Scheduling Method and System for Periodically Dependent Tasks in a Heterogeneous Central Computing Platform for Vehicles

Abstract

This application discloses a scheduling method and system for periodically dependent tasks in a vehicle heterogeneous central computing platform, relating to the field of task scheduling. The method includes: establishing a vehicle heterogeneous processing platform; abstractly modeling each dependent task based on the number of dependent tasks running in each heterogeneous computing unit of the vehicle heterogeneous processing platform to obtain an abstract model of dependent tasks for the vehicle heterogeneous processing platform; modeling the scheduling of dependent tasks on the vehicle heterogeneous platform based on the abstract model of dependent tasks to obtain a task scheduling model; the task scheduling model includes a task scheduling cost function and multiple constraints; the constraints include task execution location constraints, task execution time constraints, task round execution time constraints, task execution order constraints, and task data dependency constraints; and solving the task scheduling model to obtain a task scheduling scheme. This application can achieve efficient scheduling of periodically dependent tasks in a vehicle heterogeneous processing platform, improving the utilization rate of computing resources.

Description

Technical Field

[0001] This application relates to the field of task scheduling, and in particular to a scheduling method and system for periodically dependent tasks on a vehicle heterogeneous central computing platform. Background Technology

[0002] With the rapid development of autonomous driving and in-vehicle technologies, the dependent tasks in automotive electronic systems are becoming increasingly complex and their scale is constantly expanding. Due to considerations of space, weight, power consumption (SWaP), and cost, automotive electronic and electrical architectures are gradually shifting from distributed to centralized architectures. This trend is prompting the integration of more and more dependent tasks with different priorities into a unified heterogeneous central computing platform. The priority of a task is usually closely related to its safety criticality; the higher the priority, the more severe the consequences of failure to execute the task on time. For example, an automotive control system may simultaneously run a high-priority environmental perception task and a relatively low-priority driver voice recognition task. In this case, the user can accept a certain degree of voice recognition delay, but cannot accept the problem of the environmental perception task missing its deadline.

[0003] However, traditional scheduling algorithms struggle to effectively distinguish and schedule dependent tasks with different priorities. Furthermore, most current scheduling methods for heterogeneous computing platforms are based on heuristic algorithms, which cannot guarantee a globally optimal solution. Summary of the Invention

[0004] The purpose of this application is to provide a scheduling method and system for periodically dependent tasks in a vehicle heterogeneous central computing platform, which can achieve efficient scheduling of periodically dependent tasks in the vehicle heterogeneous central computing platform, improve the utilization of computing resources, and reduce the total cost of task execution.

[0005] To achieve the above objectives, this application provides the following solution:

[0006] Firstly, this application provides a scheduling method for periodically dependent tasks on a vehicle heterogeneous central computing platform, including:

[0007] A heterogeneous processing platform for automobiles is established; the heterogeneous processing platform for automobiles includes multiple heterogeneous computing units that are connected in communication; each heterogeneous computing unit is used to execute multiple computing tasks contained in each of a number of functions with different priorities;

[0008] Based on the number of dependent tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, an abstract model of each dependent task is created to obtain an abstract model of dependent tasks for the automotive heterogeneous processing platform.

[0009] Based on the aforementioned dependent task abstract model, a task scheduling model is obtained by modeling the dependent task scheduling of the heterogeneous automotive platform. The task scheduling model includes a task scheduling cost function and multiple constraints. The constraints include task execution location constraints, task execution time constraints, task round execution time constraints, task execution order constraints, and task data dependency constraints.

[0010] Solving the task scheduling model yields a task scheduling scheme; the task scheduling scheme includes the heterogeneous computing unit corresponding to each computing task included in each function, the start execution time, and the end execution time; the heterogeneous computing unit corresponding to the computing task is a heterogeneous computing unit used to execute the computing task.

[0011] Secondly, this application provides a scheduling system for periodically dependent tasks on a vehicle heterogeneous central computing platform, comprising:

[0012] A vehicle heterogeneous processing platform construction module is used to establish a vehicle heterogeneous processing platform; the vehicle heterogeneous processing platform includes multiple heterogeneous computing units that are connected in communication; each heterogeneous computing unit is used to execute multiple computing tasks contained in each of a number of functions with different priorities.

