JP2009113618A - Electric power steering device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an electric power steering device accurately detecting an abnormal execution state by severely monitoring the execution state of a control program. <P>SOLUTION: The electric power steering device includes a steering assistant electric motor; and a control part for executing the steering control of the vehicle by driving and controlling the steering assistant electric motor by sequentially executing a plurality of tasks. The control part measures completion times of the respective tasks and performs the detection of the abnormal execution state of the program based on the completion times of the respective measured tasks. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an electric power steering apparatus that detects an abnormal execution state of a program for performing steering control.

  In order to reduce the steering force of vehicles such as passenger cars and trucks, there is a so-called electric power steering (EPS) device that assists steering by a steering assist motor. In the EPS device, the driving force of the steering assist motor is applied to the steering shaft or the rack shaft by a transmission mechanism such as a gear or a belt via a speed reducer.

  Since the EPS device plays an important role of vehicle steering, it is particularly necessary to consider safety and reliability among the parts constituting the automobile. For this reason, the control unit control program that controls the EPS device can be dealt with before it reaches a state where abnormal steering assistance is executed. In such a case, a method for detecting an abnormal execution state of the program such as a program runaway state is required.

  In order to detect such an abnormal execution state of a program, WDT (Watch Dog Timer) is conventionally used. WDT is a timer that is reset from the program so that a timeout does not occur, and detects an abnormal execution state of the program by using the timeout when the application does not operate correctly and WDT is not reset. be able to. For example, according to Patent Document 1, a technique for dealing with a system failure using WDT is introduced.

JP 2002-251300 A

  However, in the conventional method for detecting an abnormal execution state of a program using WDT, when the WDT is periodically reset even though the program is in an abnormal execution state, Abnormality is not detected. That is, the technique using WDT has a problem that only a limited part of the abnormality is detected.

  The present invention has been made in view of the above, and it is an object of the present invention to provide an electric power steering apparatus capable of accurately monitoring an execution state of a control program and accurately detecting an abnormal execution state. And

  In order to solve the above-described problems and achieve the object, the present invention relates to an electric motor for assisting steering and a drive control of the electric motor for assisting steering by sequentially executing a plurality of tasks. In the electric power steering apparatus including a control unit that executes steering control, the control unit measures an end time of each task, and detects an abnormal execution state of the program based on the measured end time of each task. It is characterized by performing.

  According to a preferred aspect of the present invention, the control unit determines the execution order of tasks based on the measured end time of each task, and the determined execution order is executed in a predetermined order. If not, the program is detected as an abnormal execution state.

  Further, according to a preferred aspect of the present invention, the control unit determines whether or not a plurality of required tasks are all executed during a predetermined process cycle based on the measured end time of each task, It is characterized in that an abnormal execution state of a program is detected when all of a plurality of required tasks are not executed during a predetermined process cycle.

  Further, according to a preferred aspect of the present invention, the control unit determines whether each task is completed within a predetermined time based on the measured end time of each task, and determines the predetermined task for which each task is designed. If the program is not finished in time, it is detected as an abnormal execution state of the program.

  According to this invention, since the end time of each task is measured and the abnormal execution state of the program is detected based on the measured end time of each task, the execution state of the control program can be strictly monitored. This makes it possible to accurately detect an abnormal execution state of a program that cannot be detected by the watchdog timer.

  Embodiments of an electric power steering (EPS) device according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

  FIG. 1 is a diagram illustrating a general configuration of an electric power steering (EPS) device 100. In FIG. 1, a column shaft 2 of a steering handle 1 is connected to a tie rod 6 of a steering wheel via a reduction gear 3, universal joints 4a and 4b, and a pinion rack mechanism 5. The column shaft 2 is provided with a torque sensor 10 that detects the steering torque T of the steering handle 1, and a steering assist motor 20 that assists the steering force of the steering handle 1 is connected to the column shaft via the reduction gear 3. 2 is connected. Here, the steering assist motor 20 is, for example, a brushless motor or a brush motor. The control unit (ECU) 30 that controls the electric power steering apparatus is supplied with electric power from the battery 14 via the built-in power supply relay 13 and is supplied with an ignition signal from the ignition key 11. Further, the control unit 30 calculates a current command value of the steering assist motor 20 based on the steering torque T detected by the torque sensor 10 and the vehicle speed (vehicle speed) V detected by the vehicle speed sensor 12, and steering assist is performed. Based on the detected current value of the motor 20 and the current command value, the steering assist motor 20 is driven and controlled so that the detected current value of the steering assist motor 20 follows the current command value.

