WO2008062512A1

WO2008062512A1 - Multiprocessor system

Info

Publication number: WO2008062512A1
Application number: PCT/JP2006/323198
Authority: WO
Inventors: Hiromasa Takahashi; Takashi Chiba; Shunsuke Kamijo
Original assignee: Fujitsu Limited
Priority date: 2006-11-21
Filing date: 2006-11-21
Publication date: 2008-05-29
Also published as: JP5187194B2; JPWO2008062512A1

Abstract

When sensor information or mode information changes, a reception unit (33) generates an interrupt signal. A driving state judging table (12) defines the correspondence relation of a combination pattern of the sensor information and the mode information with a driving state. A priority processing information table (13) defines the correspondence relation of the driving state with an application which is to be executed. When the reception unit (33) generates the interrupt signal, an operating state management module (36) determines the application which is to be executed with reference to the driving state judging table (12) and the priority processing information table (13) to allocate the application to the corresponding processor elements (PE0 to PE3).

Description

Specification

Manolet processor system

Technical field

The present invention relates to a multiprocessor system including a plurality of processor elements, and more particularly to a multiprocessor system incorporated in a control target device whose operation mode changes depending on an external environment.

Background art

[0002] In recent years, the IT industry of automobiles has advanced, and systems for realizing various controls have been developed. In this case, in the conventional system, as shown in FIG. 1A, there are many configurations in which one ECU (Electronic Control Unit) is assigned to one application (hereinafter referred to as a distributed processing system). In the example shown in Fig. 1A, the corresponding ECUs (la to If) are assigned to six applications (front monitoring, night vision, right side monitoring, left side monitoring, driver monitoring, and rear monitoring). Each ECU executes processing independently. Therefore, the following three benefits are obtained in a distributed processing system.

(1) Programs can be developed independently for each ECU.

(2) Easy recovery when a failure occurs.

(3) To add new functions, simply add a new ECU.

[0003] However, in distributed processing systems, the number of ECUs increases as the number of applications to be processed increases. ], Problems such as increased mounting volume and increased power consumption occur. Therefore, in order to solve this problem, as shown in Fig. 1B, multiple sensors for detecting the external environment and one central processing unit 2 are connected via a LAN. A configuration for executing applications in parallel (hereinafter referred to as a centralized processing system) has been proposed.

[0004] As a conventional configuration in a centralized processing system, three methods shown in FIGS. 2A to 2C are known. The following is for processing six applications A to F in a centralized processing system! I will explain in a moment. In the first method shown in FIG. 2A, the central processing unit 2 has only one processor. Applications A to F are processed in a time-sharing manner using the task switching function of the real-time OS. In this case, in order to satisfy the required response time, it is necessary to have sufficient processing capacity of the processor even if all applications are processed. Here, the required processing capacity is approximately obtained by adding the overhead of time-sharing processing to the total processing amount of the processing units la˜: Lf shown in FIG. 1A. Therefore, as the number of applications to be processed increases, high performance, high price, high power consumption processors are required.

In the second method shown in FIG. 2B, the central processing unit 2 is a multiprocessor system. Here, the central processing unit 2 contains four processor elements (PE0 to PE3). Applications A to F are dynamically allocated to processor elements PE0 to PE3 using a real-time OS. Therefore, even in this method, in order to ensure the required responsiveness, even if six applications are executed using four processor elements, there is still a need for sufficient processing capacity. In this case, the processing capacity required is almost the same as in the first method. For this reason, the second method also requires a high-performance multiprocessor, and it is difficult to realize low cost and low power consumption.

In the third method shown in FIG. 2C, the central processing unit 2 includes the same number of processor elements as the number of applications to be executed. That is, applications A to F are fixedly assigned to processor elements PE0 to PE5, respectively! /. In this configuration, since each processor element only executes a predetermined application that does not need to perform time-sharing processing, the system configuration is relatively simple.

As described above, the centralized processing system has been proposed to solve the problems of the distributed processing system. However, the first and second methods cannot satisfy the features (1) to (3) of the distributed processing system. The third method satisfies the above (1) and (2), but cannot satisfy the above (3).

