Method for monitoring electronic system
By using orthogonal functions to generate independent internal problems and answers in the controller, the problem of high overhead of problem-response communication in the prior art is solved, faster communication and shorter error processing time intervals are achieved, and maintenance costs are reduced .
Patent Information
- Application Number
- CN202280093580.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-01-13
- Filing Date
- 2022-12-23
- Publication Date
- 2024-11-01
AI Technical Summary
The prior art has high overhead in maintaining or handling problem-answer (Q/R) communications, limiting the shortest error processing time interval, especially in modern systems such as electric vehicles, requiring faster detection and response times.
By introducing an external monitoring unit into the controller, using orthogonal functions to generate completely independent internal problems and responses, break away from the problems of the external monitoring unit, reduce the number of interrupts and service calls, achieve faster communication, and on the same communication bus Conduct communication.
Significant savings of time and storage resources, shortening error processing time intervals, reducing maintenance costs, and enabling faster system response and independent internal problem handling.
Smart Images

Figure CN118891591A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for monitoring an electronic system and to a device for carrying out the method. The invention further relates to a computer program and a machine-readable storage medium. Background Art
[0002] In the field of functional safety, the term safety integrity level is used, which is also called safety level or safety requirement level. In this context, the safety integrity level is used to assess the reliability of electrical, electronic or programmable electronic units with respect to safety functions. The level sought results in safety-oriented design principles that must be adhered to in order to minimize the risk of failure.
[0003] All modern systems supporting such safety integrity levels must implement various safety or communication mechanisms. For example, all engine control units use question response (Q / R: Query Response) communication, which is implemented as a safety mechanism between the internal main function controller and an external monitoring unit or external watchdog. The external monitoring unit ensures that a transfer to a safe state of the system occurs in the event of a failure of the internal main function controller.
[0004] The known solution provides for the following steps, see Figure 1 :
[0005] 1. The external monitoring unit evaluates the previous cycle and outputs the question Q.
[0006] 2. The controller must read out a question Q (reference numeral 8 ) from an external monitoring unit at a point in time 16 via a communication channel or a peripheral unit, which requires an interruption or interruption or other planned activities or other planned tasks and processes.
[0007] 3. The controller must keep the question Q (reference numeral 8) available to all associated internal subsystems (reference numeral 10) and inform the subsystems of the new question availability (reference numerals 20, 22, 24, 26, ...).
[0008] 4. After a given period of time and within the time window of the external monitoring unit, for a correct behavior of the controller, the controller must collect all partial replies (reference numeral 28 ) and calculate the final or final reply R (reference numeral 30 ).
[0009] 5. The controller must send a final response R (reference numeral 30) to the external monitoring unit, for which a second interrupt or other planned tasks and processes are required. This occurs at time 18, wherein this time 18 must be within a predefined time window (reference numeral 14), otherwise the response of the controller is evaluated by the external monitoring unit as invalid.
[0010] This process is repeated during the entire critical operating cycle of the controller, i.e. a new question (reference numeral 54) from the external monitoring unit at time 52 must be processed by all monitored subsystems 50 and the final answer (reference numeral 56) must be set in the external monitoring unit at time 58 within the time window (reference numeral 60). Reference is made to publication DE 102 05 809 A1.
[0011] Question 8 and new question 54 differ from each other in the case of normal correct operation of the system. If the previous answer (reference numeral 30) is wrong or the time at which the answer was placed is outside time window 14, new question 54 can be the same, possibly repeated question 8.
[0012] Since problems with external monitoring units are known to all internal monitored units, and the type of external monitoring unit or the way in which new problems are generated is also known, the monitored subsystems can also simply acquire unnecessary information about the operating or safety status of the entire system.
[0013] In any case there are no internal questions or responses. The external monitoring unit's questions are known to all internal monitored units and the way to collect the internal partial responses or their function is known and the same effort is sent to the external monitoring unit.
[0014] References to the known solutions mentioned above Figure 1 .
[0015] The main problem with this solution is the high overhead in maintaining or handling the Q / R communication and the external monitoring unit (also called external watchdog). This also limits the shortest fault handling time interval (FHTI) in a way that has an impact on the fault tolerant time interval (FTTI). Since modern systems require much shorter detection and response times for errors and faults in the system, even in the case of increasingly popular electric vehicles, new concepts for faster or even different types of communication and maintenance of external monitoring units are needed.
