JP2010187323A - Gateway device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve efficiency of data transfer in a gateway device connecting a plurality of networks in a vehicle. <P>SOLUTION: In a gateway device for performing data transfer between a CAN and a LIN, the timing of reception in the CAN is detected and the timing to start a communication schedule in the LIN is adjusted to start the communication schedule just after the reception timing. In the case where the communication schedule in the LIN is constituted of (transmission of) a transmission frame and (reception of) a reception frame, when the communication schedule in the LIN is started just after the reception timing in the CAN, data received in the CAN are immediately transferred to the LIN in accordance with the communication schedule, like a LIN communication schedule pattern (11), thereby improving transfer efficiency. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a gateway device that is used in a vehicle network mounted on a vehicle and performs data transfer processing by connecting a plurality of networks in the vehicle network.

  As a vehicle network mounted on a vehicle, one or more wire harnesses are connected between an electronic control device (hereinafter referred to as an ECU) and sensors or actuators, and individual wiring harnesses are used to connect individual devices. There is one that exchanges signals according to the specifications.

In vehicles, the amount of information to be exchanged tends to increase as control becomes more sophisticated and complicated, and more wire harnesses are required in such a vehicle network. For this reason, there existed a problem that mounting property deteriorated or a weight and cost increased.
Therefore, in recent years, a multiple communication network such as a CAN (Controller Area Network) or a LIN (Local Interconnect Network) may be used. The CAN communication specification is a CSMA / CA (Carrier Sense Multiple Access / Collection Avidance) system. In CAN, each communication device (for example, ECU) can immediately transmit data if the communication bus is free.

  On the other hand, the communication specification of LIN is a single master / multi-slave system. In the LIN, there is one master, and communication is performed according to scheduling by the master. For example, the slave cannot transmit data unless instructed by the master.

  FIG. 19 is a diagram illustrating an example of a configuration of a vehicle network.

  A vehicle network 300 illustrated in FIG. 19 includes ECUs 301 and 302. The ECUs 301 and 302 include known microcomputers, are connected via a CAN communication line 307, and are configured to be communicable according to a CAN protocol.

  The ECU 301 is connected to a control device 303 as a control target via wire harnesses 304 and 305. ECU 301 transmits and receives data to and from control device 303 via wire harnesses 304 and 305. ECU 302 is connected to control device 303 via wire harness 306. ECU 302 transmits and receives data to and from control device 303 via wire harness 306. In the example of FIG. 19, three wire harnesses are used.

  FIG. 20 is a diagram illustrating an example of a configuration of a vehicle network using CAN and LIN. The connection using the wire harnesses 304, 305, and 306 in FIG. 19 is replaced with the connection using the LIN communication line 404, and communication is performed according to the LIN protocol. It is what I did.

  A vehicle network 400 illustrated in FIG. 20 includes ECUs 401 and 402. The ECUs 401 and 402 are connected to each other via a CAN communication line 405 and configured to be communicable according to a CAN protocol.

  The ECU 401 is connected to a control device 403 as a control target via the LIN communication line 404, and transmits and receives data to and from the control device 403 via the LIN communication line 404 according to the LIN protocol. In LIN in FIG. 20, the ECU 401 is a master and the control device 403 is a slave.

  In FIG. 20, the ECU 401 also functions as a gateway that connects CAN and LIN. For example, the data received from the ECU 402 on the CAN is sent to the LIN, or the data received from the control device 403 on the LIN is sent to the CAN. Or send to. The ECU 402 can communicate with the control device 403 via the ECU 401 and can control the control device 403. For example, Patent Document 1 discloses a gateway device that transfers data by connecting a plurality of networks.

  Here, FIG. 21 is a diagram illustrating an example of a configuration of a message frame transmitted and received in the LIN and a communication schedule in the LIN (hereinafter simply referred to as a communication schedule). Hereinafter, communication in LIN is simplified and described as LIN communication. Similarly for CAN, communication in CAN is referred to as CAN communication.

  A message frame in LIN communication is composed of a header initialized by the master and a response following the header. A header is periodically sent from the master, and a response is sent from the master or slave according to the contents of the header.

  As shown in FIG. 21, the header includes three fields: Sync Break (synchronization break), Sync Field (synchronization field), and Ident Field (identifier field). Sync Break (synchronous break) represents the beginning of a frame. The Synch Field (synchronization field) is used to adjust the baud rate of the slave. An Identity Field (identifier field) contains an identifier byte. The identifier byte uniquely defines the purpose of the message frame.

  As shown in FIG. 21, the response includes two fields, Data Field (data field) and Check Sum (checksum). The Data Field (data field) stores specific data. Check Sum (checksum) performs an error detection function.

The message frame is classified into a transmission frame in which data from the master to the slave is stored and a reception frame in which data from the slave to the master is stored.
Next, an example of a communication schedule will be described. In the example of FIG. 21, if the data amount of both the transmission frame and the reception frame is 4 [bytes] and the transfer baud rate is 19.2 [kbps], the transfer time of one transmission frame or reception frame is about 6.2. [Msec]. When one communication schedule (hereinafter also referred to as one schedule) is configured by transmission / reception of two frames of a transmission frame and a reception frame, the master may transmit a header every 8 [msec], for example. The value of 8 [msec] is obtained by adding a margin to the above 6.2 [msec]. At this time, the time of one schedule is 16 [msec]. By the way, in LIN, a schedule is controlled based on a periodic task that is activated at a predetermined cycle. If the communication schedule is controlled by a periodic task activated at a cycle of 4 [msec], one schedule is completed every four cycles (every four periodic tasks).

JP 2006-319759 A

  Here, FIG. 22 is a diagram illustrating an example of data transfer between CAN and LIN. Specifically, an example of data transfer from CAN to LIN is shown. For example, in FIG. 20, it is assumed that the ECU 401 transfers data from CAN to LIN.

  It is assumed that the ECU 401 receives data to be transferred in the CAN at a cycle of 16 [msec]. Further, it is assumed that the periodic task period in the LIN in which the ECU 401 operates as a master is 4 [msec] and the communication schedule period is 16 [msec]. The communication schedule is composed of transmission / reception of transmission frames and transmission / reception of reception frames (see FIG. 21).

  When the ECU 401 transfers the data 1 received in the CAN to the LIN, the ECU 401 cannot transfer the data 1 to the LIN until the start of the next communication schedule in the LIN. For example, if data 1 is received in CAN immediately after the start of the communication schedule in LIN, in order to transfer the data 1 to LIN, approximately one period until the next communication schedule in LIN (one period of the communication schedule) Min), I have to wait.

  Next, the data transfer from CAN to LIN will be described more specifically with reference to FIG.

  FIG. 23 shows an example in which, for example, the ECU 401 (see FIG. 20) transfers data received in the CAN at a cycle of 16 [msec] to the LIN. In LIN, the communication schedule is controlled by a periodic task that starts at a cycle of 4 [msec]. One communication schedule includes transmission / reception of transmission frames and transmission / reception of reception frames. In this example, a transmission frame is transmitted and received for each communication schedule (the ECU 401 transmits a transmission frame). In other words, a transmission frame is transmitted and received every 16 [msec] corresponding to four periodic tasks (hereinafter referred to as four tasks). FIG. 23 shows four patterns (LIN communication schedule patterns (11) to (14)) in which the start timing of the LIN communication schedule differs by one task (4 [msec]).

  As shown in FIG. 23, in the communication schedule pattern (11) among the LIN communication schedule patterns (11) to (14), the data received by the CAN is transferred to the LIN in the shortest time.

  On the other hand, in the LIN communication schedule patterns (12) to (14), there is a delay compared to the LIN communication schedule pattern (11) until the data received by the CAN is transferred to the LIN.

  Specifically, the LIN communication schedule pattern (12) is delayed by one task (1/4 of the communication schedule period), and the LIN communication schedule pattern (13) is two tasks (2/4 of the communication schedule period). ), And the LIN communication schedule pattern (14) is delayed by 3 tasks (3/4 of the communication schedule period).

  In this way, when data is transferred between CAN and LIN, if the data reception cycle in CAN and the communication schedule cycle in LIN are different, there may be a delay in data transfer from CAN to LIN. .

  Next, FIG. 24 is a diagram illustrating an example of a mode in a case where the ECU 401 (see FIG. 20) transfers data received by the LIN to the CAN, for example. In LIN, the communication schedule is controlled by a periodic task that starts at a cycle of 4 [msec]. One communication schedule includes transmission / reception of transmission frames and transmission / reception of reception frames. In this example, the received frame is transmitted and received every cycle of the communication schedule (4 tasks: every 16 [msec]). FIG. 24 shows four patterns (LIN communication schedule patterns (15) to (18)) in which the start timing of the LIN communication schedule differs by one task (4 [msec]).

  As shown in FIG. 24, in the communication schedule pattern (18) among the LIN communication schedule patterns (15) to (18), the data received by the LIN is transferred to the CAN in the shortest time.

  On the other hand, in the LIN communication schedule patterns (15) to (17), there is a delay compared to the LIN communication schedule pattern (18) until the data received by the LIN is transferred to the CAN.

  Specifically, the LIN communication schedule pattern (17) is delayed by one task (1/4 of the communication schedule period), and the LIN communication schedule pattern (16) is two tasks (2/4 of the communication schedule period). ), And the LIN communication schedule pattern (15) is delayed by 3 tasks (3/4 of the communication schedule period).