As related techniques, Patent Documents 1 and 2 disclose a technique for improving the reliability of a computer system including a plurality of processors. Patent Document 1: Japanese Unexamined Patent Publication No. 2006-11992

Patent Document 2: JP 2002-259155 A

Disclosure of the invention

[0010] An object of the present invention is to provide a multi-processor system that facilitates program development, failure recovery, and function addition.

The multiprocessor system of the present invention includes a plurality of processor elements and a control target device. The multiprocessor system of the present invention is selected by a selection unit that dynamically selects an application program to be executed from among a plurality of application programs based on an operation status of the control target device, and the selection unit. And assigning means for assigning the abrasion program to the plurality of processor elements.

[0011] In the multiprocessor system, when the operation status of the control target apparatus changes, the selection unit selects an application program corresponding to the new operation status. The assigning means assigns the selected application program to a plurality of processor elements. Therefore, a necessary application program can be selectively executed in accordance with the operation status of the device to be controlled. At this time, each application program can be assigned to any processor element rather than fixedly assigned to a specific processor element.

Thus, each application program can be developed independently of other application programs, and correction work is easy. Even if some of the processor elements fail, the application program to be executed can be assigned to the remaining normal processor elements. Furthermore, application tracking is realized by changing the allocation sequence without changing the hardware configuration. Brief Description of Drawings

FIG. 1A is a diagram schematically showing a distributed processing system.

FIG. 1B is a diagram schematically showing a centralized processing type system.

FIG. 2A is a diagram (part 1) showing a configuration of a conventional centralized processing system.

FIG. 2B is a diagram (part 2) showing a configuration of a conventional centralized processing system. FIG. 2C is a diagram (part 3) showing the configuration of a conventional centralized processing system.

FIG. 3 is a diagram for explaining the outline of the present invention.

FIG. 4 is a diagram illustrating a hardware configuration of the multiprocessor system according to the embodiment.

FIG. 5 is an example of a traveling state determination table and a priority processing information table.

FIG. 6 is a diagram showing a software configuration of the multiprocessor system of the embodiment.

FIG. 7 is a flowchart showing a procedure for determining a running situation.

FIG. 8 is a flowchart of processing for assigning an application to a processor element.

FIG. 9 is a diagram schematically showing how an application is changed in accordance with changes in driving conditions.

FIG. 10 is a flowchart showing an allocation method when the number of applications is larger than the number of processor elements.

FIG. 11 shows an example of a priority processing information table.

FIG. 12 shows an example of a priority processing information table when processor elements having different abilities are included.

FIG. 13 is a flowchart showing an assignment procedure when processor elements having different abilities are included. BEST MODE FOR CARRYING OUT THE INVENTION

[0014] A multiprocessor system according to an embodiment of the present invention is used by being incorporated in a device to be controlled. Here, the device to be controlled is not particularly limited, and for example, an automobile, an aircraft, a railway vehicle, a ship, and the like are assumed. In the following description, it is assumed that the device to be controlled is an automobile.

[0015] In recent years, automobiles are generally equipped with various applications in order to control their running. For example, a function that measures the distance between vehicles and automatically activates the brake, and a function that monitors the operation of the driver and generates an alarm when necessary are widely known. However, not all of these applications are necessarily required at the same time. That is, for example, the anti-collision function is important when traveling at high speeds. The night vision function is important at night. No need for daytime. Thus, paying attention to the fact that the functions of all installed applications are not necessary in the individual driving situations of the automobile, the control system according to the embodiment of the present invention executes in accordance with the running situation of the automobile. The application to be selected is dynamically selected. The outline of the embodiment of the present invention will be described below with reference to FIG.

In FIG. 3, the multi-core processor unit 11 includes a plurality of processor elements (PE 0 to PE 3). Each processor element PEO to PE3 provides a corresponding function by executing an assigned application program. In this example, the applications include forward monitoring (collision detection), forward monitoring (white line detection), night vision, side monitoring, rear monitoring, driver monitoring, passenger monitoring for airbags, indoor comfort control, security It is assumed that monitoring and the like are prepared in advance.