[0013] The dependency task abstract model construction module is used to abstract and model each dependency task based on the number of dependency tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, so as to obtain a dependency task abstract model for the automotive heterogeneous processing platform.

[0014] The task scheduling model construction module is used to model the task scheduling of the heterogeneous automotive platform based on the dependent task abstract model to obtain a task scheduling model. The task scheduling model includes a task scheduling cost function and multiple constraints. The constraints include task execution location constraints, task execution time constraints, task round execution time constraints, task execution order constraints, and task data dependency constraints.

[0015] The task scheduling scheme calculation module is used to solve the task scheduling model to obtain a task scheduling scheme; the task scheduling scheme includes the heterogeneous computing unit corresponding to each computing task included in each of the functions, the start execution time and the end execution time; the heterogeneous computing unit corresponding to the computing task is a heterogeneous computing unit used to execute the computing task.

[0016] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0017] This application provides a scheduling method and system for periodically dependent tasks on a vehicle heterogeneous central computing platform. The method includes: constructing a vehicle heterogeneous processing platform that integrates multiple heterogeneous computing units connected via communication links. Each heterogeneous computing unit is responsible for executing multiple functions with different priorities, and each function contains multiple computing tasks. To optimize platform performance, each dependent task is abstracted based on the number of dependent tasks running in the heterogeneous computing unit, forming a dependent task abstract model specifically for the vehicle heterogeneous processing platform. Based on this abstract model, a dependent task scheduling model for the vehicle heterogeneous platform is further constructed. This model includes a task scheduling cost function and a series of constraints. These constraints comprehensively cover requirements regarding task execution location, execution time, round execution time, execution order, and data dependencies, ensuring the accuracy and effectiveness of scheduling. By solving the task scheduling model, a specific task scheduling scheme is derived. This scheme specifies in detail the heterogeneous computing units to be allocated to each computing task under each function, the task's start execution time, and the end execution time. These computing tasks are precisely assigned to heterogeneous computing units capable of efficiently executing them. The solution proposed in this application enables efficient scheduling of periodically dependent tasks in automotive heterogeneous processing platforms, significantly improving the utilization of computing resources. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a scheduling method for periodically dependent tasks on a vehicle heterogeneous central computing platform, provided as an embodiment of this application.

[0020] Figure 2 A typical automotive heterogeneous central computing platform structure diagram provided in an embodiment of this application;

[0021] Figure 3 A typical gesture recognition flowchart provided in an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of a typical gesture recognition DAG provided in an embodiment of this application;

[0023] Figure 5 This is a schematic diagram of a scheduling system structure for a vehicle heterogeneous central computing platform that periodically depends on tasks, provided as an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] Example 1

[0027] like Figure 1 As shown, this embodiment provides a scheduling method for periodically dependent tasks on a vehicle heterogeneous central computing platform, including:

[0028] Step 101: Establish a heterogeneous processing platform for automobiles; the heterogeneous processing platform for automobiles includes multiple heterogeneous computing units that are connected in communication; each heterogeneous computing unit is used to execute multiple computing tasks contained in each of a number of functions with different priorities.

[0029] Step 102: Based on the number of dependent tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, abstract model each dependent task to obtain the dependent task abstract model for the automotive heterogeneous processing platform.

[0030] Step 103: Based on the dependent task abstract model, model the dependent task scheduling of the automotive heterogeneous platform to obtain the task scheduling model; the task scheduling model includes a task scheduling cost function and multiple constraints; the constraints include task execution location constraints, task execution time constraints, task round execution time constraints, task execution order constraints, and task data dependency constraints.

[0031] Step 104: Solve the task scheduling model to obtain a task scheduling scheme; the task scheduling scheme includes the heterogeneous computing unit corresponding to each computing task included in each function, the start execution time and the end execution time; the heterogeneous computing unit corresponding to the computing task is a heterogeneous computing unit used to execute the computing task.

