JP5700426B2 - Vehicle network system - Google Patents

Description

  The present invention relates to a vehicular network in which each node in a vehicle is connected by a communication bus to construct a network, and each node regularly exchanges a regular frame and an irregularly generated event frame via the communication bus. More specifically, the present invention relates to monitoring of the load state of the communication bus.

  In recent years, in the field of vehicle (automobile) control, networking of various control ECUs and state monitoring ECUs such as sensors and switches has been progressing.

  A vehicle network system mounted on this type of vehicle is formed by a serial communication protocol network represented by CAN (Controller Area Network) or LIN (Local Interconnect Network), and this network is, for example, a CAN line type. This network is constructed by connecting the nodes of each ECU in the vehicle to a communication bus.

  FIG. 4 shows an example of the CAN vehicle network system, in which an A node 100a, a B node 100b, a C node 100c,... Are connected to a communication bus 200 to construct an in-vehicle network. Each of the nodes 100a to 100c is a master node, and is formed by an ECU having a microcomputer configuration.

  The nodes 100a to 100c,... Are assigned different IDs (identifiers), and IDs on the transmission side are attached to periodic frames that occur regularly and event frames that occur irregularly due to switching on and off. Then send. Each of the nodes 100a to 100c,... Registers the ID of the frame to be received in the register, and refers to the registered ID of the register to fetch the regular frame and event frame of the corresponding ID of the communication bus 200. Receive.

  FIG. 5 shows, for example, the message buffer 101 of the A node 100a and the non-volatile register 102 that holds the registered ID. If ID = 20 and 30 are registered in the register 102, a periodic frame with ID = 20, The event frame is received and held in, for example, the # 1 buffer area of the message buffer 101, and the periodic frame with ID = 30, and the event frame is received and held in the buffer area of the message buffer 101, for example, # 2, = 25 regular frames and event frames are not received, and are not held in any of the 32 buffers # 1 to # 32 or 64 buffers # 1 to # 64 of the message buffer 101. Depending on the node, most of the buffer of the message buffer 101 is empty.

  By the way, in such a vehicle network system, the bus load of the communication bus 200 is set so that the program update of each of the nodes I00a to 100c,... Can be performed without any problem regardless of the communication status between the nodes. It is set not to exceed a threshold value (for example, 30% with respect to 100% of the load amount).

  However, for example, if any abnormality or failure occurs in any node and the event frame of the node continuously or frequently occurs and occupies the communication bus 200, the bus load of the communication bus increases and the threshold is increased. A situation that exceeds the value occurs. If it is used without noticing this state, it becomes difficult to exchange important information for traveling between nodes, or the program update cannot be performed. Such inconvenience has become a problem since, for example, ECUs connected to a CAN network have increased in recent years.

  Therefore, in a CAN vehicle network system configured to relay a CAN network and a second network having a different communication protocol or communication speed by a gateway node, a gateway having a relatively small processing load and a sufficient processing capacity. The node (ECU) has a bus monitoring function for monitoring the bus load state of the CAN communication bus (CAN bus). Based on the monitoring result, when the bus load of the CAN communication bus increases, the gateway node temporarily It has been proposed to adjust the bus load to be reduced by thinning out the data so as not to hinder the CAN vehicle network system (see, for example, Patent Document 1).

JP 2006-287738 A

  In the case of the vehicle network system of Patent Document 1, it is necessary to provide a high-function and expensive gateway node having a bus monitoring function in the network, which is expensive.

  In particular, in the case of a conventional vehicle network system constructed by a single CAN network, for example, where the second network is not provided, a gateway node is not provided, and it is formed by an inexpensive and low-function ECU. In a general-purpose node of a network, it is difficult to monitor a bus load state of a communication bus by monitoring all periodic frames and event frames exchanged by each node of the network, such as a gateway node.

  An object of the present invention is to enable a vehicle network system to monitor a bus load of a communication bus at a low cost without providing a gateway node.

  In order to achieve the above-described object, the network system for a vehicle according to the present invention is configured such that each node in the vehicle is connected by a communication bus, a network is constructed, and a periodic frame that is periodically generated by each node and irregularly generated. A vehicle network system for exchanging event frames via the communication bus, wherein a predetermined node of the network includes a bus load monitoring unit that monitors a load state of the communication bus, and the bus load monitoring unit includes the bus load monitoring unit, Storage means for preliminarily storing a periodic load of the communication bus based on a periodic frame; calculation means for detecting the event frame of the communication bus and calculating an irregular load of the communication bus based on the event frame; The communication load is calculated from the periodic load stored by the means and the irregular load calculated by the calculation means. It is characterized in that a monitoring process means for monitoring the load state (claim 1).

  According to the first aspect of the present invention, a predetermined node having a large space in the message buffer of the network includes the bus load monitoring means.