[0017] The automobile of the embodiment is equipped with various sensors 101 for detecting an external environment and an operating state. The mounted sensors are, for example, a vehicle speed sensor that detects the traveling speed, a steering angle sensor that detects the steering angle of the steering, a G sensor that detects acceleration, a short sensor, a solar radiation sensor, and the like. The automobile also includes a mode switch 102 for designating an operation mode. Mode switches include, for example, a switch that indicates whether or not to operate ACC (Adaptive Cruis e Control), PCS (Pre-Crash Safety), VDC (Vehicle Dynamic Control), or a power or force that operates a blinker. , A switch that instructs the power to turn on the light, and an input unit that gives instructions to the automatic transmission.

[0018] The traveling state determination table 12 stores information for determining the traveling state of the vehicle based on the sensor information from which various sensor forces are obtained and the mode information from the various mode switch forces. In other words, the traveling situation can be detected by referring to the traveling situation determination table 12 using the sensor information and the mode information. In this embodiment, since it is assumed that the multiprocessor system according to the present invention is incorporated in an automobile, the force for detecting the “running situation” generally means that the operating situation is detected.

[0019] The priority processing information table 13 includes an application to be executed for each traveling state pattern. Information specifying Yong is registered. Here, the application to be executed is given a priority for each driving situation pattern.

[0020] The selection unit 14 refers to the priority processing information table 13 and selects an application to be executed in accordance with the driving situation. Since the driving situation changes every moment, the selection unit 14 dynamically selects an application to be executed according to the changing driving situation.

[0021] Allocation unit 15 allocates the application selected by selection unit 14 to processor elements PEO to PE3. Then, each of the processor elements PEO to PE3 executes the assigned application. This provides multiple functions for controlling the car in parallel.

As described above, in the system according to the embodiment, a corresponding application is dynamically selected and executed in accordance with the traveling state of the automobile. Therefore, in the system of the embodiment, even if the “number of processor elements incorporated in the multi-core processor unit 11” is smaller than the “total number of prepared applications”, the V Since the number of required applications is generally smaller than the “total number of prepared applications”, the necessary applications can be executed. According to the system of the embodiment, the following effects can be obtained.

(1) Similar to the distributed processing system shown in Fig. 1A, three advantages (program development, easy failure recovery, and easy addition of functions) can be obtained. In this respect, it is superior to the conventional centralized processing system shown in FIGS. 2A to 2C.

(2) Since it is not necessary to provide an ECU for each application, it is superior in terms of cost reduction, size reduction, weight reduction, and power consumption reduction compared to a distributed processing system.

(3) In the distributed processing system or the configuration shown in FIG. 2C, there are ECUs that may operate substantially depending on the driving situation. However, in the system of the embodiment, processor resources are used efficiently. The

(4) Each sensor force Sensor fusion that integrates and processes the obtained sensor information becomes possible, and advanced processing can be provided.

FIG. 4 is a diagram of a hard processor of a multiprocessor system that implements the control system of the embodiment. FIG. Here, it is assumed that this multiprocessor system is used by being incorporated in the automobile in order to control the running of the automobile.

[0024] The multi-core processor unit 11 includes four processor elements PEO to PE3. Note that the number of processor elements included in the multiprocessor system is not particularly limited. The processor elements PEO to PE3 are connected to the memory bus 21 and also to the IZO bus 22, respectively. Each processor element ΡΕ0 to ΡΕ3 is connected to a local area network (LAN1, LAN2) via an IZO bus 22. As shown in FIG. 1B, various sensors (including cameras) are connected on the local area network, and each processor element PE0 to PE3 is a sensor collected by each sensor via this local area network. Receive information.

The memory 31 is SDRAM, for example, and is connected to the memory bus 21. The memory 31 is provided with the traveling state determination table 12 and the priority processing information table 13 described with reference to FIG. The application program is loaded from the nonvolatile memory 32 to the memory 31. Each of the processor elements PE0 to PE3 can access the memory 31 via the memory bus 21.

The nonvolatile memory 32 is a flash ROM, for example, and is connected to the I / O bus 22. The non-volatile memory 32 holds information to be loaded into the memory 31 (including the application program, the traveling state determination table 12, and the priority processing information table 13).