[0032] In some embodiments, when performing step 101, the specific steps may be as follows:

[0033] like Figure 2As shown, in the automotive heterogeneous processing platform, CPUs (general-purpose processors), GPUs (graphics processing units), and MCUs (microcontrollers) are heterogeneous computing units, connected together via a high-speed bus. Specifically, each CPU, GPU, and MCU contains multiple cores. The entire heterogeneous computing platform is defined as follows:

[0034] P = {P1, P2, ..., P} |P|}

[0035] Where |P| represents the number of heterogeneous computing units interconnected via an internal high-speed bus in the heterogeneous system, P i It represents a specific heterogeneous computing unit, specifically referring to the processing core encapsulated in a cluster as a heterogeneous computing unit.

[0036] In some embodiments, when performing step 102, the specific steps may be as follows:

[0037] Based on the number of dependent tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, an automotive dependent task set T is established;

[0038] For each dependent task in the automotive dependent task set T, the dependent task is abstracted into a DAG to obtain a dependent task abstraction model for automotive heterogeneous processing platforms.

[0039] Specifically, the abstract modeling of periodically dependent tasks in a heterogeneous processing platform for automotive hybrid critical systems includes the following steps:

[0040] 1) Establish a set of car-dependent tasks T.

[0041] T = {T1,T2,...,T} |T|}

[0042] Where |T| represents the number of dependent tasks running on the heterogeneous processing platform described in step one in the vehicle system, T k It represents a separate dependency task.

[0043] 2) Perform abstract modeling of individual dependent tasks.

[0044] For a single dependent task T k ,by Figure 3 , Figure 4 For example, Figure 3 This is a diagram illustrating a single dependent task. Figure 4 Is with Figure 3 Correspondingly, each subtask is abstracted as a node (e.g., face detection is node 12), and the set of subtasks is as follows:

[0045] V = {V1,V2,...,V} |V|} represents the dependent task Tk All subtasks.

[0046] The dependencies between subtasks (also known as data dependencies) are abstracted as directed edges, and the set of directed edges E is a |V|×|V| matrix.

[0047] This embodiment will use a single dependent task T k Abstracted as a DAG, it is represented as:

[0048] T k ={V,W,E,Per,Prio}.

[0049] Where V = {V1, V2, ..., V} V|} represents the dependent task T k All subtasks. W is a |P|×|V| matrix, where W represents the worst-case execution time of the dependent tasks on different heterogeneous computing units. ij Represents the dependent task T k Subtask N j In heterogeneous computing unit P i The worst-case execution time is set, which can be obtained through the operating system's timer or debugging tools such as Lauterbach. If a heterogeneous computing unit does not support the subtask, its execution time is set to a positive number much larger than the normal execution time. E represents the directed edges in the DAG, which is a |V|×|V| matrix. If the subtask V j With V j' If the existence of the dependency is between them, then E jj' =1. Per and Prio represent the duration and priority of the dependent task.

[0050] In some embodiments, when performing step 103, the specific steps may be as follows:

[0051] Modeling the scheduling of dependent tasks for heterogeneous automotive platforms based on the dependent task abstraction model specifically includes:

[0052] 1) Find the supercycle of all periodically dependent tasks: the least common multiple of the cycles of all dependent tasks, and the number of times each dependent task is executed within a supercycle:

[0053] {N1,N2,...,N |T|}

[0054] 2) Establish subtask execution constraints, as follows:

[0055]

[0056] The proposed scheduling method does not consider task replication; for any dependent task T... k Subtask V jIt must be executed by one and only one heterogeneous computing unit. Where X... ijkl For binary decision variables (which can only take the values ​​0 or 1), when depending on task T k During the l-th scheduling, subtask V j In heterogeneous computing unit P i The value is 1 when the above is executed, otherwise it is 0.

[0057] 3) Establish time constraints for the execution of subtasks on heterogeneous computing units, as follows:

[0058] S ijkl +C ijkl ≤X ijkl ·L

[0059]

[0060] In the formula, S ijkl and C ijkl They represent the dependent tasks T respectively. k The l-th scheduled subtask V j In heterogeneous computing unit P i The start and end times of the execution are decision variables, and their values ​​range from [0, +∞). ijk For dependent task T k Subtask V j In heterogeneous computing unit P i The worst execution time.