  As described above, when data is transferred between CAN and LIN, if the data transmission cycle in CAN and the communication schedule cycle in LIN are different, there may be a delay in data transfer from LIN to CAN. .

  The present invention has been made in view of these problems, and an object thereof is to improve the efficiency of data transfer in a gateway device that connects a plurality of networks in a vehicle.

  In order to achieve the above object, the invention described in claim 1 is a gateway device connected to a plurality of networks in a vehicle and transferring data between the networks, and at least periodically by a master device. In a gateway apparatus for transferring data between a network (hereinafter referred to as a first network) in which data is transmitted and received according to a communication schedule activated by the network (hereinafter referred to as a second network) It is configured to operate as a master device and activate a communication schedule.

  Then, the first network and the second network are monitored, transmission / reception timing when the transfer target data is transmitted / received on the first network, and the transfer target data is transmitted / received on the second network. Detection means for detecting the transmission / reception timing in this case, and because the transmission timing on the first network and the reception timing on the second network are synchronized based on the detection result of the detection means, or on the first network Calculating means for calculating a start timing of the communication schedule for synchronizing the reception timing at the transmission and the transmission timing on the second network, so that the communication schedule is started at the start timing calculated by the calculating means It has become.

  According to such a gateway device, since the communication schedule in the first network is activated so that the transmission timing on the first network and the reception timing on the second network are synchronized, the first network The data received at can be immediately transmitted to the second network. In addition, since the communication schedule in the first network is activated so that the reception timing on the first network and the transmission timing on the second network are synchronized, the data received on the second network is immediately Can be transmitted to the first network. That is, data to be transferred can be transferred in a shorter time between the first network and the second network.

  Further, by providing the detection means, it is possible to detect the transmission / reception timing for the desired data. For this reason, the data of the transmission / reception timing detection target (in other words, the target of synchronization) may be changed according to the situation, and the transfer can be optimized according to the time.

  In particular, when there are a plurality of data to be transferred, it is preferable to configure as in claim 2.

  The gateway device according to claim 2 is the gateway device according to claim 1, wherein there are a plurality of types of data to be transferred, the plurality of types of data includes information indicating importance, and the detection means Of the data having the highest importance, the transmission / reception timing on the first network and the transmission / reception timing on the second network are detected.

  According to such a gateway apparatus, since the transmission / reception timing on the first network and the transmission / reception timing on the second network are detected for the data having the highest importance, the data having the highest importance is detected. The transmission timing on the first network and the reception timing on the second network are synchronized, or the reception timing on the first network and the transmission timing on the second network are synchronized. It becomes like this. Therefore, the data having the highest importance level is transferred in a shorter time.

  Further, when there are a plurality of data to be transferred, a configuration as in claim 3 may be adopted.

  The gateway device according to claim 3 is the gateway device according to claim 1, wherein there are a plurality of types of data to be transferred, and the calculating means calculates a frequency distribution of transmission / reception timings of the data to be transferred on the second network, Based on the frequency distribution, the transmission timing on the first network and the reception peak on the second network are synchronized, or the reception timing on the first network and the second network The start timing of the communication schedule for synchronizing with the peak of transmission is calculated.

  In this case, the transmission timing on the first network is synchronized with the timing at which a relatively large number of receptions occur on the second network. Alternatively, the reception timing of the first network is synchronized with the timing at which transmissions are relatively frequent on the second network. According to such a configuration, the transfer time can be shortened as a whole for a plurality of data.

  Next, the gateway device according to claim 4 is characterized in that, in the gateway device according to claims 1 to 3, the activation timing of the communication schedule is gradually changed by a predetermined amount to coincide with the activation timing calculated by the calculating means. Yes.

  When the activation timing of the communication schedule is changed to the activation timing calculated by the calculating means, there is a concern that if the change amount is large, communication is affected. For example, when the activation timing is greatly delayed as a result of the change, there is a possibility that the receiving side may determine that communication is interrupted.

  For example, it is assumed that there is an example in which the first communication schedule starts at time T1, and thereafter the communication schedule starts every 10 [msec]. That is, the activation timing of the communication schedule is T1, T1 + 10, T1 + 20, T1 + 30, T1 + 40, T1 + 50.

  Assume that after starting the communication schedule at T1, it is desired to change the start timing of the communication schedule such as T1 + 16, T1 + 26, T1 + 36, T1 + 46, T1 + 56. Then, after starting at T1, it is only necessary to start the next communication schedule after waiting for 16 [msec]. However, 6 [msec] is delayed from the original T1 + 10, and no data is transmitted or received during that time. There may be inconveniences.

  In this regard, in the invention of claim 4, it is gradually changed by a predetermined amount to approach the target activation timing. For example, as described above, when it is desired to change to the start timings of T1 + 16, T1 + 26, T1 + 36, T1 + 46, T1 + 56... An example of starting every 10 [msec] to T1 + 46, T + 56. According to this, it is possible to finally match the target start timing while gradually shifting the start timing. As described above, since the amount of change until the target activation timing is matched can be reduced, for example, it is possible to avoid a situation in which it is erroneously determined that communication is interrupted.

  By the way, when data transfer is more urgent, it may be configured as in claim 5.

  The gateway apparatus according to claim 5 is the gateway apparatus according to claims 1 to 3, wherein the data to be transferred includes information indicating an urgency level, and determines whether or not an urgent situation is required based on the information indicating the urgency level. If it is determined that an emergency is required, the currently operating communication schedule is canceled and a new communication schedule is activated.

  According to this, it is not necessary to wait until the currently operating communication schedule is completed, and a new communication schedule can be started immediately at a desired timing. Therefore, the transmission timing on the first network and the reception timing on the second network, or the reception timing on the first network and the transmission timing on the second network are immediately synchronized. Is also possible.

  Next, the gateway device according to claim 6 does not change again in the gateway device according to claims 1 to 5 until a predetermined time elapses after the activation timing of the communication schedule is changed. ing.

  If the start timing of the communication schedule is frequently changed, the start timing of the communication schedule may not be constant, and the transfer efficiency may be deteriorated. Therefore, if it is configured as in claim 6, it is possible to suppress frequent changes in the start timing, and it is possible to prevent the occurrence of problems such as poor communication efficiency.

  Next, the gateway device according to claim 7 includes dummy data transmission / reception means for transmitting / receiving dummy data at the same timing as the data to be transferred instead of the predetermined data to be transferred in the gateway device according to claim 1, The detection means detects transmission / reception timing when dummy data is transmitted / received on the first network and transmission / reception timing when dummy data is transmitted / received on the second network. The transmission timing on the first network and the reception timing on the second network are synchronized, or the reception timing on the first network and the transmission timing on the second network are synchronized. The start timing of the communication schedule is calculated.

  According to such a configuration, the activation timing of the communication schedule is adjusted in advance using dummy data, for example, when the gateway device does not perform data transfer or control of the control target (for example, during initialization processing immediately after activation). can do.

  By adjusting in advance, it is not necessary to change the start timing of the communication schedule, for example, when it is necessary to frequently transfer data or during control of a control target.

  Next, the gateway device according to claim 8 is the gateway device according to any one of claims 1 to 7, wherein information indicating that the activation timing of the communication schedule is changed is provided to both or one of the first network and the second network. It is supposed to be sent out.

  According to this, the communication devices on the first network and the second network can recognize that the data transmission / reception timing is changed.

  For example, when the transmission / reception timing is changed, if the communication device cannot recognize the change, it may be determined that communication is interrupted in some cases. On the other hand, if the communication device can recognize the change in transmission / reception timing, it is possible to invalidate the communication interruption determination process in the communication device, and the communication device determines that the communication interruption is unnecessarily unnecessary. The situation can be avoided.

It is drawing which shows the vehicle network of this embodiment. It is drawing which shows the structure of ECU101 (gateway) of this embodiment. It is a flowchart showing the flow of a precondition determination process. It is a figure for demonstrating a precondition (the 1). It is a figure for demonstrating a precondition (the 2). It is a figure for demonstrating a precondition (the 3). It is a figure for demonstrating a precondition (the 4). It is a figure for demonstrating a precondition (the 5). It is a figure for demonstrating a precondition (the 6). 3 is a flowchart illustrating processing executed in a communication synchronization device 202 of the ECU 101. 3 is a flowchart showing processing executed in a LIN communication scheduler 206 of the ECU 101. It is a figure for demonstrating the modification 1 (transfer from CAN to LIN). It is a figure for demonstrating the modification 1 (transfer from LIN to CAN). 10 is a diagram for explaining a second modification. 10 is a diagram for explaining a third modification. 10 is a diagram for explaining a modification example 4; It is drawing for demonstrating the modification 5 (time chart). It is drawing for demonstrating the modification 5 (graph). It is drawing which shows an example of the conventional vehicle network. It is drawing which shows an example of the conventional vehicle network (multiplex communication). It is drawing explaining the structure of the flame | frame in LIN. It is drawing which shows an example of the aspect of the data transfer from CAN to LIN. It is a figure showing the pattern of the aspect of the data transfer from CAN to LIN. It is a figure showing the pattern of the aspect of the data transfer from LIN to CAN.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[First Embodiment]
FIG. 1 is a diagram illustrating a vehicle network according to the present embodiment. The vehicle network 100 shown in FIG. 1 differs from the vehicle network 400 shown in FIG. 20 only in reference numerals and has the same configuration as the vehicle network 400. For this reason, the description of the configuration of the vehicle network 100 is omitted here.