  Since the regular frame exchanged by each node of the network is determined, the bus load of the communication bus based on the regular frame is stored in advance as the regular load in the storage unit of the bus load monitoring unit without monitoring the exchange. .

  On the other hand, since the event frame exchanged by each node cannot be predicted, the exchange is actually monitored, and only the irregular load of the communication bus based on the event frame is calculated by the calculation unit of the bus load monitoring unit based on the monitoring result. To do.

  Then, the monitoring processing means of the bus load monitoring means monitors the load state of the communication bus from the periodic load stored in the storage means and the irregular load calculated by the calculation means.

  In this case, since it is possible to monitor the load state of the communication bus by monitoring only the event frame exchanged by each node, for example, in a vehicle network system constructed with a single CAN network and not including a gateway node. Even in such a case, the bus load state of the communication bus can be monitored by a predetermined node of the network formed by an inexpensive general-purpose ECU, and the cost can be reduced.

It is a block diagram of one embodiment of the network system for vehicles of the present invention. FIG. 2 is a detailed block diagram of a predetermined node in FIG. 1. 3 is a flowchart for explaining the operation of a predetermined node in FIG. 2. It is a block diagram of an example of the conventional vehicle network system. It is explanatory drawing of the frame reception of each node of FIG.

  Next, in order to describe the present invention in more detail, an embodiment of the present invention will be described in detail with reference to FIGS.

  FIG. 1 shows a CAN system of a vehicle (automobile) 1 which is a vehicle network system according to the present embodiment. In this CAN system, each node 2a, 2b, 2c,. 3 is constructed with a network of units connected to 3. Each of the nodes 2a to 2c,... Is not a gateway node but a general-purpose node formed by an inexpensive microcomputer-type ECU, and all operate as a master node. Note that the ECUs of the nodes 2a to 2c,... Are specifically ECUs for various controls such as in-vehicle engine control and brake control, and ECUs for each sensor.

  Each node 2a to 2c,... Is assigned a different ID (identifier) in the same manner as the nodes 100a to 100c,... In FIG. 4, and is sent to a periodic frame that occurs regularly and an event frame that occurs irregularly. The ID is attached and transmitted. Each of the nodes 2a to 2c,... Has at least an ID of an event frame to be received and processed registered in a register 21 described later, and the corresponding ID of the communication bus 3 with reference to the registered ID of the register 21. These periodic frames and event frames are received and received in a message buffer 22 described later. Like the message buffer 101 of FIG. 5, the message buffer 22 of each of the nodes 2a to 2c has 32 buffer areas # 1 to # 32 or 64 buffer areas # 1 to # 64. Nodes and the like have many buffer areas available. Further, in order to secure a space between the nodes 2a to 2c,... So that program update or the like can be performed without any trouble, a threshold (for example, a load amount of 70%) at which the bus load of the communication bus 3 is set is set. It is set not to exceed.

  Therefore, a node having a large space in the message buffer 22, such as a node of the steering angle sensor, is set in advance as a predetermined node of the present invention. In FIG. 1, the node 2a is a predetermined node.

  FIG. 2 shows a configuration of a part of a predetermined node 2a, which includes a nonvolatile register 21 and a message buffer 22, and a bus load monitoring unit 23 (to be described later) formed by executing a set bus load monitoring program. Non-volatile storage means 24, calculation means 25, and monitoring processing means 26 are provided.

  The register 21 holds one or a plurality of IDs to be received and received by the node 2a as normal processing IDs, and holds all system IDs including these IDs as reception monitoring IDs. Each ID for reception monitoring may be simply the ID of the node on the transmission side, but in this embodiment, in order to classify and monitor the exchange of frames between the nodes in more detail, the IDs on the transmission side and the reception side A node ID is combined with the ID of the node, and frames exchanged according to these IDs are classified. Specifically, in order to simplify the description, assuming that the nodes connected to the communication bus 3 are three nodes 2a to 2c and the IDs of the nodes 2a to 2c are ID = a, b, c, = (Sender ID, Receiver ID) = (a, b), (a, c), (b, a), (b, c), (c, a), (c, b) Sort by ID combination. In practice, the number of ID combinations is less than the above-mentioned maximum because the number of nodes that receive event frames transmitted from each ID node on the transmission side is limited.

  The message buffer 22 captures and receives the periodic frame and event frame of each ID to be received and received by the node 2a in one or a plurality of predetermined buffer areas registered and set in the register 21, and the remaining free space In each buffer area, the event frame of each ID generated on the communication bus 3 during the set period T0 is received by receiving a combination of IDs on the transmitting side and the receiving side. The buffer area of each ID is set to a capacity larger than the load amount (information amount) of the regular frame or event frame. The predetermined period T0 is, for example, 1 to several seconds, and several tens of frames are exchanged via the communication bus 3 during that period.