[0061] This embodiment defines when the dependent task T k Subtask V j In heterogeneous computing unit P i During execution, its value represents the actual start and end time of the subtask, respectively; otherwise, the value is 0. L represents a positive number much larger than the system timeout. If dependent task T... k Subtask V j Not in heterogeneous computing unit P i If executed above, then X ijkl =0, using the first formula above to set S ijkl and C ijkl The constraint is 0; if it depends on task T k Subtask V j In heterogeneous computing unit P i If executed above, then X ijkl =1, in the first formula of 3) above, L is a very large positive number, equivalent to S ijkl and C ijkl Without constraints, the second formula in 3) above represents the dependent task T. k The l-th scheduled subtask V j In heterogeneous computing unit P iThe constraint that the difference between the start and end times of the task is at least equal to the difference in time between the start and end times of the task on the heterogeneous computing unit P is also true. i The execution time on the heterogeneous computing unit. If the subtask cannot be executed on the heterogeneous computing unit, its processing time M is a large positive number relative to other processing times, which is obviously not the optimal scheduling scheme, and the result will be discarded during the optimization process.

[0062] 4) Establish execution time constraints for different scheduling rounds of dependent tasks, as follows:

[0063]

[0064] Among them, S kl Represents the dependent task T k The start time of the l-th scheduling, i.e., T k The start time of the first subtask executed during the l-th scheduling; where C kl Represents the dependent task T k The end execution time of the l-th scheduling, i.e., T k The execution time of the last subtask executed during the l-th scheduling is given by the formula above, which represents the execution time of the dependent task T. k The execution in the l-th round needs to begin after the start of the l-th execution cycle. The fourth formula above represents the dependent task T. k The execution in round l requires the execution to begin after round l-1. The fifth formula above represents the dependent task T. k The execution must end within the l-th execution cycle to complete the execution in the l-th round, delta kl Indicates the dependency task T k The measure of exceeding its deadline in the l-th cycle.

[0065] 5) Establish execution order constraints for subtasks on heterogeneous computing units, as follows:

[0066] S ijkl ≥C ij'k'l' -(Y ijj'kk'll' )·L

[0067]

[0068] In the formula, Y ijj'kk'll' C is a binary decision variable. ij'k'l' For dependent task T k' The l'th scheduling subtask V j' In heterogeneous computing unit P i The start time decision variable for execution, S ij'k'l' For dependent task T k' The l'th scheduling subtask V j'In heterogeneous computing unit P i The start time decision variable for execution.

[0069] The proposed scheduling method does not consider task preemption; that is, once a subtask begins execution on a heterogeneous computing unit, it will occupy that unit until the subtask completes. Each heterogeneous computing unit can only operate on a single thread, meaning it can only process one subtask at a time. Assigning different subtasks to the same heterogeneous computing unit requires deciding their computation order. Define the dependency task T... k The l-th scheduled subtask V j In heterogeneous computing unit P i Prior to dependent task T k' The l'th scheduling subtask V j' Binary decision variable Y during execution ijj'kk'll' =1 equals 1, otherwise it equals 0. These two constraints constrain the end time of the preceding subtask and the start time of the following subtask in two tasks executed sequentially on the same heterogeneous computing unit. Specifically, k and k' can be equal, and l and l' can be equal. In other words, the above formula can model the execution order of subtasks with different dependencies and different scheduling rounds, and is also applicable to modeling the execution order of subtasks with the same dependency and different scheduling rounds.

[0070] 6) Establish data dependency constraints for subtasks, as follows:

[0071]

[0072] Each dependent task has subtasks that have data dependencies; these subtasks must receive data from their predecessor task before they can begin execution. and They represent the dependent tasks T respectively. k During the l-th scheduling, subtask V j and V j' Start time and end time of execution.

[0073] 7) Task scheduling cost function modeling, as detailed below:

[0074]

[0075] In the formula, These are weighting coefficients related to the priority of dependent tasks. Higher priority tasks have larger weighting coefficients; typically, a one-level increase in priority increases the weighting coefficient tenfold or more. The different real-time requirements of dependent tasks with different priorities are reflected by a weighted sum of the results exceeding the deadline. To represent the dependent task T k Delta is the ratio of the completion time to the deadline of the l-th cycle.kl To represent the dependent task T k The measure of exceeding its deadline in the l-th cycle.