  FIG. 2 is a configuration diagram showing a specific configuration of ECU 101 in FIG.

  As shown in FIG. 2, the ECU 101 of this embodiment includes a communication synchronization device 202, a CAN communication driver 203, a database 204, a LIN communication driver 205, and a LIN communication scheduler 206.

  The CAN communication driver 203 receives data from the CAN communication line 207, stores the received data in the database 204, and sends a data reception completion notification to the communication synchronization apparatus 202.

  Upon receiving a reception completion notification from the CAN communication driver 203, the communication synchronization apparatus 202 sends a communication schedule adjustment instruction to the LIN communication scheduler 206.

  The LIN communication scheduler 206 adjusts the communication schedule (communication start time) based on the adjustment instruction from the communication synchronization apparatus 202 and then sends the communication instruction to the LIN communication driver 205.

  Upon receiving a communication instruction from the LIN communication scheduler 206, the LIN communication driver 205 acquires the corresponding data from the database 204 based on the communication instruction, and sends the acquired data to the LIN communication line 208. When the data transmission is completed, the LIN communication scheduler 206 sends a transmission completion notification to the communication synchronization apparatus 202.

  In this way, data is transferred from CAN to LIN.

  The LIN communication driver 205 receives data from the LIN communication line 208, stores the received data in the database 204, and sends a data reception completion notification to the LIN communication scheduler 206.

  The LIN communication scheduler 206 sends a reception completion notification to the communication synchronization apparatus 202. Further, the communication synchronization apparatus 202 sends a notification that the reception completion notification has been received from the LIN communication scheduler 206 to the CAN communication driver 203.

  The CAN communication driver 203 acquires the corresponding data from the database 204 based on the notification from the communication synchronization apparatus 202, and sends the acquired data to the CAN communication line 207. Here, if the transmission cycle or transmission timing of the transmission target data is predetermined, the CAN communication driver 203 transmits the data according to the transmission cycle or transmission timing. At this time, if the CAN communication line 207 is free, data is transmitted from the CAN communication driver 203 onto the CAN communication line 207. On the other hand, if the CAN communication line 207 is not free due to transmission / reception of other data, transmission processing is performed again. Then, when the transmission of data is completed, the CAN communication driver 203 sends a transmission completion notification to the communication synchronization apparatus 202.

  In this way, data is transferred from LIN to CAN.

  In this embodiment, the communication synchronization device 202 of the ECU 101 executes the process of FIG. 10, and the LIN communication scheduler 206 of the ECU 101 executes the process shown in FIG. Communication schedule).

  Here, in the present embodiment, the transmission / reception cycle (hereinafter also referred to as CAN transmission / reception cycle, Tc) of CAN to be transferred and the LIN communication schedule cycle (hereinafter, also referred to as LIN communication schedule cycle, Tl). To determine whether or not a precondition for the period is satisfied.

  FIG. 3 shows a flowchart showing the flow of processing related to the determination as to whether or not the precondition for this cycle is satisfied. The precondition determination process shown in FIG. 3 is executed, for example, in a computer (not shown) that virtually constructs the vehicle network 100 and simulates the operation. Then, the user sees the processing result in the computer, and confirms whether or not the precondition for the cycle is satisfied. 3 may be executed in the ECU 101 of the vehicle network 100, but the processing load on the ECU 101 is suppressed by being executed in another computer like the former. be able to.

  In the precondition determination process of FIG. 3, the CPU of the computer (not shown) first determines in S10 whether the CAN transmission / reception cycle and the LIN communication schedule cycle are equal, and if it is determined that they are equal (S10: YES), it is valid. (The precondition for the period is satisfied) and the process is terminated.

  On the other hand, if it is determined in S10 that the CAN transmission / reception period is not equal to the LIN communication schedule period (S10: NO), the process proceeds to S20. Note that the content of the processing of S20 will be supplemented in the description of FIGS.

  In S20, in the period of the least common multiple of the CAN transmission / reception cycle and the LIN communication schedule cycle, it is determined whether or not all possible patterns appear as patterns of deviation between the CAN transmission / reception timing and the start timing of the LIN communication schedule, If it is determined that all the patterns do not appear (S20: YES), it is determined that the pattern is valid (the precondition regarding the cycle is satisfied), and the process ends.

  On the other hand, if it is determined in S20 that all patterns appear (S20: NO), it is determined that the pattern is invalid (the precondition regarding the cycle is not satisfied), and the process ends.

  In S20, when YES is determined (when it is determined that the precondition for the cycle is satisfied) and when NO is determined as negative (when it is determined that the precondition for the cycle is not satisfied), This will be specifically described with reference to FIGS.

  FIGS. 4 to 6 show aspects of a pattern of a shift in period between CAN and LIN when data received in CAN is transferred to LIN.

  FIGS. 7 to 9 show aspects of the pattern of the shift of the period between the LIN and the CAN when the data received in the LIN is transferred to the CAN.

Here, in LIN, a periodic task is activated at a predetermined cycle, and a communication schedule is started in accordance with the periodic task. The upper part of FIGS. 4 to 9 shows periodic tasks, and serial numbers that continue from “1” are given for each period. Hereinafter, for the sake of convenience, the transmission / reception cycle in the CAN is also expressed on the basis of the periodic task.
<Example of FIG. 4: CAN->LIN>
First, FIG. 4 shows a case where the CAN reception cycle (hereinafter also referred to as a CAN reception cycle) is for four regular tasks and the LIN communication schedule cycle is for six regular tasks. The data reception timing in the CAN corresponds to the timing of the regular task 4, the regular task 8, the regular task 12, the regular task 16, the regular task 20, the regular task 24,. To do. The received data is sequentially referred to as data (1), data (2), data (3), data (4), data (5), data (6). Further, data received before data (1) is data (0), and data received before that is data (-1). Such data (-1) to (6) received at the CAN is transferred to the LIN according to the communication schedule of the LIN.

  There are six possible patterns in the example of FIG. 4 for the start timing of the communication schedule in LIN relative to the reception timing in CAN (offset = 0 to 5).

  In the stage of offset = 0 to 5, when the communication schedule in LIN is started with the periodic tasks 1, 7, 13, 19... (when offset = 0), the periodic tasks 2, 8, 14, When started with 20... (Offset = 1), when started with periodic tasks 3, 9, 15, 21... (When offset = 2), scheduled tasks 4, 10, 16,. When started with 22... (When offset = 3), when started with periodic tasks 5, 11, 17, 23... (When offset = 4), scheduled tasks 6, 12, 18,. This represents the case where the period starts at 24... (When offset = 5).

  Here, with respect to each pattern of offset = 0 to 5, it is examined whether the timing at which data is received by the CAN corresponds to the number of regular tasks as viewed from the start timing of the communication schedule in the LIN. In addition, the number showing this number is also written in the figure. This is also shown in FIGS. Further, hereinafter, a number representing the number of times is defined as Ncl.

  In the example of offset = 0, data is first received by CAN at the fourth periodic task (hereinafter referred to as the fourth task), counting from the start timing of the first communication schedule (periodic task 1). Has been.

  Subsequently, data is received by the CAN at the second task (Ncl = 2) counting from the start timing of the next communication schedule (periodic task 7). Furthermore, data is received by CAN at the sixth task (Ncl = 6) counting from the start timing of the same communication schedule (periodic task 7).

  In this way, the task at which the reception timing in CAN is viewed from the start timing of the LIN communication schedule (that is, the value of Ncl) is the period of the least common multiple of the CAN reception period and the LIN communication schedule period (that is, periodic In the example of offset = 0, Ncl can take a value of (2, 4, 6).

  In the same way, when extracting for the example of offset = 1, Ncl can take the value of (1, 3, 5).

  In the case of the example of offset = 2, 4, Ncl takes the value of (2, 4, 6) as in the case of offset = 0, and in the case of the example of offset = 3, 5, Ncl is equal to (1 , 3, 5).

  Thus, in the example of FIG. 4, Ncl takes either (1, 3, 5,) or (2, 4, 6), and all of (1, 2, 3, 4, 5, 6) Since there is no example of taking a value (pattern), in S20 in the precondition determination process of FIG. 3, an affirmative determination is made that all patterns do not appear (S20: YES), and it is valid (the precondition for the period is satisfied). ). In other words, it is determined that the effectiveness of synchronizing the CAN reception cycle and the LIN communication schedule cycle can be obtained.

  In order to examine the presence or absence of effectiveness, after receiving data at the CAN for the period of the least common multiple of the CAN reception period and the LIN communication schedule period (that is, the period of the periodic tasks 1 to 12), Consider the total time required to transfer to the LIN (until the LIN communication schedule starts) (hereinafter referred to as the total transfer time).

  In the example of offset = 0, data (0) and (1) are transferred to LIN during the period of the regular tasks 1 to 12. Regarding data (0), the reception timing in CAN (not shown, immediately before the regular task 1) and the transfer timing to the LIN (communication schedule start timing, the regular task 1) are synchronized, There is no delay. Next, the data (1) is received by the CAN at a timing corresponding to the regular task 4, and then the transfer to the LIN is started (the LIN communication schedule is started) by the regular task 7, until the transfer starts. Has a delay of two tasks. From the above, the total transfer time is two tasks.

  Similarly, considering the total transfer time for the case of offset = 1 to 5, there are 4 tasks for the example of offset = 1, 2 tasks for the example of offset = 2, and 4 for the example of offset = 3. There are two tasks in the example of offset = 4, and four tasks in the example of offset = 5.