  FIG. 2 is a diagram showing a hardware configuration of the control unit 30 of FIG. As shown in FIG. 2, the control unit 30 includes a power supply relay 13, an MCU (micro control unit) 110, a motor drive circuit 108, a current detection circuit 120, a position detection circuit 130, and the like.

  The MCU 110 includes a CPU 101, a ROM 102, a RAM 103, an EEPROM 104, an interface (I / F) 105, an A / D converter 106, a PWM controller 107, and the like, and these are connected by a bus. The CPU 101 executes a control program stored in the ROM 102 and controls the electric power steering apparatus.

  The ROM 102 is used as a memory for storing a control program for the steering assist motor 20, and the RAM 103 is used as a work memory for operating the program. The EEPROM 104 is a memory that can retain the stored contents even after the power is shut off, and stores control data input and output by the control program.

  The A / D converter 106 inputs the steering torque T from the torque sensor 10, the current detection value Im of the steering assist motor 20 from the current detection circuit 120, the motor rotation angle signal θ from the position detection circuit 130, and the like. Convert to digital signal. The interface 105 is connected to an in-vehicle network such as CAN, and is for receiving a vehicle speed signal V (vehicle speed pulse) from the vehicle speed sensor 12.

  The PWM controller 107 outputs a PWM control signal for each phase of UVW based on the current command value of the steering assist motor 20. The motor drive circuit 108 is configured by an inverter circuit or the like, and drives the steering assist motor 20 based on a signal output from the PWM controller 107. The current detection circuit 120 detects the current value of the steering assist motor 20 and outputs the current detection value Im to the A / D converter 106. The position detection circuit 130 outputs an output signal from the position sensor 25 such as a resolver to the A / D converter 106 as a motor rotation angle signal θ.

  Next, the main part of the first embodiment will be described. FIG. 3 is a configuration diagram of functions realized by the cooperation of the respective units in the MCU 110 illustrated in FIG.

  In FIG. 3, a control function 1101, a timer function 1102, an abnormality detection function 1103, a storage function 1104, and an interface function 1105 are realized in the MCU 110. The control function 1101 is a function realized by a control program being read and executed by the CPU 101, and is a function for controlling the operation of the EPS apparatus 100 through the control unit 30. The control function 1101 actually operates by executing several processes.

  FIG. 4 is a diagram illustrating an example in which a process is executed in the control function 1101. The process includes a torque control process for calculating a steering assist torque command value based on the steering torque T and the vehicle speed V, a current control process for driving the steering assist motor 20 based on the steering assist torque command value, and communication of each part in the MCU 110. A communication process for controlling the device, a diagnostic process for performing a self-diagnosis of the EPS device 100, and the like. Each process periodically executes one or more tasks. For example, although the process 1 in FIG. 4 executes the tasks 11 to 16, these tasks are sequentially executed in series, and when the process is returned, the process 1 is executed again from the task 11. Here, the number of processes and the number of tasks included in one process may be different from the example shown in FIG.

  Each task updates a time stamp indicating the end time of the task when the task ends. The time stamp may be designed to be acquired by installing a free-run counter or the like, for example. FIG. 5 is a diagram for explaining the processing flow of the task 11 in FIG. As shown in FIG. 5, each task shown in FIG. 4 updates the time stamp just before the return, and records the time when the task ends. The storage function 1104 holds a task execution history including the completed task (task ID) and the updated time stamp of the task.

  In FIG. 3, the timer function 1102 interrupts each process execution at a predetermined execution cycle, and acquires a task execution history held in the storage function 1104. Then, the abnormality detection function 1103 determines whether the process is normally executed as designed from the acquired task execution history. If the process is not normally executed, the abnormality detection function 1103 performs abnormality processing such as resetting or stopping the program. . At this time, the timer function 1102 may be realized by a software timer. However, since the timer function 1102 is for detecting an abnormal execution state of the program, the timer function is preferably realized by hardware.

  A storage function 1104 is a function of the ROM 102, the RAM 103, and the EEPROM 104, and holds the above-described time stamp. In addition to this, a control program that is read by the CPU 101 to realize each function and data used by the control program are held.