[0076] In some embodiments, when performing step 104, the specific steps may be as follows:

[0077] Choose a programming language (such as C, C++, JAVA, or Python) and its IDE, model the above mathematical model according to the interface requirements of a solver (such as Gurobi, SCIP, or COPT), call the corresponding solver to solve it, and obtain the optimal scheduling result.

[0078] In this embodiment, the final scheduling result focuses on the solution results of three decision variables: X, S, and C. X reflects which heterogeneous computing unit each subtask is executed on, while S and C reflect the start and end times of each subtask's execution on a specific heterogeneous computing unit.

[0079] Example 2

[0080] like Figure 5 As shown, this embodiment provides a scheduling method for periodically dependent tasks on a vehicle heterogeneous central computing platform, including:

[0081] The automotive heterogeneous processing platform construction module 501 is used to establish an automotive heterogeneous processing platform; the automotive heterogeneous processing platform includes multiple heterogeneous computing units that are connected in communication; each heterogeneous computing unit is used to execute multiple computing tasks contained in each of a number of functions with different priorities.

[0082] The dependency task abstract model construction module 502 is used to abstract and model each dependency task according to the number of dependency tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, so as to obtain a dependency task abstract model for the automotive heterogeneous processing platform.

[0083] The task scheduling model construction module 503 is used to model the task scheduling of the heterogeneous automotive platform based on the dependent task abstract model to obtain a task scheduling model. The task scheduling model includes a task scheduling cost function and multiple constraints. The constraints include task execution location constraints, task execution time constraints, task round execution time constraints, task execution order constraints, and task data dependency constraints.

[0084] The task scheduling scheme calculation module 504 is used to solve the task scheduling model to obtain a task scheduling scheme; the task scheduling scheme includes the heterogeneous computing unit corresponding to each computing task included in each of the functions, the start execution time and the end execution time; the heterogeneous computing unit corresponding to the computing task is a heterogeneous computing unit used to execute the computing task.

[0085] In summary, this application has the following technical effects:

[0086] 1) The heterogeneous computing platform periodic task scheduling method provided in this application determines the optimal scheduling strategy of the system through mixed integer linear programming, ensuring the orderly execution of each task application, improving the overall vehicle safety and the utilization efficiency of computing resources.

[0087] 2) The abstraction and modeling method of the heterogeneous central computing platform for automobiles proposed in this application does not require detailed internal chip layout of the heterogeneous central computing platform and does not require specific models of heterogeneous computing units.

[0088] 3) The task-dependent abstract modeling method proposed in this application takes into account the heterogeneity, parallelism and priority of the system.

[0089] 4) This application establishes a mathematical model for scheduling periodically dependent tasks from a theoretical perspective, enabling the method to obtain the optimal scheduling for the entire system.

[0090] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0091] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A scheduling method for periodically dependent tasks on a vehicle heterogeneous central computing platform, characterized in that, include: Establish a heterogeneous processing platform for automobiles; The automotive heterogeneous processing platform includes multiple heterogeneous computing units that are connected in communication. Each heterogeneous computing unit is used to execute multiple computing tasks contained in each of a number of functions with different priorities; Based on the number of dependent tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, an abstract model of each dependent task is created to obtain an abstract model of dependent tasks for the automotive heterogeneous processing platform. Based on the aforementioned dependent task abstract model, a task scheduling model is obtained by modeling the dependent task scheduling of the heterogeneous automotive platform. The task scheduling model includes a task scheduling cost function and multiple constraints. The constraints include task execution location constraints, task execution time constraints, task round execution time constraints, task execution order constraints, and task data dependency constraints. Solving the task scheduling model yields a task scheduling scheme; the task scheduling scheme includes the heterogeneous computing unit corresponding to each computing task included in each function, the start execution time, and the end execution time; the heterogeneous computing unit corresponding to the computing task is a heterogeneous computing unit used to execute the computing task; The aforementioned automotive heterogeneous processing platform specifically comprises: P={P1,P2,...,P |P| }; Where |P| represents the number of heterogeneous processing units interconnected via an internal high-speed bus in the heterogeneous system, P i This represents the i-th heterogeneous processing unit; The task execution location constraints in the task scheduling model are specifically as follows: Among them, X ijkl For binary decision variables, |V k |For dependent task T k The number of subtasks in the process; N k Let l be the number of times the dependent task is executed, where l is the l-th execution. The task execution time constraints in the task scheduling model are specifically as follows: Among them, S ijkl and C ijkl They represent the dependent tasks T respectively. k The l-th scheduled subtask V j In heterogeneous computing unit P i The start and end times of the execution are the decision variables, W. ijk For dependent task T k Subtask V j In heterogeneous computing unit P i Worst execution time; The task execution order constraints in the task scheduling model are specifically as follows: Among them, Y ijj'kk'll' C is a binary decision variable. ij'k'l' For dependent task T k' The l'th scheduling subtask V j' In heterogeneous computing unit P i The start time decision variable for execution, S ij'k'l' For dependent task T k' The l'th scheduling subtask V j' In heterogeneous computing unit P i The start time decision variable for execution; The task round execution time constraints are as follows: Among them, S kl Represents the dependent task T k The start time of the l-th scheduling, C kl Represents the dependent task T k The completion time of the l-th cycle, lPer k Represents the dependent task T k At the end of the l-th period, delta kl Indicates the dependency task T k C is the measure of exceeding its deadline in the l-th period. kl-1 Represents the dependent task T k The end execution time of the (l-1)th scheduling.