In the example of FIG. 4, when offset = 1, 3, and 5, the start timing of the communication schedule in the LIN is adjusted so that the total transfer time is 2 tasks (the CAN reception cycle and the LIN communication schedule cycle). Can be synchronized). In other words, the time required for transfer can be shortened (minimized), and the effectiveness of synchronization can be obtained.
<Example of FIG. 5: CAN → LIN>
Next, FIG. 5 shows a case where the CAN reception cycle is four regular tasks and the LIN communication schedule cycle is five regular tasks. The data reception timing in CAN corresponds to the timings of the regular task 4, the regular task 8, the regular task 12, the regular task 16, the regular task 20,.

  There are five possible patterns in the communication schedule start timing in LIN with respect to the reception timing in CAN in the example of FIG. 5 (offset = 0 to 4).

  In the stage of offset = 0 to 4, when the communication schedule in LIN is started by the periodic tasks 1, 6, 11, 16,... (when offset = 0), the periodic tasks 2, 7, 12, When started with 17 ... (when offset = 1), when started with periodic tasks 3, 8, 13, 18 ... (when offset = 2), with periodic tasks 4, 9, 14, This represents a case where it is started at 19... (Offset = 3), and a case where it is started at periodic tasks 5, 10, 15, 20.

  Here, for each pattern of offset = 0 to 4, it is examined how many regular tasks the timing at which data is received by the CAN corresponds to the start timing of the communication schedule in the LIN. That is, consider the value of Ncl.

  In the example of offset = 0, first, data is received by CAN at the fourth task (Ncl = 4), counting from the start timing of the first communication schedule (periodic task 1). In addition, data is received by CAN at the third task (Ncl = 3) from the start timing of the next communication schedule (periodic task 6). Furthermore, data is received by CAN at the second task (Ncl = 2), counting from the start timing of the next communication schedule (periodic task 11). Further, data is received by CAN at the first task (Ncl = 1) counting from the start timing of the next communication schedule (periodic task 16). Then, data is received by CAN at the fifth task (Ncl = 5) counting from the start timing of the same communication schedule (periodic task 16).

  In this way, the task at which the reception timing in CAN is viewed from the start timing of the LIN communication schedule (that is, the value of Ncl) is the period of the least common multiple of the CAN reception period and the LIN communication schedule period (that is, periodic In the example of offset = 0, Ncl can take a value of (1, 2, 3, 4, 5,).

  Similarly, in the case of the example of offset = 1 to 4, Ncl can take a value of (1, 2, 3, 4, 5,).

  In this way, in the example of FIG. 5, Ncl takes all values (patterns) of (1, 2, 3, 4, 5,), so all patterns appear in S20 in the precondition determination process of FIG. Thus, a negative determination is made (S20: NO), and it is determined to be invalid (a prerequisite for the period is not satisfied). In other words, it is determined that the effectiveness of synchronizing the CAN reception cycle and the LIN communication schedule cycle cannot be obtained.

  In order to examine the presence or absence of effectiveness, after receiving data at the CAN for the period of the least common multiple of the CAN reception period and the LIN communication schedule period (that is, the period of the periodic tasks 1 to 20), Consider the total time (total transfer time) required to transfer to the LIN (until the LIN communication schedule starts).

  In the example of offset = 0, data (0) to (3) are transferred to the LIN during the period of the regular tasks 1 to 20. Regarding data (0), the reception timing in CAN (not shown, immediately before the regular task 1) and the transfer timing to the LIN (communication schedule start timing, the regular task 1) are synchronized, There is no delay. Next, data (1) is received by the CAN at a timing corresponding to the regular task 4, and then the transfer is started by the regular task 6 (the LIN communication schedule is started). There is a delay of tasks (minutes of periodic task 5). The data (2) is received by the CAN at a timing corresponding to the regular task 8, and then the transfer is started by the regular task 11 (the LIN communication schedule is started). There is a delay of tasks 9 and 10). The data (3) is received by the CAN at a timing corresponding to the regular task 12, and then the transfer is started by the regular task 16 (the LIN communication schedule is started). There is a delay of tasks 13, 14, and 15). As described above, one task, two tasks, and three tasks are totaled to obtain a total transfer time of six tasks.

Similarly, considering the total transfer time for the case of offset = 1 to 4, all of them are for 6 tasks. In other words, the total transfer time cannot be reduced beyond this. In other words, the effectiveness of synchronizing the CAN reception cycle and the LIN communication schedule cycle cannot be obtained.
<Example of FIG. 6: CAN → LIN>
Next, FIG. 6 shows a case where the CAN reception cycle is four regular tasks and the LIN communication schedule cycle is two regular tasks. The data reception timing in CAN corresponds to the timings of the regular task 4, the regular task 8, the regular task 12, the regular task 16, the regular task 20,.

  There are two possible patterns of the start timing of the communication schedule in LIN relative to the reception timing in CAN in the example of FIG. 6 (offset = 0, 1).

  In the stage of offset = 0, 1, when the communication schedule in LIN is started by the periodic tasks 1, 3, 5, 7, 9,... (when offset = 0), the periodic tasks 2, 4 respectively. , 6, 8, 10... (When offset = 1).

  Here, in each of the patterns of offset = 0 and 1, it is examined how many times the task receives data in the CAN from the start timing of the communication schedule. That is, consider the value of Ncl.

  In the example of offset = 0, data is received by CAN at the second task (Ncl = 2), counting from the communication schedule start timing (periodic task 3). In addition, data is received by CAN at the second task (Ncl = 2) counting from the start timing of the communication schedule (periodic task 7). Similarly, data is received by CAN at the second task (Ncl = 2) counting from the start timing of the immediately preceding communication schedule.

  In this way, the task at which the reception timing in CAN is seen from the start timing of the LIN communication schedule (that is, the value of Ncl) is the period of the least common multiple of the CAN reception period and the LIN communication schedule period (that is, In the example of offset = 0, Ncl can take the value of (2).

  In the same way, when extracting for the example of offset = 1, Ncl can take the value of (1).

  Thus, in the example of FIG. 6, there is no example in which Ncl takes either (1) or (2) and takes all values (patterns) of (1, 2). In S20 in the process, an affirmative determination is made so that all patterns do not appear (S20: YES), and it is determined that the condition is valid (a prerequisite for the period is satisfied). In other words, it is determined that the effectiveness of synchronizing the CAN reception cycle and the LIN communication schedule cycle can be obtained.

  In order to examine the presence or absence of effectiveness, after receiving data at the CAN for the period of the least common multiple of the CAN reception period and the LIN communication schedule period (that is, the period of the periodic tasks 1 to 4), Consider the total time (total transfer time) required to transfer to the LIN (until the LIN communication schedule starts).

  In the example of offset = 0, data (0) is transferred to LIN during the period of the periodic tasks 1 to 4. When data (0) is received at CAN (not shown, but received immediately before periodic task 1), it is immediately transferred to LIN in the communication schedule started by periodic tasks 1 and 3 thereafter. Therefore, there is no transfer delay. From the above, the total transfer time is 0 tasks.

  In the example of offset = 1, data (0) is transferred to LIN during the period of the periodic tasks 1 to 4. When data (0) is received at the CAN, it is transferred to the LIN in accordance with a communication schedule started at the periodic tasks 2 and 4 after that. )). From the above, the total transfer time is one task.

  In the example of FIG. 6, in the case of offset = 1, the start timing of the communication schedule in LIN is adjusted so that the total transfer time is 0 tasks (synchronization between the CAN reception cycle and the LIN communication schedule cycle). ). In other words, the time required for transfer can be shortened (minimized), and the effectiveness of synchronization can be obtained.

FIGS. 7 to 9 show aspects of the pattern of the shift in the period between the LIN and the CAN when the data received in the LIN is transferred to the CAN as described above.
<Example of FIG. 7: LIN → CAN>
First, FIG. 7 shows a case where a CAN transmission cycle (hereinafter also referred to as a CAN transmission cycle) is for four regular tasks and a LIN communication schedule cycle is for six regular tasks. The data transmission timing in CAN corresponds to the timing of the regular task 4, the regular task 8, the regular task 12, the regular task 16, the regular task 20, the regular task 24,. .

  There are six possible patterns in the LIN communication schedule start timing for the CAN transmission timing in the example of FIG. 7 (offset = 0 to 5).

  In the stage of offset = 0 to 5, when the communication schedule in LIN is started with the periodic tasks 1, 7, 13, 19... (when offset = 0), the periodic tasks 2, 8, 14, When started with 20... (Offset = 1), when started with periodic tasks 3, 9, 15, 21... (When offset = 2), scheduled tasks 4, 10, 16,. When started with 22... (When offset = 3), when started with periodic tasks 5, 11, 17, 23... (When offset = 4), scheduled tasks 6, 12, 18,. This represents the case where the period starts at 24... (When offset = 5).

  Here, in each pattern of offset = 0 to 5, it is examined whether the timing at which data is transmitted by CAN corresponds to the number of periodic tasks as seen from the start timing of the communication schedule in LIN. In the following, the number representing this number is defined as Nlc.

  In the example of offset = 0, first, data is transmitted by CAN at the fourth task (Nlc = 4), counting from the start timing of the first communication schedule (periodic task 1).