  The interface function 1105 is realized by the interface 105 and the A / D converter 106. The interface function 1105 communicates with the torque sensor 10, the vehicle speed sensor 12, the motor drive circuit 108, the current detection circuit 120, and the position detection circuit 130. Do.

  An operation for detecting an abnormal execution state of each process according to the first embodiment of the present invention executed in the above configuration will be described. FIG. 6 is a flowchart for explaining the operation of detecting the abnormal execution state of each process.

  In FIG. 6, the interruption by the timer function 1102 is assumed to be executed at a cycle that is ½ of the execution cycle of the process to which the interruption is applied. First, when the process is executed, the timer function 1102 starts counting at the same time, and a first interrupt is generated when half of the process design execution time has elapsed (step S61). The anomaly detection function 1103 acquires the task execution history recorded in the storage function 1104 by each task, and whether or not the tasks are executed in a predetermined order when half the execution cycle of this process has elapsed. Judging. If not executed in the designed order (No in step S61), the abnormality monitoring function 1103 determines that this process is in an abnormal execution state, and proceeds to step S64 to perform abnormality processing such as program reset. Do. When the tasks are executed in the designed order (step S61, Yes), the abnormality monitoring function 1103 determines that this process is currently executed normally, and proceeds to step S62.

  Subsequently, since the interruption by the timer function 1102 is executed at a half cycle of this process, the second interruption is executed at the time of the process design return (step S62). In step S62, the abnormality detection function 1103 acquires the task execution history at this time again, and determines whether the tasks have been executed in the order as designed when the design process execution period has elapsed. When the abnormality detection function 1103 determines that the tasks have not been executed in the designed order (step S62, No), this process is determined to be in an abnormal execution state, and abnormality processing is performed (step S64). If it is determined that the process has been executed normally (step S62, Yes), the process proceeds to step S63. In step S63, the abnormality detection function 1103 resets the time stamp of each task held in the storage function 1104. Then, control goes to a step S61.

  As described above, the abnormal execution state of the program can be detected by detecting the abnormalities when the tasks are not executed in the designed order.

  Next, what case is detected as an abnormal execution state by the abnormal execution state detection operation of each process according to the first embodiment will be described using a more specific example. FIG. 8 is a diagram for explaining a specific example.

  In FIG. 8, the process 1 described as an example in FIG. 4 is taken up. In FIG. 8, as shown in the uppermost part of FIG. 8, process 1 operates at an execution cycle of 20 ms, task 11 is 1 ms, task 12 is 3 ms, task 13 is 4 ms, task 14 is 3 ms, and task 15 is Assume that the task 16 is executed in 4 ms, 4 ms, and is further designed to have a 1 ms blank. Next, as shown in the second row from the top in FIG. 8, an example is considered in which the task 12 and the task 13 are reversed and moved due to some trouble. At this time, at the time of the first interruption in step S61, the task execution history is in the order of task 11, task 13, and task 12. According to the check by the abnormality detection function 1103, it is determined that this is an abnormal execution state that is not executed in the designed order.

  Further, in the case of execution example 2 shown in the third row from the top in FIG. 8, the order of each task is normal, but task 11 and task 15 operate by exceeding the design execution time by 1 ms. ing. As a result, the process 1 exceeds the design execution period of 20 ms. At this time, since the task order is correct at the time of the first interrupt in step S61, it is not detected as abnormal. However, at the time of the check by the second interrupt in step S62, execution of the task determined from the time stamp of each task. The histories are task 11, task 12, task 13, task 14, and task 15, and the time stamp recording of task 16 has not been completed yet. Therefore, in this case, the abnormality detection function 1103 determines whether or not all the required tasks included in the process are executed during a predetermined process cycle, and all the required tasks are executed. If not, it is detected as an abnormal execution state of the program.

  Incidentally, the abnormality detection method using WDT (watchdog timer) is configured to cause any task to output a WDT reset signal, but in the cases of execution example 1 and execution example 2, all tasks Is executed, so it is not possible to detect anomalies.

  By the way, in the first embodiment, it has been described that the confirmation of the task order is executed only for the tasks belonging to one process. However, when the task execution order is fixed, the task order is changed over different processes. Confirmation may be made as abnormal when a task is not executed over different processes in the order as designed.