2. The scheduling method for periodically dependent tasks of a vehicle heterogeneous central computing platform according to claim 1, characterized in that, The heterogeneous computing unit includes an image processor, a general-purpose processor, and a microcontroller.

3. The scheduling method for periodically dependent tasks of a vehicle heterogeneous central computing platform according to claim 2, characterized in that, Based on the number of dependent tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, an abstract model is created for each dependent task to obtain an abstract model of dependent tasks for the automotive heterogeneous processing platform, specifically including: Based on the number of dependent tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, an automotive dependent task set T is established; For each dependent task in the automotive dependent task set T, the dependent task is abstracted into a DAG to obtain a dependent task abstraction model for automotive heterogeneous processing platforms.

4. The scheduling method for periodically dependent tasks of a vehicle heterogeneous central computing platform according to claim 3, characterized in that, The specific abstract model of the dependent task is as follows: T k ={V,W,E,Per,Prio}; Where V = {V1, V2, ..., V} V| } is the dependent task T k All subtasks; W is a |P|×|V| matrix, where W represents the worst execution time of the dependent task on different heterogeneous computing units, and E represents the directed edges in the DAG. Per and Prio represent the period and priority of the dependent task.

5. The scheduling method for periodically dependent tasks in a vehicle heterogeneous central computing platform according to claim 4, characterized in that, The task scheduling cost function in the task scheduling model is specifically as follows: in, These are weighting coefficients related to the priority of dependent tasks. To represent the dependent task T k The ratio of the completion time to the deadline of the l-th cycle.

6. A scheduling system for periodically dependent tasks on a vehicle heterogeneous central computing platform, used to implement the scheduling method for periodically dependent tasks on a vehicle heterogeneous central computing platform as described in any one of claims 1-5, characterized in that, include: The automotive heterogeneous processing platform building module is used to establish an automotive heterogeneous processing platform. The automotive heterogeneous processing platform includes multiple heterogeneous computing units that are connected in communication. Each heterogeneous computing unit is used to execute multiple computing tasks contained in each of a number of functions with different priorities; The dependency task abstract model construction module is used to abstract and model each dependency task based on the number of dependency tasks running by each heterogeneous computing unit in the automotive heterogeneous processing platform, so as to obtain a dependency task abstract model for the automotive heterogeneous processing platform. The task scheduling model construction module is used to model the task scheduling of the heterogeneous automotive platform based on the dependent task abstract model to obtain a task scheduling model. The task scheduling model includes a task scheduling cost function and multiple constraints. The constraints include task execution location constraints, task execution time constraints, task round execution time constraints, task execution order constraints, and task data dependency constraints. The task scheduling scheme calculation module is used to solve the task scheduling model to obtain a task scheduling scheme; the task scheduling scheme includes the heterogeneous computing unit corresponding to each computing task included in each of the functions, the start execution time and the end execution time; the heterogeneous computing unit corresponding to the computing task is a heterogeneous computing unit used to execute the computing task.

Images