  Subsequently, data is transmitted by CAN to the second task (Nlc = 2), counting from the start timing of the next communication schedule (periodic task 7). Furthermore, data is transmitted by CAN at the sixth task (Nlc = 6) counting from the start timing of the same communication schedule (periodic task 7).

  In this way, the task at which the CAN transmission timing is viewed from the start timing of the LIN communication schedule (that is, the value of Nlc) is the period of the least common multiple of the CAN transmission cycle and the LIN communication schedule cycle (that is, periodic Nlc can take the values (2, 4, 6,).

  Similarly, when extracting for the example where offset = 1, Nlc can take the value of (1, 3, 5).

  In the example of offset = 2, 4, Nlc takes the value of (2, 4, 6) as in offset = 0, and in the example of offset = 3, 5, Nlc is equal to (1 , 3, 5).

  Thus, in the example of FIG. 7, Nlc takes either (1, 3, 5,) or (2, 4, 6), and all of (1, 2, 3, 4, 5, 6) Since there is no example of taking a value (pattern), in S20 in the precondition determination process of FIG. 3, an affirmative determination is made that all patterns do not appear (S20: YES), and it is valid (the precondition for the period is satisfied). ). In other words, it is determined that the effectiveness of synchronizing the CAN transmission cycle and the LIN communication schedule cycle can be obtained.

  In order to examine whether or not there is effectiveness, CAN transmission is started from the start point of the LIN communication schedule for the least common multiple period of the CAN transmission period and the LIN communication schedule period (that is, the period of the periodic tasks 1 to 12). Consider the total time (total transfer time) until (transfer) occurs.

  In the example of offset = 0, the LIN communication schedule starts in the periodic task 1, and CAN transmission occurs at the timing corresponding to the periodic task 4, during which 3 tasks (periodic tasks 1, 2 and 3) I have time. In addition, the communication schedule starts at the regular task 7, and CAN transmission occurs at the timing corresponding to the regular tasks 8 and 12, and during that period, one task (for the regular task 7) and five tasks (for the regular task) Task 7, 8, 9, 10, 11 minutes). From the above, the total transfer time is 9 tasks by adding 3 tasks, 1 task, and 5 tasks.

  Similarly, considering the total transfer time in the case of the example of offset = 1 to 5, it becomes 6 tasks in the example of offset = 1, 9 tasks in the example of offset = 2, and 6 in the example of offset = 3. In the example of offset = 4, there are 9 tasks, and in the example of offset = 5, there are 6 tasks.

In the example of FIG. 7, in the case of offset = 0, 2, 4 example, the start timing of the communication schedule in LIN is adjusted so that the total transfer time is 6 tasks (CAN transmission cycle and LIN communication schedule) Should be synchronized with the period). In other words, the time required for transfer can be shortened (minimized), and the effectiveness of synchronization can be obtained.
<Example of FIG. 8: LIN → CAN>
Next, FIG. 8 shows a case where the CAN transmission cycle is four regular tasks and the LIN communication schedule cycle is five regular tasks. For example, it is assumed that the transmission timing of data in CAN corresponds to the timing of the regular task 4, the regular task 8, the regular task 12, the regular task 16, the regular task 20,.

  There are five possible patterns in the example of FIG. 8 for the start timing of the communication schedule in LIN relative to the transmission timing in CAN (offset = 0 to 4).

  In the stage of offset = 0 to 4, when the communication schedule in LIN is started by the periodic tasks 1, 6, 11, 16,... (when offset = 0), the periodic tasks 2, 7, 12, When started with 17 ... (when offset = 1), when started with periodic tasks 3, 8, 13, 18 ... (when offset = 2), with periodic tasks 4, 9, 14, This represents a case where it is started at 19... (Offset = 3), and a case where it is started at periodic tasks 5, 10, 15, 20.

  Here, in each pattern of offset = 0 to 4, it is examined how many regular tasks the timing at which data is transmitted by CAN corresponds to the start timing of the communication schedule in LIN. That is, consider Nlc.

  In the example of offset = 0, first, data is transmitted by CAN at the fourth task (Nlc = 4), counting from the start timing of the first communication schedule (periodic task 1). In addition, data is transmitted by CAN at the third task (Nlc = 3) from the start timing of the next communication schedule (periodic task 6). Further, data is transmitted by CAN at the second task (Nlc = 2), counting from the start timing of the next communication schedule (periodic task 11). Further, data is transmitted by CAN at the first task (Nlc = 1) counting from the start timing of the next communication schedule (periodic task 16). Then, data is transmitted by CAN at the fifth task (Nlc = 5) counting from the start timing of the same communication schedule (periodic task 16).

  In this way, the task at which the CAN transmission timing is viewed from the start timing of the LIN communication schedule (that is, the value of Nlc) is the period of the least common multiple of the CAN transmission cycle and the LIN communication schedule cycle (that is, periodic Nlc can take the values (1, 2, 3, 4, 5).

  Similarly, in the case of the example of offset = 1 to 4, Nlc can take a value of (1, 2, 3, 4, 5).

  In this way, in the example of FIG. 8, since Nlc takes all values (patterns) of (1, 2, 3, 4, 5,), all patterns appear in S20 in the precondition determination process of FIG. Thus, a negative determination is made (S20: NO), and it is determined to be invalid (a prerequisite for the period is not satisfied). In other words, it is determined that the effectiveness of synchronizing the CAN transmission cycle and the LIN communication schedule cycle cannot be obtained.

  In order to examine whether or not there is effectiveness, CAN transmission is started from the start point of the LIN communication schedule for the period of the least common multiple of the CAN transmission period and the LIN communication schedule period (that is, the period of the periodic tasks 1 to 20). Consider the total time (total transfer time) until (transfer) occurs.

  In the example of offset = 0, the LIN communication schedule is activated in the periodic task 1, and CAN transmission occurs at a timing corresponding to the periodic task 4, during which 3 tasks (periodic tasks 1, 2 and 3) I have time. In addition, the communication schedule is activated by the regular task 6, CAN transmission occurs at a timing corresponding to the regular task 8, and there are two tasks (periods 6 and 7) in the meantime. In addition, a communication schedule is activated by the periodic task 11, CAN transmission occurs at a timing corresponding to the periodic task 12, and there is a time corresponding to one task (for the periodic task 11). In addition, the communication schedule is activated by the regular task 16, and CAN transmission occurs at the timings corresponding to the regular tasks 16 and 20, respectively, for 0 task and 4 tasks (periodic tasks 16, 17, 18, There are 19 minutes). From the above, the total transfer time is 10 tasks by summing 3 tasks, 2 tasks, 1 task, and 4 tasks.

Similarly, if the total transfer time is examined for the cases of offset = 1 to 4, all are 10 tasks. In other words, the total transfer time cannot be reduced beyond this. In other words, the effectiveness of synchronizing the CAN reception cycle and the LIN communication schedule cycle cannot be obtained.
<Example of FIG. 9: LIN → CAN>
Next, FIG. 9 shows a case where the CAN transmission cycle is four regular tasks and the LIN communication schedule cycle is two regular tasks. It is assumed that the data transmission timing in the CAN corresponds to the timings of the regular task 4, the regular task 8, the regular task 12, the regular task 16, the regular task 20,.

  There are two possible patterns in the LIN communication schedule start timing for the CAN transmission timing in the example of FIG. 9 (offset = 0, 1).

  In the stage of offset = 0, 1, when the communication schedule in LIN is started by the periodic tasks 1, 3, 5, 7, 9,... (when offset = 0), the periodic tasks 2, 4 respectively. , 6, 8, 10... (When offset = 1).

  Here, in each pattern of offset = 0, 1, it is examined how many regular tasks the timing at which data is transmitted by CAN corresponds to the start timing of the communication schedule in LIN. That is, consider Nlc.

  In the example of offset = 0, data is transmitted by CAN at the second task (Nlc = 2) counting from the communication schedule start timing (periodic task 3). In addition, data is transmitted by CAN at the second task (Nlc = 2) counting from the start timing of the communication schedule (periodic task 7). Similarly, data is transmitted by CAN to the second task (Nlc = 2) counting from the start timing of the immediately preceding communication schedule.

  In this way, the task at which the transmission timing in the CAN is viewed from the start timing of the LIN communication schedule (that is, the value of Nlc) is the period of the least common multiple of the CAN transmission cycle and the LIN communication schedule cycle (that is, When extracting for the period of the periodic tasks 1 to 4, Nlc can take the value of (2).

  Similarly, when extracting for the example of offset = 1, Nlc can take the value of (1).

  Thus, in the example of FIG. 9, there is no example where Nlc takes either (1) or (2) and takes all values (patterns) of (1, 2). In S20 in the process, an affirmative determination is made so that all patterns do not appear (S20: YES), and it is determined that the condition is valid (a prerequisite for the period is satisfied). In other words, it is determined that the effectiveness of synchronizing the CAN reception cycle and the LIN communication schedule cycle can be obtained.

  In order to examine whether or not it is effective, CAN transmission is started from the start point of the LIN communication schedule for the period of the least common multiple of the CAN transmission period and the LIN communication schedule period (that is, the period of the periodic tasks 1 to 4). Consider the total time (total transfer time) until (transfer) occurs.

  In the example of offset = 0, CAN transmission occurs at a timing corresponding to the periodic task 4 after the LIN communication schedule is started in the periodic task 3, and there is a time corresponding to one task (for the periodic task 3). . From the above, the total transfer time is one task.