  The storage unit 24 stores the size (load amount) of the regular frame in advance for each combination of the transmission side ID and the reception side ID scheduled to be exchanged in each period T0. Note that the ID on the transmitting side and the ID on the receiving side that are scheduled to be exchanged in each period T0 are known in the program, and the load of the periodic frame of each ID of the combination is set individually for each combination of nodes. However, it is simply set by dividing the average load amount of the regular frame of the communication bus 3 by the number of IDs.

  The calculation means 25 calculates the event of the communication bus 3 for each ID from the load amount of the event frame of each ID of the combination of the transmission side and the reception side taken into each buffer area of the message buffer 22 during the period T0. The frame is detected, and the irregular load of the communication bus 3 based on the event frame in the period T0 is calculated.

  The monitoring processing unit 26 adds the periodic load stored in the storage unit 24 and the irregular load calculated by the calculation unit 25 to add the total load of the communication bus 3 in each period T0 and the combination of the transmission side and the reception side. The individual load amount for each ID is calculated, and the load state of the communication bus 3 is monitored at the period T0 from the total load amount and the individual load amount.

  If any abnormality or failure occurs in any one of the nodes 2a to 2c,... And the total load on the communication bus 3 exceeds a set threshold value due to continuous or frequent transmission of that node, the monitoring processing means 26, an ID exceeding a set individual threshold is detected from the individual load for each ID of the combination of the transmitting side and the receiving side, and an abnormality or failure has occurred based on this ID combination The node is identified and stored as abnormality information in the internal non-volatile memory and the fail-safe process is executed. For example, the warning lamp on the meter display panel of the vehicle is turned on (or flashes), causing the driver to The generation is notified, and communication stop other than important information related to traveling of the vehicle 1 is communicated to the other nodes 2b, 2c,... Of the communication bus 3 to reduce the load on the communication bus 3.

  FIG. 3 shows a procedure of a bus load monitoring process for a communication bus based on the execution of the bus load monitoring program of a predetermined node 2a. This bus load monitoring process is performed by an ignition (hereinafter referred to as IG) key, an IG switch, The operation is turned on by the driver, for example, when the engine is started. Thereafter, the IG key and the IG switch are repeatedly operated at a substantially predetermined period T0 until the driver turns off the IG key or IG switch.

  Then, as shown in steps S1 to S16 of FIG. 3, first, all IDs of the event frames are read from the register, and the node 2a is placed in each buffer area # 1 to # 32 (# 64) of the message buffer 22. After each ID to be received and each ID to be monitored are set (step S1), the timer for the predetermined period T0 of the monitoring processing means 26 is initially reset to T0 = 0 and starts measuring the predetermined period T0. (Step S2) Then, the monitoring processing means 26 reads out the magnitude of the load of the regular frame for each ID that combines the ID on the transmission side and the ID on the reception side held in the storage means 24 (Step S3). .

  Further, to simplify the explanation, the nodes connected to the communication bus 3 are three nodes 2a to 2c as described above, and the IDs of the nodes 2a to 2c including the predetermined node 2a are ID = a, b, c, the maximum ID = (transmitting side, receiving side) = (a, b), (a, c), (b, a), (b, c), (c, a), (c, For example, if an a event frame is generated in the node 2a with ID = a based on another classification of the six combinations of IDs in b) (YES in step S4), the node 2a registered in the register 21 is sent to the transmission side. The a event frame of the communication buffer corresponding to ID = (a, b), (a, c) is written in the corresponding buffer area of the message buffer 22, and the a event of each written ID Calculation means 2 from information amount of frame 2 There is calculated latest load of a event frame (information amount) for each ID (step S5). Similarly, when a b event frame transmitted from the node 2b with ID = b occurs (YES in step S6), a combination of IDs with the node 2b registered in the register 21 as the transmission side, for example, ID = (b, The b event frames corresponding to a) and (b, c) are written in the corresponding buffer areas of the message buffer 22, and the calculation means 25 calculates the latest b event frame from the information amount of the written b event frame of each ID. A load amount (information amount) is calculated for each ID (step S7). Similarly, when a c event frame transmitted from the node 2c with ID = c occurs (YES in step S8), a combination of IDs with the node 2c registered in the register 21 as the transmission side, for example, ID = (c, a) The c event frame corresponding to (c, b) is written in the corresponding buffer area of the message buffer 22, and the calculation means 25 calculates the latest c event frame from the information amount of the c event frame of each ID written. A load amount (information amount) is calculated for each ID (step S9). This is because if the IDs on the transmitting side are the same but the IDs on the receiving side are different, transmission is repeated by the number of IDs on the receiving side even for frames with the same content, and the load increases.