  In addition, it has been described that the interrupt and the abnormality detection operation by the timer function 1102 and the abnormality detection function 1103 are executed in an execution cycle that is ½ of the execution cycle of the process, but may be 1 / n (n is a positive integer). It does not matter how it is set. At this time, the speed of abnormality detection increases as the value of n increases. However, since the load on the CPU increases, it is necessary to set it appropriately at the design stage.

  Further, when an abnormality is detected, the abnormality process is simply performed in step S64. However, when the abnormality process is reset, the time stamp may be reset and the process may proceed to step S61 again.

  As described above, in the first embodiment, the task execution order is determined based on the measured end time of each task, and if the determined execution order is not executed in a predetermined order, a program error is detected. And detecting whether or not all the required tasks are being executed during the predetermined process cycle based on the measured end time of each task. When all of the tasks are not executed, it is detected as an abnormal execution state of the program, so the execution state of the control program can be strictly monitored, and cannot be detected by the watchdog timer. It is possible to accurately detect an abnormal execution state of a simple program.

  In the first embodiment, the abnormality detection function 1103 acquires a time stamp at the end of each task, checks whether each task is operating in the designed order, and whether a plurality of required tasks are included in a predetermined process cycle. By checking whether everything was running, it was determined whether the process was running normally. In the second embodiment, the abnormality detection function 1103 calculates the operation period of each task from the time stamp of each task, and also checks whether each task is executed according to the designed operation period. Determine whether is running normally. Since the configuration of the electric power steering apparatus according to the second embodiment is the same as that of the electric power steering apparatus according to the first embodiment, only the detection operation of the abnormal execution state of the process, which is a portion different from the first embodiment, will be described here.

  FIG. 7 is a flowchart for explaining the detection operation of the abnormal execution state of the process according to the second embodiment. Here, as in the description of the first embodiment, the interrupt by the timer function 1102 is assumed to be executed in a cycle that is ½ of the execution cycle of the process to which the interrupt is applied. First, when the process is executed, the timer function 1102 starts counting at the same time, and a first interrupt is generated when half of the execution time in the design of the process has elapsed (step S71). The anomaly detection function 1103 acquires the time stamp of each task recorded in the storage function 1104 by each task, and determines whether the tasks are executed in the designed order when half of the execution cycle of this process has elapsed. . When it is determined that the processes are not executed in the designed order (step S71, No), the abnormality detection function 1103 determines that there is an abnormality in the execution state of the process, proceeds to step S76, and resets the program. Abnormal processing is performed. When it is determined that the tasks are executed in the designed order (step S71, Yes), the process proceeds to step S72.

  In step S72, the abnormality detection function 1103 derives the execution time of each task from the fetched time stamp, and determines whether each task is executed with the execution time as designed. When half of the execution period of this process has elapsed, if the abnormality detection function 1103 finds a task that is not executed in the designed execution time (No in step S72), it is determined that there is an abnormality in the process execution state. Then, the abnormality process is executed (step S76). If all the execution times are as designed (Yes at Step S72), the process proceeds to Step S73.

  Subsequently, since the interrupt by the timer function 1102 is executed at a half cycle of this process, the second interrupt is executed at the timing when the design process execution cycle has elapsed in step S73. In step S73, the abnormality detection function 1103 acquires the time stamp of each task again, and determines whether each task has been executed in the designed order. When the abnormality detection function 1103 determines that each task has not been executed in the designed order (No at Step S73), it is determined that there is an abnormality in the process execution state, and abnormality processing is performed (Step S76). When it is determined that the process has been executed normally (step S73, Yes), the process proceeds to step S74.

  In step S74, the abnormality detection function 1103 derives the execution time of each task from the fetched time stamp of each task, and determines whether each task is executed with the designed execution time. If the abnormality detection function 1103 finds a task that has not been executed with the designed execution time when the design execution cycle of this process has elapsed (step S74, No), the abnormality detection function 1103 indicates the process execution state. Is determined to be abnormal, and abnormality processing is executed (step S76). If all are executed with the execution time as designed (step S74, Yes), the process proceeds to step S75.

  In step S75, since no abnormality has been seen so far, the time stamp is reset, and the process proceeds to step S71.

  As described above, in the second embodiment, in addition to the task execution order, the execution time of each task is also confirmed, and an abnormality in the execution state of the program is detected.

  Next, what case is detected as an abnormal execution state by the abnormal execution state detection operation of each process according to the second embodiment will be described more specifically with reference to FIG.