[0016] The publication DE 10205809 A1 describes a method and a device for monitoring the control of an operating process in a vehicle. In the case of this method, control functions and monitoring functions are performed in a control unit. Provision is made for a monitoring module to transmit at least one question to the control unit, wherein a first monitoring function is provided in the control unit. A partial answer is calculated from the question in a second time slot. An answer is formed from the partial answers. In a first time slot, the formation of an answer is activated, wherein the control unit transmits the answer to the monitoring module. Based on the answer, the monitoring module detects an error in the execution of the control function. In this case, the first time slot and the second time slot are asynchronous with respect to each other. Summary of the invention
[0017] Against this background, a method according to claim 1 and a device having the features of claim 13 are proposed. Furthermore, a computer program according to claim 14 and a machine-readable storage medium according to claim 15 are proposed. Embodiments emerge from the dependent claims and the description.
[0018] The proposed method is used for monitoring an electronic system which has a plurality of subsystems in a control unit, wherein the method is implemented using an external monitoring unit and comprises the following steps:
[0019] - an external monitoring unit evaluates previous operating cycles of the system and, based on this evaluation, provides external questions to the controller,
[0020] the internal driver of the external monitoring unit in the control unit, ie the driver of the external monitoring unit in the control unit, asks the internal monitoring unit in the control unit at least one new question,
[0021] - after a certain period of time, the controller collects the internal partial replies of all internal monitored units in the controller,
[0022] A driver of the external monitoring unit in the controller outputs an external response to the external monitoring unit.
[0023] The method sets the following steps in the design scheme:
[0024] 1. The external monitoring unit evaluates the previous cycle and outputs the question Q.
[0025] 2. After a given time period and within the time window of the external monitoring unit, regarding the correct behavior of the controller, the controller reads the question Q of the external monitoring unit via the communication channel or the peripheral unit and immediately responds with a deliberately correct or deliberately incorrect response R. Based on the subsystem's response F, the controller also outputs a new question F to the subsystem or a new, different question to the subsystem, where only interruptions or other planned tasks and processes are required.
[0026] In this regard, refer to Figure 2.
[0027] The proposed method has a number of advantages, at least in some embodiments:
[0028] 1. Communications with external computing units can be encapsulated into larger communication subgroups and thus enable faster communications through the same communication bus or the same communication interface.
[0029] 2. The number of required interrupts or service calls is halved. This means that time and storage resources are significantly saved. This also enables the processing or even implementation of shorter FHTIs.
[0030] 3. The processing of the internal system and the subsystems SUBSYS_1 to SUBSYS_m with the final connection function F provides a much larger time interval compared to the previous system with two calls.
[0031] 4. Question Qn and question Fn can be completely decoupled. This means that the external and internal questions are completely separated and none of the internal monitored subsystems of the controller knows the exact question of the external monitoring unit, except for the processing unit or driver that generates the internal question by an orthogonal function to the question that generates the external question. This also means that even if the question of the next monitoring unit is repeated in the next cycle, the internal question is generated completely anew, independently of the external question. Therefore, no monitored unit has information about the overall functionality.
[0032] This also enables completely new sets of results, speeds up development and reduces maintenance costs, since the operation and maintenance of external monitoring units becomes independent of the system itself. Qn and question Fn can be interconnected, or completely new functions can be used, but ultimately the response Rn only matches the question Qn if the response Fn matches the question Fn, or the response Rn must differ from the correct response to the question Qn if the response Fn does not match the correct response to the question Fn.
[0033] This ultimately means that the adaptation function can only be expressed by a single bit, even though all questions and answers must be kept sufficiently far apart. However, such a single-bit result is "hidden" within the Q / R function within the scope of the proposed method, and not only the questions but also the answers use other variables or another set of variables. In other words, the function H in block AUX is a gating function for the calculated correct answer CR, or if the answer F does not correspond to the correct answer of the system to the question F, the block H in block AUX can stop the calculation or distribution of the correct answer CR for the question Qn, update to the answer Rn during the processing of the question Qn by the external monitoring unit and the direct evaluation of the question, which is sent to the external monitoring unit.
[0034] As already stated, the external problem of the monitoring unit and the internal problem in the controller can be decoupled, i.e. the internal problem and the external problem are different problems generated by a table or by an orthogonal function. The internal problem is generated by the external problem. This also implies that the internal problem is changed even if the external problem remains the same.
[0035] Therefore, except for the processing unit or driver that generates the internal question through the orthogonal function, none of the monitored subsystems of the controller knows the exact question of the external monitoring unit, thereby generating the external question. This means that the internal subsystem cannot estimate the value of the external question and the actual state of the external monitoring unit.