  In the example of offset = 1, CAN transmission occurs at a timing corresponding to the periodic task 4 after the LIN communication schedule starts in the periodic task 4, and there is no time lag during that period. From the above, the total transfer time is 0 tasks.

  In the example of FIG. 9, in the case of offset = 0, the start timing of the communication schedule in the LIN is adjusted so that the total transfer time is 0 tasks (synchronization between the CAN reception cycle and the LIN communication schedule cycle). ). In other words, the time required for transfer can be shortened (minimized), and the effectiveness of synchronization can be obtained.

  Next, the process of FIG. 10 will be described. As described above, the communication schedule adjustment process of FIG. 10 is a process executed by the communication synchronization apparatus 202 of the ECU 101 at predetermined intervals.

  In the communication schedule adjustment process of FIG. 10, the communication synchronization apparatus 202 first measures the transmission time (transmission timing) or reception time (reception timing) of the data in the CAN for the data to be transferred, and transmits the transmission period or Calculate the reception period. Here, the transmission time (transmission timing) is measured based on the transmission completion notification from the CAN communication driver 203, and the reception time (reception timing) is measured based on the reception completion notification from the CAN communication driver 203. The data transmission time (transmission timing) is measured at the time of data transfer from LIN to CAN, and the data reception time (reception timing) is measured at the time of data transfer from CAN to LIN.

  Next, the process proceeds to S120, and it is determined whether or not the start timing of the LIN communication schedule can be adjusted. Here, the determination is made based on whether or not both a transfer start condition and an interval condition, which will be described later, are satisfied, and if it is determined that both are satisfied, it is determined that adjustment is possible.

  The transfer start condition is a condition for confirming that data transfer has been started reliably. For example, (a) the ignition switch (not shown) of the vehicle is in an ON state, (b) the first reception of data to be transferred after activation is completed, and (c) the communication schedule is started at LIN. This is listed as a requirement for establishing the transfer start condition. In the present embodiment, it is determined that the transfer start condition is satisfied when any of the requirements (A) to (C) is satisfied.

  The interval condition is a condition for confirming whether or not a predetermined time (hereinafter also referred to as interval time) has elapsed since the previous adjustment of the start timing of the communication schedule. That is, when the interval time has elapsed since the previous adjustment of the start timing of the communication schedule, it is determined that the interval condition is satisfied. This is intended to prevent frequent occurrence of adjustment of the communication schedule. This is because if the communication schedule is frequently adjusted, data transfer may become inefficient or the communication schedule may not be started in LIN. The interval time can be set to, for example, a time obtained by multiplying the least common multiple of the CAN transmission / reception cycle (Tc) and the LIN communication schedule cycle (Tl) by the time of the regular task.

  If it is determined in S120 that adjustment is possible (S120: YES), the process proceeds to S130. If it is determined in S120 that adjustment is not possible (S120: NO), the process returns to S110 again. In S110 returned from S120, the transmission time (transmission timing) or the reception time (reception timing) is measured for other transfer target data. Then, the subsequent processing is continued.

  In S130, it is determined whether or not the CAN transmission / reception cycle for the data to be transferred and the LIN communication schedule cycle are the same. If it is determined that they are the same (S130 :: YES), the process proceeds to S140 and is not the same. (S130: NO), the process proceeds to S150. If the CAN transmission / reception cycle and the LIN communication schedule cycle are the same, the CAN transmission / reception timing and the LIN communication schedule start timing completely coincide with each other by synchronizing their start timings. On the other hand, when the CAN transmission / reception cycle and the LIN communication schedule cycle are not the same, adjustment may not be necessary as shown in the examples of FIGS.

  In S140, the number of tasks to be offset is calculated. Specifically, the adjustment amount of the start timing of the LIN communication schedule is calculated.

  Here, at the time of data transfer from CAN to LIN, if the LIN communication schedule is started in the periodic task immediately after reception at CAN, the delay in data transfer from CAN to LIN is minimized. Therefore, in the process of S140, the number of tasks offset at the time of data transfer from CAN to LIN is calculated by the following equations (a) and (b).

OFScl = n1 · Tc− (Tl−Ncl) (a) Expression n1 = CEILING ((Tl−Ncl) / Tc) (b) Expression OFScl: Number of tasks to be offset CEILING: Round up the decimal point Function Tl: Communication schedule cycle (number of periodic tasks)
Tc: CAN transmission / reception cycle (number of periodic tasks)
Next, when transferring data from LIN to CAN, the delay in data transfer from LIN to CAN is minimized when the LIN communication schedule ends in a periodic task immediately before transmission in CAN occurs. Therefore, in the process of S140, the number of tasks offset at the time of data transfer from LIN to CAN is calculated by the following equations (c) and (d).

OFScl = (n2 · Tc−1) − (T1−Nlc) (c) Expression n2 = 1 + FLOOR ((T1−Nlc) / Tc) (d) Expression OFScl: Number of tasks to be offset FLOOR: Decimal point After the truncation function S140, the process proceeds to S180, and a communication schedule adjustment instruction is output to the LIN communication scheduler 206 based on the number of tasks to be offset calculated in S140. And it returns to S110 again. When the number of tasks to be offset is zero, no adjustment instruction is notified.

  On the other hand, in S150, where a negative determination is made in S130 and the process proceeds, the above-described Ncl or Nlc appearance pattern is determined based on the measurement result in S110. For example, the appearance pattern can be discriminated by measuring the value of Ncl or Nlc during the aforementioned interval time.

  After S150, the process proceeds to S160, and it is determined whether or not adjustment of the start timing of the communication schedule is necessary. In the present embodiment, the ECU 101 stores in advance information indicating whether adjustment of the start timing of the communication schedule is effective for each Ncl or Nlc appearance pattern (hereinafter also referred to as adjustment necessity information). Then, the necessity of adjustment is determined by referring to the adjustment necessity information. For example, in the example of FIG. 4, when the appearance pattern of Ncl is (4, 2, 6) as in the case of offset = 0, 2, 4, 4, it is set that no adjustment is necessary, and offset = When the appearance pattern of Nlc is (3, 1, 5) as in the case of 1, 3, 5, the adjustment is set as necessary.

  If it is determined in S160 that no adjustment is necessary (S160: NO), the process returns to S110 again. On the other hand, if it is determined in S160 that adjustment is required (S160: YES), the process proceeds to S170. In S170, the number of tasks to be offset is calculated. In this embodiment, the number of tasks to be offset in order to minimize the total transfer time is stored in the ECU 101 in advance for each pattern as shown in FIGS. Then, the number of tasks to be offset can be calculated with reference to the information. For example, in the example of FIG. 4, in the case of offset = 1, if the start timing of the communication schedule is offset by one task (accelerated), it matches the example of offset = 0, and the total transfer time is two tasks. Decrease to. Therefore, for example, when the pattern of offset = 1 in FIG. 4 is taken, information that the number of tasks to be offset is 1 is previously stored.

  After S170, the process proceeds to S180, and a communication schedule start timing adjustment instruction is output to the LIN communication scheduler 206.

  Next, the process of FIG. 11 will be described. As described above, the communication schedule start process of FIG. 11 is executed by the LIN communication scheduler 206 of the ECU 101 every predetermined period.

  In the communication schedule start process of FIG. 11, the LIN communication scheduler 206 first determines whether or not a communication schedule adjustment instruction has been received from the communication synchronization apparatus 202 in S210. The communication schedule adjustment instruction is output from the communication synchronization apparatus 202 in S180 of FIG.

  If it is determined in S210 that an adjustment instruction has not been received (S210: NO), the process proceeds to S230, and a communication schedule is started (activated) at a preset timing (periodic task). Thereafter, the process is terminated.

  On the other hand, if it is determined in S210 that an adjustment instruction has been received (S210: YES), the process proceeds to S220, and it is determined whether or not a time corresponding to the number of tasks to be offset has elapsed from a preset timing. Information on the number of tasks to be offset is included in the adjustment instruction from the communication synchronization apparatus 202.

  If it is determined in S220 that the time corresponding to the number of tasks to be offset has not elapsed (S220: NO), the processing in S220 is repeated again (waiting until the time corresponding to the number of tasks to be offset has elapsed). On the other hand, when it is determined in S220 that the time corresponding to the number of tasks to be offset has elapsed (S220: YES), the synchronization timing is determined, and the process proceeds to S230 to start the communication schedule. Thereafter, the process is terminated.

  As described above, in this embodiment, the ECU 101 connected to the CAN and the LIN determines the communication schedule in the LIN so that the time required for the data transfer between the CAN and the LIN is minimized. The start timing is adjusted. Specifically, at the time of data transfer from CAN to LIN, the communication schedule is started at LIN so that the reception timing at CAN and the transmission timing at LIN are synchronized. For this reason, data received by the CAN is immediately transferred to the LIN. Further, at the time of data transfer from the LIN to the CAN, the communication schedule is started at the LIN so that the transmission timing at the CAN and the reception timing at the LIN are synchronized. For this reason, the data received by the LIN is immediately transferred to the CAN.

  In this embodiment, the occurrence of a time lag can be suppressed during data transfer from CAN to LIN or data transfer from LIN to CAN. According to this, data transfer from CAN to LIN or data transfer from LIN to CAN becomes smooth, which can contribute to improvement of transfer efficiency.

In the present embodiment, the process of S110 corresponds to the detection unit, and the processes of S140 and S170 correspond to the calculation unit.
[Modification 1]
Next, Modification 1 of the above embodiment will be described with reference to FIGS. 12 shows a mode of data transfer from CAN to LIN, and FIG. 13 shows a mode of data transfer from LIN to CAN.

