory (storage device)
110 computer 111 processor 112 memory 113 network interface card (NIC)
200 base station 210 core network 220 WLAN
221 WLAN access point 230 WAN
240 cloud (data center)
241 server

Claims (10)

a storage unit for pre-storing communication flow information for each state of the mobile body with respect to a network mounted on the mobile body;
a state detection unit that detects the state of the moving object;
a communication flow selection unit that refers to the storage unit and selects at least one communication flow corresponding to the state of the mobile body;
a route determination unit that determines at least one route from among a plurality of routes of the selected communication flow based on a cost calculation result of each route;
A network control device comprising: 2. The route determination unit according to claim 1, wherein, in response to detection of a change in the state of said mobile unit by said state detection unit, said route determination unit determines a route of a communication flow corresponding to a changed state. Network controller. Cost information of communication flow is set in the storage unit for each state of the mobile body,
The communication flow selection unit uses the cost information of the communication flows as a priority, and selects communication flows corresponding to the state of the mobile body in descending order of priority. Item 3. The network control device according to item 2. 4. The route determining unit uses the cost information set corresponding to the state for the selected communication flow to calculate the cost of each route of the communication flow. The network control device according to . 5. The route determination unit adds the cost of the communication flow for which the route has already been determined to the cost of the route of another communication flow including the route and corresponding to the state. The network control device according to . 6. The network according to any one of claims 1 to 5, characterized in that the route determining unit determines a primary route for the selected communication flow and subsequently determines a secondary route for the communication flow. Control device. The state detection unit is
presence or absence of movement of the moving body;
surrounding environment of the moving object;
an operation mode of the moving body;
content of the operation,
7. The network control device according to any one of claims 1 to 6, wherein the state of said moving object is determined based on any one or a combination of: Completing the cost calculation for each path of the communication flow in advance,
The route determining unit, when the state of the mobile object changes, determines the route by referring to the route calculation result for which the cost calculation has been completed, with respect to the communication flow corresponding to the changed state. A network control device according to any one of claims 1 to 7. pre-storing communication flow information for each state of the mobile body with respect to a network mounted on the mobile body in a storage unit;
detecting the state of the moving object;
selecting at least one communication flow corresponding to the detected state by referring to the storage unit;
A network control method, comprising determining at least one route from among a plurality of routes of the selected communication flow, based on a cost calculation result for each route. a process of detecting the state of a moving object;
a process of selecting at least one communication flow corresponding to a detected state by referring to a storage unit pre-stored with communication flow information for each state of the mobile body regarding the network mounted on the mobile body;
a process of determining at least one route from among a plurality of routes of the selected communication flow based on a cost calculation result of each route;
A program that makes a computer run

Images