[0036] Internal and external responses can be linked and disconnected from each other, i.e. internal and external responses are different. This also implies that the internal response is changed even if the external response remains the same. This means that the internal response is not the response that is sent to the external monitoring unit, but that the external response only corresponds to the expected response and only if the format and value of the internal response with respect to the internal question are correct. If the internal response does not correspond to the expected value and format of the internal question, the external question has an incorrect value or an incorrect time. This is ensured by the processing unit, which checks the internal response in terms of time and value and sends the correct external response to the function or table.
[0037] The proposed device is used to carry out the described method and is implemented, for example, in hardware and / or software. In this case, the external monitoring unit and / or the controller may be (one or more) components of the device.
[0038] Further advantages and embodiments of the invention are apparent from the description and the accompanying drawings.
[0039] It goes without saying that the features mentioned above and the features still to be explained below can be used not only in the respectively stated combination but also in other combinations or alone, without departing from the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 A system for performing the known method is shown.
[0041] FIG. 2 shows a system for carrying out an embodiment of the proposed method. DETAILED DESCRIPTION
[0042] The present invention is schematically shown according to embodiments in the drawings and is described in detail below with reference to the drawings.
[0043] Figure 1A system for carrying out the known method is shown, wherein the system is generally designated by reference numeral 10. Furthermore, a timeline 12 and a window 14 are depicted, in which an interruption Q 16 and an interruption R 18 are plotted. In the system 10, m subsystems are provided, of which subsystems SUBSYS_1 20, SUBSYS_2 22, SUBSYS_m-1 24 and SUBSYS_m 26 are shown. Furthermore, an evaluation unit 28, also referred to as an evaluator, is provided in the system 10. The evaluation unit 28 outputs a response Rn 30 to the window 14.
[0044] To clarify the time sequence, the previous window 32 is plotted with a dashed line on the left, which received the response Rn-134. On the right, the system is plotted with a dashed line at a later time 50 and the assigned window 52 is plotted, in which the subsequent interruption Q 52 and the subsequent interruption R 54 are also entered. The system 52 outputs the response Rn+156 to the window 52.
[0045] 2 shows a system for carrying out an embodiment of the proposed method, which is generally designated by reference numeral 95. The proposed solution introduces a linking concept and consists of an external monitoring unit 100, a monitoring driver 180 for the external monitoring unit 100, and a monitored system 142. The monitored system 142 consists of one or more monitored subsystems 150, 152, 154, 156, 158.
[0046] The external monitoring unit 100 has an error counter. If the external monitoring unit 100 does not receive a correct response or receives the correct response outside a time window, this increases the error counter. Only when a correct response is received within an open time window is the error counter decremented. If the error counter exceeds a predetermined threshold value, the monitored system or the monitored controller 205 is considered to be faulty and the external monitoring unit 100 issues an initialization to the controller 205 in order to transfer the overall system to a safe state.
[0047] The monitoring unit driver 180 in the controller 205 also monitors the external monitoring unit 100 and compares the behavior of the monitoring unit 100 with the reference model taking into account as many parameters as possible. If the behavior of the external monitoring unit 100 deviates from the expected behavior, the monitoring unit driver 180 in the controller increments the internal error counter INEC. When the external monitoring unit 100 behaves as expected again according to the reference model, this internal error counter can be repaired, i.e. put into an error-free state. The monitoring unit driver 180 can initialize the overall system as well as the external monitoring unit 100 even when the internal error counter INEC in the monitoring unit driver exceeds its threshold value.
[0048] To state in more detail:
[0049] The external monitoring unit 100 requires the correct communication as well as the time type and value type.
[0050] The correct communication of the values is ensured by a correct answer 130, 230 to each question 120, 220 outputted. Since not only the questions but also the answers are only communicated externally, these are referred to as "external" questions and "external" answers.
[0051] The correct communication of time is ensured by checking the time point of the response reception. Each such test cycle starts with a given time period 110. When the time period expires, a given time window 112 is opened. If the response does not arrive, the time window 112 is closed after a certain time, which means a defective system, or if the response arrives, it is closed immediately, i.e. in normal operation. When the time window 112 is closed, the next cycle of the external monitoring unit 100 begins. This means that in a real system, the time periods 112 and 212 are not exactly the same.
[0052] The internal driver 180 in the controller also generates questions and expects correct answers to the questions, but these questions differ from the external questions in format and also in value. Since not only the questions 240, 340 ... but also the answers 170, 270 ... are only communicated internally in the controller, the questions and answers are called "internal" questions and "internal" answers.