  In the first modification, as shown in FIG. 12, a LIN communication schedule is configured by transmitting and receiving four frames. The LIN communication schedule cycle is 4 tasks. Specifically, a communication schedule is configured in the order of transmission of data A, reception of data B, reception of data C, transmission of data D, transmission of data A, reception of data B, reception of data C, data Each transmission of D occupies one periodic task.

  In FIG. 12, when the CAN reception cycle and the LIN communication schedule cycle are synchronized by the processes of FIGS. 10 and 11, the period immediately after the CAN reception timing is set as shown in the LIN communication schedule pattern (1). The task starts the LIN communication schedule. As described above, since the communication schedule in LIN is configured in the order of transmission of data A, reception of data B, reception of data C, and transmission of data D, data A can be transferred from CAN to LIN in the shortest time.

  For example, when the data D is to be transferred in the shortest time, the start timing of the LIN communication schedule is offset (delayed) by one task as shown in the LIN communication schedule pattern (2). As a result, the task “transmission of data D” occurs immediately after reception in CAN, so that data D can be transferred from CAN to LIN in the shortest time.

  Next, in FIG. 13, when the CAN transmission cycle and the LIN communication schedule are synchronized by the processing of FIGS. 10 and 11, the data C is transferred from the LIN to the CAN as shown in the LIN communication schedule pattern (3). Can be transferred in the shortest time.

  Here, for example, when data B is to be transferred in the shortest time, as shown in the LIN communication schedule pattern (4), the start timing of the LIN communication schedule is offset (delayed) by one task. As a result, the task of “reception of data B” occurs immediately before transmission in CAN, so that data B can be transferred from LIN to CAN in the shortest time.

  As described above, in the first modification, the CAN transmission / reception cycle and the LIN communication schedule cycle are synchronized, and an offset is additionally executed according to the data to be transferred in the shortest time.

  According to this, when the LIN communication schedule is configured by transmitting and receiving a plurality of frames, the transfer time can be minimized for a desired frame (desired data) among the plurality of frames. . According to this, it is possible to more reliably realize transfer of desired data in the shortest time.

The data to be transferred in the shortest time may be set in advance, but the data to be transferred in the shortest time may be dynamically changed as in Modification 2 shown below. .
[Modification 2]
Modification 2 will be described with reference to FIG. The second modification can be applied even when it is determined that the precondition determination process in FIG. 3 is invalid (the precondition regarding the period is not satisfied).

  In the example of FIG. 14, the LIN communication schedule is configured in the order of transmission / reception of transmission frame α (transmission / reception of data A) and transmission / reception of transmission frame β (transmission / reception of data B). The LIN communication schedule cycle is 6 tasks. Specifically, transmission / reception of the transmission frame α and transmission / reception of the transmission frame β occupy three tasks.

  On the other hand, in CAN, it is assumed that CAN frame 1 (data A) and CAN frame 2 (data B) are received at a period corresponding to 6 tasks. For example, CAN frame 1 is received immediately after regular task 2 and immediately after regular task 8, and CAN frame 2 is received, for example, immediately after regular task 3 and immediately after regular task 9, respectively.

  When data is transferred in synchronization with CAN frame 1 (data A) (when data A is transferred in the shortest time), as shown in LIN communication schedule pattern (5) in FIG. The start timing of the communication schedule is adjusted so that the task starts the communication schedule (the processes of FIGS. 10 and 11). As a result, the data A can be transferred in the shortest time.

  On the other hand, for data B, in the LIN communication schedule pattern (5), CAN frame 2 (data B) is received by CAN immediately after periodic task 3, and then transmission frame β (data There will be a delay in the transfer until B) is sent to LIN.

  Here, in the second modification, when data B is to be transferred in the shortest time, a synchronization instruction for synchronizing the data B with the CAN frame 2 storing the data B (hereinafter also referred to as a data B synchronization instruction). ) Is stored.

  In the CAN frame 2, when the data B synchronization instruction is stored together with the data B, the start timing of the LIN communication schedule is adjusted so that the data B can be transferred in the shortest time. Specifically, as shown in the LIN communication schedule pattern (6), the start timing of the LIN communication schedule so that the task “transmission / reception of transmission frame β” occurs in the periodic task immediately after reception of the CAN frame 2. Is offset.

  The offset procedure will be described more specifically. First, the start timing of the LIN communication schedule is adjusted so that the reception period of the CAN frame 2 and the LIN communication schedule period are synchronized (the process of FIG. 10). In addition, the start timing of the LIN communication schedule is subtracted (accelerated) by the time required for transmission / reception of the transmission frame α (three tasks in the example of FIG. 14). According to such a procedure, a task “transmission / reception of transmission frame β” occurs in a periodic task immediately after reception of CAN frame 2, and data B can be transferred in the shortest time.

Although the case of data transfer from CAN to LIN has been described here, the same method can be used for data transfer from LIN to CAN.
[Modification 3]
Next, Modification 3 will be described with reference to FIG.

  The example of FIG. 15 represents a mode in which data is transferred from CAN to LIN. The CAN reception cycle is 4 tasks. For example, it is assumed that reception occurs at the CAN immediately after the regular task 4, immediately after the regular task 8, just after the regular task 12, and just after the regular task 16.

  The LIN communication schedule is configured in the order of transmission / reception of transmission frames and transmission / reception of reception frames, and the LIN communication schedule cycle is assumed to be four tasks.

  As shown in the example of “before synchronization” in FIG. 15, if the start timing of the LIN communication schedule is regular tasks 3, 7, 11, 15,... In order to transfer the processed data to the LIN in the shortest time, it is necessary to wait for 2/4 period (that is, two tasks) for the start of the LIN communication schedule. Specifically, the LIN communication schedule needs to be started by the periodic tasks 5, 9, 13.

  In this third modification, the start timing of the LIN communication schedule is gradually shifted to synchronize the CAN reception cycle and the LIN communication schedule cycle.

  First, as shown in the stage of “synchronization process (first time)” in FIG. 15, the start timing of the LIN communication schedule is delayed by one task in the first synchronization process (waits for one task). Specifically, the communication schedule originally started by the periodic task 7 is started by the periodic task 8 after one task.

  Subsequently, as shown in the stage of “synchronization process (second time)” in FIG. 15, the start timing of the LIN communication schedule is further delayed by one task in the second synchronization process. Thereafter, although not shown, the communication schedule is started at a cycle of 4 tasks.

  As a result, the start timing of the LIN communication schedule matches the target timing. That is, the LIN communication schedule starts with the periodic tasks 13, 17,. In other words, the LIN communication schedule is started in the periodic task immediately after the occurrence of reception in the CAN.

  As in the third modification example, the adjustment amount per time is adjusted by gradually adjusting the start timing of the LIN communication schedule in multiple synchronization processes instead of adjusting the start timing in one synchronization process. (Offset amount) can be reduced. In other words, the start timing of the communication schedule can be prevented from being significantly changed by one adjustment. If the start timing of the communication schedule is changed significantly by a single adjustment, a device or the like that cannot recognize the change on the network (for example, the control device 103 in FIG. 1) has a communication failure or communication interruption. Although it may be erroneously determined that it has occurred, according to the configuration of the third modification, it is possible to avoid the occurrence of such a problem.

Although the case of data transfer from CAN to LIN has been described here, the same method can be used for data transfer from LIN to CAN.
[Modification 4]
Next, Modification 4 will be described with reference to FIG. Note that the fourth modification can be applied even when it is determined that the precondition determination process in FIG. 3 is invalid (the precondition regarding the period is not satisfied).

  The example of FIG. 16 represents an aspect in the case where data is transferred from CAN to LIN. In CAN, it is assumed that CAN frame 1 (data A) and CAN frame 2 (data B) are received at a period corresponding to 6 tasks. More specifically, the CAN frame 1 is received, for example, immediately after the regular task 2 and the timing immediately after the regular task 8, and the CAN frame 2 is received, for example, immediately after the regular task 5, and immediately after the regular task 11. Received respectively.

  The LIN communication schedule is configured in the order of transmission / reception of transmission frames (transmission / reception of data A and data B) and transmission / reception of reception frames (transmission / reception of data C), and the LIN communication schedule cycle is assumed to be 6 tasks.

  In the example of the LIN communication schedule pattern (7) in FIG. 16, the data A is configured to be transferred in the shortest time. Specifically, the LIN communication schedule is started by a periodic task immediately after the reception of CAN frame 1 (data A). On the other hand, for data B, for example, CAN frame 2 (data B) is received by CAN immediately after periodic task 5, and then transmission frames (data A and data B) are transmitted by LIN in periodic task 9. There will be a delay in the transfer until it is transmitted and received.

  In the fourth modification, for example, when data B is urgently transferred, the data B is data to be urgently transferred to the CAN frame 2 storing the data B (hereinafter also referred to as emergency data). Is stored.

  In the CAN frame 2, when emergency data information is stored together with the data B, the start timing of the LIN communication schedule is adjusted so that the data B can be transferred in an emergency (shortest). Specifically, the current communication schedule is canceled and a new communication schedule is started so that the data B can be transferred immediately as shown in the LIN communication schedule pattern (8).