  Next, the latest load amount of the event frame for each ID is added by the calculation means 25 to calculate the total load (unscheduled load) of the event frame on the communication bus 3 (step S10).

  Then, the calculation of the load amount and the total load amount for each ID on the communication bus 3 for the event frame is repeated within a predetermined period T0 (NO in step S11), and when the predetermined period T0 has elapsed, the timer is stopped, The monitoring processing means 26 calculates the total load per unit time (for example, 1 second) of the communication bus 3 from the load obtained by adding the regular load and the irregular load in the predetermined period T0 (step S12).

  If the calculated total load is within a set threshold value (for example, 30%), the monitoring processing unit 26 is normal with no abnormality or failure occurring in any of the nodes 2a to 2c. Judgment is made (YES in step S13), and nothing is done and the process returns to step S2 to monitor the next predetermined period T0.

  On the other hand, when the total load of the communication bus 3 exceeds the set threshold value, the monitoring processing unit 26 has some abnormality or failure in any of the nodes 2a to 2c, and the event frames of the node continue. Alternatively, it is determined that a situation occurs frequently that the communication bus 200 is occupied (step S13), and a load amount (individual load amount) for each ID for a predetermined period T0 and a set threshold value. (Individual threshold value) is compared, and a node in which an abnormality or failure has occurred is identified from an ID that exceeds the individual threshold value (step S14), and an abnormality (failure) has occurred in that node An event frame is transmitted to notify a specific node where an abnormality (failure) has occurred (step S15), and communication from outside the node can be stopped. DoFurther, fail-safe processing is executed, and the remaining nodes are notified of the prohibition of transmission of important information related to the travel of the vehicle 1 other than the frame (step S16), and the free space of the communication bus 3 is secured as much as possible. To do.

  As described above, in the case of the present embodiment, the predetermined node 2 a having a large vacancy in the network message buffer 22 of the vehicle network system of the vehicle 1 includes the bus load monitoring unit 23.

  Since the regular frames exchanged between the nodes 2a to 2c of the network are determined, the bus load of the communication bus 3 based on the regular frames is stored in the storage means 24 of the bus load monitoring means 23 without monitoring the exchanges. Is stored in advance as a periodic load.

  On the other hand, since the event frame exchanged between the nodes 2a to 2c cannot be predicted, the exchange is actually monitored by the bus load monitoring means 23, and only the irregular load of the communication bus based on the event frame is based on the monitoring result. Is calculated by the calculating means 25.

  The monitoring processing unit 26 of the bus load monitoring unit 23 can monitor the load state of the communication bus 3 from the periodic load stored in the storage unit 24 and the irregular load calculated by the calculation unit 25. Furthermore, based on the monitoring result, it is possible to identify a node where an abnormality or failure has occurred and notify the node that an abnormality has occurred. In addition, transmission of important information other than frames related to the traveling of the vehicle 1 to the nodes 2a to 2c,... Is prohibited, the free space of the communication bus 3 is ensured as much as possible, and troubles such as program update occur. In addition, it is possible to prevent and avoid a fatal crisis of the system based on the abnormality of the bus load by maintaining the exchange between nodes as much as possible.

  And in the vehicle network system of this embodiment constructed by a single network of CAN and not including a gateway node, only the event frames exchanged by the inexpensive general-purpose nodes 2a to 2c, ... are monitored and communicated. The load state of the bus 3 can be monitored at low cost, and the cost can be reduced.

  Also, based on the monitoring results, identify the node where an abnormality or failure has occurred, easily track the software bug from the event frame sent by the node, etc., quickly investigate the cause of the failure, recover, etc. can do.

  The present invention is not limited to the above-described embodiment, and various modifications other than those described above can be made without departing from the spirit thereof. For example, the configuration of the bus load monitoring means is The number of nodes connected to the communication bus is not limited to the configuration shown in FIG.

  In the above embodiment, the present invention is applied to a vehicle network system constructed with a CAN network. However, the present invention can be applied to a vehicle network system constructed with a LIN network in the same manner. The present invention can be applied to a vehicle network system for various vehicles.

DESCRIPTION OF SYMBOLS 1 Vehicle 2a-2c Node 3 Communication bus 23 Bus load monitoring means 24 Storage means 25 Calculation means 26 Monitoring processing means

Claims (1)

A network system for a vehicle in which each node in a vehicle is connected by a communication bus, and a network is constructed. ,
A predetermined node of the network includes bus load monitoring means for monitoring a load state of the communication bus;
The bus load monitoring means includes
Storage means for storing in advance a periodic load of the communication bus based on the periodic frame;
Calculating means for detecting the event frame of the communication bus and calculating an irregular load of the communication bus based on the event frame;
A vehicle network system comprising: a monitoring processing unit that monitors a load state of the communication bus from the periodic load stored in the storage unit and the irregular load calculated by the calculation unit.

Images