[0027] The receiving unit 33 periodically collects sensor information obtained by each sensor and mode information obtained by each mode switch. The collected sensor information and mode information are written in the driving status register 34. When detecting that at least one of the sensor information or the mode information has changed, the reception unit 33 outputs a status change signal indicating that the driving status of the vehicle has changed. Then, when receiving the situation change signal, the interrupt controller 35 transmits an interrupt signal caused by the change in the running situation to the processor element including the operation state management module 36. Sensor information and Z or mode information may be given to the receiving unit 33 via a network of control systems installed in the automobile. [0028] The processor element activates the operation state management module 36 when receiving an interrupt signal resulting from a change in the running state. The operation state management module 36 refers to the traveling state determination table 12 and the priority processing information table 13 and selects an application to be executed. The selected application is assigned to the corresponding processor element. As a result, an application corresponding to the driving situation of the car is executed.

[0029] It should be noted that the operation state management module 36 may be provided in another processor element that is provided in the processor element PE3 in the example shown in FIG. However, the operation state management module 36 is preferably provided in a processor element to which a low priority application is assigned. For example, in an application allocation rule, it is determined that the highest priority application is assigned to the processor element PEO, and the second and third highest priority applications are assigned to the processor elements PE1 and PE2, respectively. The operating state management module 36 preferably operates on the processor element PE3! /! /.

FIG. 5 is an example of the traveling state determination table 12 and the priority processing information table 13. In the driving situation determination table 12, corresponding driving situations are defined for each combination of sensor information and mode information. Here, in this embodiment, each sensor information is represented by a comparison result with a threshold value. For example, sensor information related to the vehicle speed sensor is represented by “high speed Z low speed”, and information related to the solar radiation sensor is represented by “strong Z weak”. The mode information is basically represented by “ON / OFF”. When a set of sensor information and mode information is acquired, a driving situation uniquely corresponding to the sensor information and mode information is obtained. In the example shown in Fig. 5, "Vehicle speed: High speed", "Steering angle: Small", "Solar radiation: Strong", "Acceleration: Small", "Jorate: Small", "Throttle: Small", "HL: Off", "ACC: On" When “PCS: ON” and “WINKER: OFF” are detected, “Running condition 1 (Highway mode 1)” is obtained.

[0031] In the priority processing information table 13, the application to be executed and the priority order of the application to be executed are defined for each traveling situation. In the example shown in FIG. 5, for example, in the driving situation 1, the applications to be executed are “forward monitoring 3” “side monitoring L” and “driver monitoring”. And the highest for “front monitoring H” Priority is given, “side monitoring L” is given the second highest priority, and “driver monitoring” is given the lowest priority. “H” means high-precision control, and “L” means low-precision control. In addition, the load factor of the processor element when each application is executed is also described.

In this way, in the traveling state determination table 12, “traveling state” is defined for the combination of sensor information and mode information. In the priority processing information table 13, applications to be executed and their priorities are defined for each “running situation”. Therefore, by referring to these tables using sensor information and mode information, the application to be executed (and its priority) can be deduced.

In the embodiment shown in FIG. 5, the traveling state determination table 12 and the priority processing information table 13 are separated. Therefore, according to this configuration, it is possible to determine the driving situation and select an application to be executed independently of each other. However, the present invention is not limited to this configuration, but it is also possible to provide one table that integrates the contents of these tables.

FIG. 6 is a diagram illustrating a software configuration of the multiprocessor system according to the embodiment.

As shown in Figure 6, a real-time OS runs on each processor element. This real time OS has a communication library between processor elements. Each application runs on this real-time OS. The multiprocessor system of this embodiment includes an operation state management module 36 that operates in a layer between the OS layer and the application layer. The operation management module 36 may be incorporated as a function in the OS. As will be described in detail later, the operation state management module 36 refers to the traveling state determination table 12 and the priority processing information table 13, selects an application to be executed, and assigns the selected application to a corresponding processor element. In the example shown in FIG. 6, applications A, B, E, and G are selected and assigned to processor elements PE0, PE1, PE2, and PE3, respectively.

[0035] The operation state management module 36 may be mounted on one processor element, or may be mounted on a plurality of processor elements. Next, the operation of the multiprocessor system configured as described above will be described with reference to FIG. Here, it is assumed that the operation state management module 36 operates in the processor element PE3. Further, it is assumed that the running condition determination table 12 and the priority processing information table 13 shown in FIG.