  For example, in the example of FIG. 16, after the CAN frame 2 is received by the CAN, the task “transmission / reception of received frame (transmission / reception of data C)” that should have occurred in the periodic task 6 is canceled ( The current communication schedule is canceled), and a new communication schedule is started in the periodic task 7. As a result, the CAN frame 2 (data B) received by the CAN can be immediately transferred to the LIN.

  According to the fourth modification, the desired data B can be immediately transferred from the CAN to the LIN without waiting for the end of transmission / reception of the received frame (transmission / reception of the data C).

Although the case of data transfer from CAN to LIN has been described here, the same applies to the case of data transfer from LIN to CAN. Information indicating emergency data may be stored in the received frame in the LIN. When information indicating emergency data is stored in the received frame, the start timing of the communication schedule in the LIN may be adjusted so that the reception timing of the received frame is synchronized with the transmission timing in the CAN.
[Modification 5]
Next, Modification 5 will be described with reference to FIGS. 17 and 18.

  In the embodiment described above, the start timing of the communication schedule in the LIN is adjusted so that certain data is transferred in the shortest time. In the fifth modification, a plurality of data is targeted, and the data are transferred as much as possible without delay.

  In the example of FIG. 17, it is assumed that the CAN frame 1, CAN frame 2, and CAN frame 3 are received as data to be transferred by the CAN. Note that the reception cycle of CAN frame 1 and CAN frame 2 is 16 [ms], and the reception cycle of CAN frame 3 is 24 [ms].

  Here, in the present modification, the ECU 101 executes the following process.

  First, the reception timing of each frame is detected for each section having a predetermined length. Here, the reception timing is detected with a period of 16 [msec] as one section.

  Then, the detection results of a plurality of sections are overlaid. Specifically, the detection results in the sections of 1 to 16 [msec], 16 to 32 [msec],... N × 16 to (n + 1) × 16 [msec] are overlapped.

  Then, a reception timing histogram is created.

  A histogram in this case is shown in FIG. 1 to 16 [msec], 16 to 32 [msec],... N × 16 to (n + 1) × 16 [msec] are overlapped, and first, CAN frame 1, CAN frame 2, and CAN For each frame 3, a reception timing histogram is created.

  Since CAN frame 1 is received at the timing of periodic task [n × 16 + 1], the peak reception timing of CAN frame 1 appears at the location of periodic task [n × 16 + 1].

  Similarly, since CAN frame 2 is received at the timing of regular task [n × 16 + 3], the peak reception timing of CAN frame 1 appears at the location of regular task [n × 16 + 3].

  The CAN frame 3 is received at the timing of the periodic task [n × 16 + 4] and the periodic task [n × 16 + 12], and the peak reception timing of the CAN frame 3 is the periodic task [n × 16 + 4] and the periodic task [n. X16 + 12].

  Next, all the histograms of CAN frame 1, CAN frame 2, and CAN frame 3 are combined. Then, as shown in FIG. 18, it can be seen that peaks appear at the timing of the regular task [n × 16 + 2] or the regular task [n × 16 + 3] and the regular task [n × 16 + 12]. large. That is, it can be seen that the CAN reception is most frequently generated at the timing of the regular task [n × 16 + 2] or the regular task [n × 16 + 3].

  Therefore, in the fifth modification, the start timing of the LIN communication schedule is adjusted so that the LIN communication schedule is started in the periodic task immediately after the peak.

  According to the fifth modification, it is possible to prevent the transfer delay from increasing on average for a plurality of data.

Although the case of data transfer from CAN to LIN has been described here, the same applies to the case of data transfer from LIN to CAN. In this case, a transmission timing histogram in CAN is created, and the start timing of the communication schedule may be adjusted so that the peak transmission timing in CAN and the reception timing in LIN are synchronized.
[Modification 6]
In the above embodiment, for data that is actually transmitted and received by CAN and LIN, the data to be transferred is determined in the shortest time, the transmission / reception timing of the data is detected, and the transmission / reception timing at CAN and the transmission / reception timing at LIN (Schedule start timing).

  On the other hand, instead of data actually transmitted and received by CAN and LIN, the start timing of the LIN communication schedule may be adjusted using dummy data.

  Specifically, dummy data may be transmitted / received by CAN via CAN communication driver 203 (see FIG. 2), and dummy data may be transmitted / received by LIN via LIN communication driver 205. The transmission / reception timing of dummy data can be set in advance. For example, by virtually constructing the vehicle network 100 in an external computer and executing an operation simulation, the transmission / reception timing of predetermined data can be determined, and the transmission / reception timing thus determined is preliminarily stored in the ECU 101. Set it up.

  According to the configuration in which the start timing of the communication schedule in the LIN is adjusted using the dummy data, for example, when the ECU 101 has to frequently transfer data or during the control of the control device 103, the LIN communication schedule For example, it is possible to adjust the start timing of the LIN communication schedule during the initialization process immediately after startup. According to this, it can contribute by efficiency improvement of data transfer.

In this case, for example, the CAN communication driver 203 and the LIN communication driver 205 correspond to dummy data transmission / reception means.
[Modification 7]
In the above embodiment, the ECU 101 may send information indicating that the start timing is changed to the LIN in accordance with the change of the start timing of the communication schedule in the LIN.

  According to this, for example, the communication device on the LIN can recognize the change of the start timing of the communication schedule. For this reason, it is possible to prevent the communication apparatus on the LIN from being erroneously determined to have an abnormality in communication due to the change or erroneously determined to be a communication interruption.

  As mentioned above, although one Embodiment of this invention was described, this invention is not limited to the said embodiment, A various form can be taken within the technical scope of this invention.

  For example, in the above embodiment, the configuration of the vehicle network 100 is an example, and for example, a configuration in which a plurality of ECUs and control targets are further connected may be used.

  In the determination of the transfer start condition of the above embodiment, (b) the ignition switch (not shown) of the vehicle is in the ON state, (b) the initial reception of the data to be transferred after activation is completed, and (c) LIN. It may be determined that the transfer start condition is satisfied when all of the communication schedule has started.

  Further, in the above-mentioned modification 5, the length of the predetermined section is 16 [msec], but any length may be used.

  In addition, you may combine the structure of the said modifications 1-7.

DESCRIPTION OF SYMBOLS 100 ... Vehicle network, 202 ... Communication synchronization apparatus, 203 ... CAN communication driver, 204 ... Database, 205 ... LIN communication driver, 206 ... LIN communication scheduler, 207 ... CAN communication line, 208 ... LIN communication line, 300 ... Vehicle network, DESCRIPTION OF SYMBOLS 303 ... Control apparatus, 304 ... Wire harness, 306 ... Wire harness, 307 ... CAN communication line, 400 ... Vehicle network, 403 ... Control apparatus, 404 ... LIN communication line, 405 ... CAN communication line

Claims (8)

A gateway device connected to a plurality of networks in a vehicle and transferring data between the networks, at least,
Data transfer between a network (hereinafter referred to as a first network) in which data is transmitted and received and another network (hereinafter referred to as a second network) in accordance with a communication schedule periodically started by the master device In the gateway device that performs
It is configured to operate as a master device and start a communication schedule,
When the first network and the second network are monitored and transmission target data is transmitted / received on the first network, and the transmission target data is transmitted / received on the second network Detection means for detecting transmission / reception timing;
Based on the detection result of the detection means, the transmission timing on the first network and the reception timing on the second network are synchronized, or the reception timing on the first network and the second network Calculating means for calculating the start timing of the communication schedule for synchronizing with the transmission timing of
A gateway apparatus, wherein a communication schedule is activated at an activation timing calculated by a calculation means. The gateway device according to claim 1,
There are multiple types of data to be transferred.
The multiple types of data contain information representing importance,
The detecting means is configured to detect transmission / reception timing on the first network and transmission / reception timing on the second network for data having the highest importance among a plurality of types of data. Gateway device. The gateway device according to claim 1,
There are multiple types of data to be transferred.
The calculating means calculates a frequency distribution of transmission / reception timing of the data to be transferred on the second network, and based on the frequency distribution, the transmission timing on the first network and the reception timing on the second network. In order to synchronize with the peak, or to calculate the start timing of the communication schedule to synchronize the reception timing on the first network and the transmission peak on the second network. A gateway device characterized. The gateway device according to any one of claims 1 to 3,
A gateway apparatus characterized by gradually changing the activation timing of the communication schedule by a predetermined amount to match the activation timing calculated by the calculation means. The gateway device according to any one of claims 1 to 3,
The data to be transferred contains information indicating the urgency,
Based on the information indicating the degree of urgency, it is determined whether or not an emergency is required, and if it is determined that an emergency is required, the communication schedule that is currently operating is canceled and a new communication schedule is started. Gateway device. The gateway device according to any one of claims 1 to 5,
A gateway apparatus characterized in that, after changing the activation timing of the communication schedule, the change is not made again until a predetermined time elapses. The gateway device according to claim 1,
In place of predetermined transfer target data, dummy data transmission / reception means for transmitting / receiving dummy data at the same timing as the transfer target data is provided,
The detection means detects transmission / reception timing when dummy data is transmitted / received on the first network, and transmission / reception timing when dummy data is transmitted / received on the second network,
For the dummy data, the calculation means synchronizes the transmission timing on the first network and the reception timing on the second network, or the reception timing on the first network and the second network. The gateway apparatus characterized by calculating the starting timing of the communication schedule for synchronizing with the transmission timing of. The gateway device according to any one of claims 1 to 7,
A gateway apparatus characterized in that information representing changing the activation timing of a communication schedule is sent to both or one of the first network and the second network.