  Since the present embodiment also includes the operation of detecting the abnormal execution state of each process according to the first embodiment, the cases of execution example 1 and execution example 2 in FIG. 8 can be detected as abnormal. In the execution example 3, the operation of the task 11 takes 1 ms longer than the operation of the designed process 1. In this case, no abnormality is detected in the detection operation at the time of two interruptions in the first embodiment. However, in the detection operation in the present embodiment, the execution time of each task is also compared with the design execution time, so that an abnormality can be detected even in the case of execution example 3.

  By the way, also in the second embodiment, when the task execution order is fixed, the task order is confirmed over different processes, and the task is not executed over different processes in the designed order. You may make it judge that it is abnormal.

  In addition, the interrupt and the abnormality detection operation by the timer function 1102 and the abnormality detection function 1103 have been described as being executed in an execution cycle that is ½ of the process execution cycle, but may be 1 / n (n is a positive integer). It does not matter how it is set.

  Further, when abnormality is detected, the abnormality process is simply performed in step S76. However, when the abnormality process is reset, the time stamp may be reset and the process may proceed to step S71 again.

  As described above, in the second embodiment, the end time of each task is measured, the order in which the tasks are executed from the end time, and whether a plurality of required tasks are all executed during a predetermined process cycle, The execution time of each task is confirmed, and if it is not executed as designed, it is determined that there is an abnormality. Therefore, in addition to the effect of the electric power steering device according to the first embodiment, a minute delay in the task execution time is also abnormal. It is possible to detect as

  In the first embodiment, the first process for confirming the execution order of each task and the second process for confirming whether a plurality of required tasks are all executed during a predetermined process cycle are executed. In the second embodiment, in addition to the first and second processes, the third process for checking the execution time of each task is performed. However, the first to third processes are performed independently. May be. Furthermore, an abnormal task execution state other than the first to third processes may be detected based on the measured end time of each task. That is, in the present invention, the end time of each task is measured, and an abnormal execution state of the program is detected based on the measured end time of each task.

  As described above, the electric power steering apparatus according to the present invention is suitable for application to an electric power steering apparatus that detects an abnormality in the execution state of a program.

It is a figure which shows the general structure of an electric power steering apparatus. It is a figure which shows the hardware constitutions of a control unit. It is a figure which shows the function structure implement | achieved by MCU. It is a figure explaining the example of the process which a control function performs. It is a flowchart explaining the example of operation | movement of a task. 3 is a flowchart illustrating an operation for detecting an abnormal execution state of a process in the first embodiment. 12 is a flowchart illustrating an operation for detecting an abnormal execution state of a process in the second embodiment. It is a figure explaining the specific example of an abnormal execution state.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Steering handle 2 Column shaft 3 Reduction gear 4a, 4b Universal joint 5 Pinion rack mechanism 6 Tie rod 10 Torque sensor 11 Ignition key 12 Vehicle speed sensor 13 Power supply relay 14 Battery 20 Steering auxiliary motor 25 Position sensor 30 Control unit 100 EPS device 110 MCU
101 CPU
102 ROM
103 RAM
104 EEPROM
105 Interface 106 A / D Converter 107 PWM Controller 108 Motor Drive Circuit 120 Current Detection Circuit 130 Position Detection Circuit 1101 Control Function 1102 Timer Function 1103 Abnormality Detection Function 1104 Storage Function 1105 Interface Function

Claims (4)

In an electric power steering apparatus comprising: an electric motor for steering assistance; and a control unit that drives and controls the electric motor for steering assistance by executing a plurality of tasks in order,
The control unit measures an end time of each task, and detects an abnormal execution state of the program based on the measured end time of each task.   The control unit determines the execution order of the tasks based on the measured end time of each task, and if the determined execution order is not executed in a predetermined order set as an abnormal execution state of the program The electric power steering device according to claim 1, wherein the electric power steering device is detected.   The control unit determines whether or not all of a plurality of required tasks are executed during a predetermined process cycle based on the measured end time of each task, and the plurality of required tasks during the predetermined process cycle 2. The electric power steering apparatus according to claim 1, wherein when the program is not executed, an abnormal execution state of the program is detected.   The control unit determines whether each task is finished within a predetermined time based on the measured end time of each task, and if each task is not finished within the designed predetermined time, the program The electric power steering apparatus according to claim 1, wherein the electric power steering apparatus is detected as an abnormal execution state.

Images