(1) The reception unit 33 periodically collects sensor information and mode information. Newly collected information is compared with previously collected information. Here, the previously collected information is held in the travel status register 34. When the previously collected information and the newly collected information are inconsistent with each other, it is determined that the driving state of the vehicle has changed, and a state change signal is output. Then, when receiving the situation change signal, the interrupt controller 35 transmits an interrupt signal resulting from the change in the running situation to the processor element PE3 including the operation state management module 36. At this time, the newly collected sensor information and mode information are written in the travel status register 34.

[0037] (2) When this interrupt signal is given, the processor element PE3 calls the operation state management module 36. The operation state management module 36 accesses the reception unit 33 via the IZO bus 22 and reads the latest sensor information and mode information from the traveling state register 34. Then, the operation state management module 36 extracts a traveling state number for identifying the traveling state corresponding to the sensor information and the mode information from the traveling state determination table 12.

(3) The operation state management module 36 accesses the priority processing information table 13 using the traveling state number extracted from the traveling state determination table 12. As a result, the application to be executed in the current driving situation and its priority are obtained.

(4) The operation state management module 36 assigns the application to be executed to the processor elements ΡΕ0 to ΡΕ3 according to the priority order. At this time, these applications may be assigned to the processor elements ΡΕ0, ΡΕ1, ΡΕ2, and ΡΕ3 in order of priority. In this case, for example, in the traveling situation 1 shown in FIG. 5, “front monitoring Η”, “side monitoring L”, and “driver monitoring” are assigned to the processor elements PE0, PE1, and PE2, respectively. Also, in the driving situation 2, “front monitoring H”, “night vision”, “side monitoring L”, and “driver monitoring” are the processor elements PE0, PE1, PE2, P, respectively. Assigned to E3. As a result, an application corresponding to the driving situation is executed.

[0040] (5) When the traveling state of the automobile changes, at least a part of the sensor information or the mode information should be changed. Therefore, by executing the above (1) to (4) using the latest sensor information or mode information, a new application is allocated to the processor elements PEO to PE3 in accordance with changes in the driving situation. That is, the application to be executed is dynamically switched according to the traveling state of the vehicle.

FIG. 7 is a flowchart showing a procedure for determining the traveling situation. Steps Sl to S4 are executed by the reception unit 33 (and the interrupt controller 35), and steps S5 to S7 are executed by the operation state management module 36.

[0042] In steps S1 to S2, sensor information and mode information are periodically collected to check whether the driving situation has changed. If the running situation has changed, an interrupt signal is output to the multi-core processor unit 11 in step S3. Then, in step S4, the sensor information and monitor information collected in step S1 are written into the travel status register 34.

[0043] Step S5 is a step of waiting for an interrupt signal resulting from a change in the driving situation. When the interrupt signal is received, the operation state management module 36 is activated. In step S6, the sensor information and mode information held in the travel status register 34 are read. Then, in step S7, by referring to the driving condition determination table 12, a driving condition number for identifying the current driving condition is acquired. In this way, the current driving situation is detected by executing the processing of the flowchart shown in FIG.

FIG. 8 is a flowchart of processing for assigning an application to a processor element. This process is executed by the operation state management module 36 subsequent to step S7 in the flowchart shown in FIG. Here, it is assumed that the operation state management module 36 operates in the processor element PE3. Furthermore, the number of applications to be executed in each driving situation shall be the same as or less than the number of processor elements.

[0045] In steps S11 to S12, the priority processing information tape is obtained using the obtained traveling situation number. See Figure 13. This detects applications to run and their priorities. In steps S13 to S14, the application to be executed is assigned to the processor element according to the priority order. In this case, the highest priority application is first assigned to the processor element PEO. Subsequently, the second highest priority application is assigned to processor element PE1. Similarly, the third and fourth priority applications are assigned to the processor elements PE2 and PE3. However, if there are 3 or less applications to be executed, there are processor elements to which no application is assigned. Also, if there are more than 5 applications to be executed, it will not be executed !, and there will be applications.

[0046] In step S15, a PEZ application allocation table is created. Here, it is assumed that applications A, B, C, and D are assigned to the profiler elements PEO, PE1, PE2, and PE3, respectively.

[0047] In step S16, the OS is called in the processor element PE3, and the tasks of the processor elements PE0 to PE2 are terminated using the remote task termination function using the inter-PE communication library. In step S17, the OS is called in the processor element PE3, and the task of the processor element PE3 is terminated.