[0053] The process is as follows:
[0054] 1. The external monitoring unit 100 evaluates the previous cycle and outputs an external question EQ 120 .
[0055] 2. The controller 205 must read the question Q 120 of the external monitoring unit 100 via the communication channel or the peripheral unit and calculate the final external answer 130 after a given period of time and within the time window 112 of the external monitoring unit 100 regarding the correct behavior of the controller 205. This external answer 130 is only correct or incorrect and is based on the internal answer 170 of the monitored system. This is ensured by the gating unit 182, which enables a correct answer 124 to the external question 120, a correct external answer 224 to the external question 220, etc.
[0056] 3. Immediately after determining the external response 130, the monitoring unit driver outputs at least one new internal question or a plurality of different internal questions 240. The internal question 240 is independent of the external question 120, 220. The internal question 240 is changed even if the external question remains the same. The new or multiple internal questions are available to the monitored system units 150, 152, 156, 158 in the system 142.
[0057] 4. The monitored subsystem calculates a partial response.
[0058] 5. The subsequent cycles are continued in a specific manner. After a specific time, the internal answers are collected by the collection unit 160 and the final internal answer 270 is calculated. The final internal answer 270 is calculated as an exclusive OR function of all internal answers. During this time, the external monitoring unit 100 evaluates the previous cycle and outputs the external question EQ 220. The controller 205 must read the question Q 220 of the external monitoring unit 100 via the communication channel or the peripheral unit through the monitoring unit driver 180 and calculate the final external answer 230 for the correct behavior of the controller 205 after a given period of time and within the time window 212 of the external monitoring unit 100. This external answer is only correct or incorrect and is based on the internal answer 270 of the monitored system or the monitored controller 205. This is ensured by the same gating unit 182, which enables the correct external answer 224 to the external question 220. This means that if the external monitoring unit 100 selects a question from a total of 16 possible questions, for example, it determines the correct answer value 124 or 224 corresponding to the external question 120 or 220 .
Claims
1. A method for monitoring an electronic system (142) having a plurality of subsystems (150, 152, 154, 156, 158) in a controller (205), wherein the method is performed using an external monitoring unit (100) and comprises the following steps: - an external monitoring unit (100) evaluates a previous operating cycle of the system (142) and provides an external question to the controller (205) based on the evaluation, - an internal driver of the external monitoring unit (100) in the controller (205) asks at least one new question to the internal monitoring unit in the controller (205), - after a certain period of time, the controller (205) collects the internal partial responses of all internal monitored units in the controller (205), - A driver of the external monitoring unit (100) in the controller (205) outputs an external response to the external monitoring unit (100).
2. The method according to claim 1, wherein the controller (205) reads the problem of the external monitoring unit (100) via a communication channel or a peripheral unit.
3. The method according to claim 1 or 2, wherein the response of the controller is correct or incorrect.
4. The method according to claim 1, wherein the external monitoring unit (100) of the controller (205) is initialized when too many incorrect responses are detected in terms of time or value.
5. The method according to any one of claims 1 to 4, wherein the controller (205) outputs a new internal question to at least one of the subsystems (150, 152, 154, 156, 158).
6. The method according to any one of claims 1 to 5, wherein the controller (205) outputs a new internal question to all subsystems (150, 152, 154, 156, 158).
7. The method according to any one of claims 1 to 6, wherein the controller (205) outputs new questions for each subsystem (150, 152, 154, 156, 158) or a subgroup of subsystems (150, 152, 154, 156, 158), and these questions are different from each other.
8. The method according to any one of claims 1 to 7, wherein the inner question is completely regenerated in each cycle even if the outer question is the same.
9. The method according to any one of claims 1 to 8, wherein a decoupling is performed between external problems of the external monitoring unit (100) and internal problems in the controller (205).
10. The method according to any one of claims 1 to 9, wherein the number of monitored subsystems (150, 152, 154, 156, 158) is determined continuously.
11. The method according to any one of claims 1 to 9, wherein the number of monitored subsystems (150, 152, 154, 156, 158) is updated in each cycle.
12. The method according to any one of claims 1 to 11, wherein the internal response and the external response are linked and decoupled from each other.
13. A device for monitoring an electronic system (142), the device being configured to carry out the method according to any one of claims 1 to 14. 14 . A computer program having program code means which are designed to carry out the method according to claim 1 , when the computer program is executed on a computing unit, in particular in a device according to claim 13 .
15. A machine-readable storage medium having stored thereon the computer program according to claim 12.