[0048] In step S18, the processor element PE3 calls the OS, and uses the remote task activation function using the inter-PE communication library to activate the processor element PE0 for A. Similarly, in steps S19 and S20, the processor elements P El and PE2 start the applications B and C, respectively. In step S21, the processor element PE3 calls the OS and starts the application D. By the above procedure, applications A to D are executed by processor elements PE0 to PE3, respectively.

FIG. 9 is a diagram schematically showing how an application is changed in accordance with a change in the driving situation. Here, the case where the driving state of the car has shifted from state 1 to state 2 is depicted. In the driving situation 1, in the processor elements PE0 to PE3, “forward monitoring (collision detection) J” forward monitoring (white line detection) J “side monitoring” and “driver monitoring” are executed. When the driving state 2 is entered, the processor elements PEO to PE3 perform “forward monitoring (collision detection)”, “night vision”, “forward monitoring (white line detection)”, and “side monitoring”.

FIG. 10 is a flowchart showing another allocation method. Note that the processing in this flowchart is executed when the number of applications to be executed is larger than the number of processor elements. Here, it is assumed that the detected driving situation is the state k shown in FIG.

[0051] First, in steps S11 to S12, the priority processing information table 13 is referenced to detect an application to be executed. At this time, the priority and load factor of each application are also detected. In steps S13 and S31, the first to fourth priority applications (ie, applications A to D) are assigned to the processor elements PEO to PE3, respectively.

[0052] In step S32, it is checked whether there are any applications that have not been assigned to the processor elements. If such an application remains, in step S33, the processing capability of each processor element is checked. Here, applications A to D have already been assigned to processor elements PEO to PE3, respectively. Therefore, the remaining processing power of processor elements PEO, PE1, PE2, and PE3 is 20%, 50%, 50%, and 60%, respectively!

[0053] In step S34, an application that has not been assigned to a processor element.

(Here, applications E and F) are assigned to the remaining processor elements. Here, the processor element PEO has only 20% processing capacity. That is, the processor element PEO cannot execute the applications E and F. Thus, application E is assigned to processor element PE1. In addition, by re-executing steps S32 to S34, application F is assigned to processor element PE2.

The subsequent processing is basically the same as the procedure described with reference to FIG. That is, when the creation of the PEZ application allocation table is completed in step S35, the task of each processor element is ended in steps S36 to S37. Subsequently, each of the applications A to F is started in steps S38 to S41! Thus, according to the allocation method shown in FIG. 10, more applications than the number of processor elements can be executed in parallel. Therefore, the low cost of the multiprocessor system can be achieved. In addition, since the total processing amount of applications allocated to each processor element does not exceed the processing capacity of that processor element, real-time performance of each application is ensured.

Meanwhile, the processing capabilities of the processor elements included in the multiprocessor system are not necessarily the same. In other words, a part of the prepared application may be executed only by a specific processor element (or preferably executed by a specific processor element). Yes.

[0057] For example, it is assumed that the processor element PEO has a mechanism for efficiently processing “forward monitoring H”, and the processor element PE3 has a mechanism for efficiently processing “night vision”. In other words, “Night Vision”, which is preferably assigned to processor element PEO, is preferably assigned to processor element PE3. In this case, for these specific applications, the processor elements to be allocated are fixedly defined in the priority processing information table 13 as shown in FIG.

FIG. 13 is a flowchart showing the allocation procedure when processor elements having different capabilities are included. Here, in the priority processing information table 13, it is assumed that the application B force processor element PEO included in the application to be executed is fixedly associated.

[0059] The assignment procedure in the case where processor elements having different capacities are included is basically the same as the procedure shown in FIG. 8 (or FIG. 10). However, in step S51, it is checked whether or not the application includes a specific processor element specified in the application to be executed. If such an application exists, in step S52, the application is assigned to the designated processor element. In this example, application B is assigned to processor element PEO prior to other applications. [0060] In steps S53 to S54, the remaining applications are processed in the same manner as steps S13 to S14. This assigns applications B, A, C, and D to processor elements PEO, PE1, PE2, and PE3, respectively. The subsequent processing is basically the same as the procedure described with reference to FIG. 8 (or FIG. 10).

[0061] <Other Embodiments>

(1) The contents of processing executed in the multiprocessor system may be displayed on the display device. For example, an application being executed is displayed on the display device. In this case, if the application to be executed changes as the driving situation changes, the display contents will also change.

[0062] (2) You may get information on the navigation system as sensor information!

(3) The operation state management module may operate on the processor element that executes the application with the lowest priority. In this case, it becomes easier to guarantee the responsiveness of the application with the higher priority.

[0063] (4) When changing the assignment of applications in accordance with changes in driving conditions, each processor element may be reset and then rebooted.

(5) When the processing amount of a certain application is large, or when the number of applications to be executed is smaller than the number of processor elements, one application may be assigned to multiple processor elements. .

[0064] (6) In the above-described embodiment, the driving situation is determined based on both the sensor information and the mode information, but the driving situation is determined based on either the sensor information or the mode information. Try to judge.

Claims

The scope of the claims

[1] A multiprocessor system having a plurality of processor elements and controlled devices,

Selection means for dynamically selecting an application program to be executed among a plurality of application programs based on the operation status of the control target device;

Allocating means for allocating the application program selected by the selecting means to the plurality of processor elements;

A multiprocessor system.

[2] A multiprocessor system according to claim 1,

Sensor power for detecting the external environment or operating state of the device to be controlled Corresponding relationship information indicating a corresponding relationship between a plurality of operating situations defined by the obtained sensor information and an application program to be executed in each operating status And further storing means for storing

The selection unit selects a corresponding application program by referring to the storage unit based on the sensor information.

A multiprocessor system characterized by that.

[3] The multiprocessor system according to claim 1,

Receiving means for receiving sensor information for detecting an external environment or an operating state of the control target device and mode information for specifying an operation mode of the control target device;

Storage means for storing correspondence information representing a correspondence relationship between a plurality of operation situations defined by a combination of the sensor information and the mode information and an application program to be executed in each operation situation;

The selection means selects a corresponding application program by referring to the storage means using the sensor information and mode information.

A multiprocessor system characterized by that.

[4] A multiprocessor system according to claim 3,

The correspondence information is defined by a combination of the sensor information and mode information. Information specifying the application program to be executed and the processor element to execute each application program.

A multiprocessor system characterized by that.

[5] The multiprocessor system according to claim 3,

The correspondence information includes information that specifies an application program to be executed and a priority order of each application program for each operation status defined by a combination of the sensor information and mode information.

A multiprocessor system characterized by that.

[6] A multiprocessor system according to claim 3,

The correspondence information specifies the number of application programs equal to or less than the number of processor elements as application programs to be executed for each operation state defined by the combination of the sensor information and mode information.

A multiprocessor system characterized by that.

[7] A multiprocessor system according to claim 3,

The correspondence information includes information indicating the application program to be executed, the priority of each application program, and the load factor of each application program for each operation status defined by the combination of the sensor information and the mode information. ,

The selection means assigns a high priority application program to each corresponding processor element, and further corresponds a low priority application program in a range where the sum of load factors does not exceed the processing capability of the processor element. Assign to processor element

A multiprocessor system characterized by that.

[8] A multiprocessor system according to claim 1,

The selecting means and the assigning means are provided in a predetermined processor element among the plurality of processor elements.

A multiprocessor system characterized by that.

[9] The multiprocessor system according to claim 1,

The selection means and the assignment means are provided in the processor element to which the application program having the lowest priority among the plurality of processor elements is assigned! /

A multiprocessor system characterized by that.

[10] A multiprocessor system according to claim 1,

The assigning means uses an OS task switching function to change an application program executed by each processor element.

A multiprocessor system characterized by that.

[11] A multiprocessor system according to claim 3,

If the first application program is included in the application program to be executed, the correspondence information is fixedly assigned to the first processor element. Contains information

A multiprocessor system characterized by that.

[12] A control system for controlling the operation of a control target device,

Multiple processor elements;

A sensor for detecting an external environment or an operating state of the device to be controlled;

Prepared in advance based on an operation state defined by a combination of input means for inputting information specifying the operation mode of the device to be controlled, sensor information obtained from the sensor, and mode information obtained from the input means. A selection means for dynamically selecting an application program to be executed from among a plurality of application programs,

Having a control system.