Controller configuration optimization method, intelligent device and computer readable storage medium

Abstract

The invention relates to the technical field of electronic and electrical architecture optimization, particularly provides a controller configuration optimization method, intelligent equipment and a computer readable storage medium, and aims to solve the problems that scheme comparison cannot be effectively performed and a configuration optimization scheme cannot be automatically determined when a controller is configured at an intelligent equipment end in the prior art. In order to achieve the purpose, the method provided by the invention comprises the following steps: constructing a multi-objective optimization model of a configuration controller at an intelligent equipment end based on a decision variable, a constraint condition and a plurality of objective functions; the multi-objective optimization model is converted into a single-objective optimization model, and topological structure parameters and model configuration parameters of the intelligent equipment end are used as input; the number is set for different controllers, the single-target optimization model is solved, and corresponding configuration parameters are obtained; and obtaining a target configuration parameter of the controller based on the configuration parameters of the plurality of controller setting numbers. According to the method, the controller configuration can be automatically evaluated, the advantages and disadvantages of a plurality of configuration schemes are compared, and the optimal configuration scheme is determined.

Description

Technical Field

[0001] The present application relates to the technical field of electronic and electrical architecture optimization, and in particular to a controller configuration optimization method, an intelligent device, and a computer-readable storage medium. Background Art

[0002] With the development of smart electric vehicles and the emergence of more and more new functions and application scenarios, the functional requirements for input and output devices installed on the vehicle body, such as sensors and actuators, are also increasing.

[0003] The increase in input and output devices has brought challenges to the design of the vehicle's electronic and electrical architecture. Traditional solutions mainly rely on the design experience of engineers, so it is impossible to effectively quantify the design solutions. There are also problems such as inconsistent standards, poor reusability of the evaluation system, and inability to compare the advantages and disadvantages of multiple design solutions. Summary of the invention

[0004] In order to overcome the above-mentioned defects, the present application is proposed to solve or at least partially solve the technical problem that the existing controller configured on the device side cannot effectively compare solutions and automatically determine the optimization solution.

[0005] In a first aspect, a controller configuration optimization method is provided, comprising:

[0006] Based on decision variables, constraints and objective functions, a multi-objective optimization model for configuring a controller on the smart device side is constructed; wherein the objective function includes a delay cost objective function, a harness length objective function and an equipment cost objective function;

[0007] Based on the delay cost objective function, the harness length objective function, and the equipment cost objective function, converting the multi-objective optimization model into a single-objective optimization model by weighting;

[0008] Obtaining topological structure parameters and model configuration parameters of the smart device end, and inputting them into the single-objective optimization model, wherein the model configuration parameters at least include a maximum number of controller settings and a preset objective function weight;

[0009] For different numbers of controller settings, the single-objective optimization model is solved respectively to obtain configuration parameters corresponding to the number of controller settings; the target configuration parameters of the controller are obtained based on multiple configuration parameters of the number of controller settings; the number of controller settings is less than or equal to the maximum number of controller settings.

[0010] In some embodiments, the smart device terminal includes a plurality of functional areas, and the functional areas are provided with input and output devices to be connected with the controller;

[0011] The decision variables include at least one of a controller deployment position variable, a controller connection path variable, a direct connection path variable between the functional area of ​​the smart device and the controller, a wiring harness connection variable, a computing resource occupancy variable of the input and output devices corresponding to the functional area, a path selection variable, and a controller deployment type variable;

[0012] The constraint conditions include at least one of the node controller quantity and type allocation constraints of each functional area of ​​the smart device, the communication connectivity rule constraints between different empty controllers, the connectivity extension relationship constraints of the functional area without controller, the controller input and output capacity constraints, the chip computing power constraints and the connectivity rule constraints.

[0013] In some embodiments, the model configuration parameters also include controller types, including domain controllers, electronic controllers, and interface centers.

[0014] In some embodiments, the method further comprises constructing a topological structure diagram of the smart device terminal using a graph theory method, wherein the topological structure diagram comprises a plurality of nodes and an edge connecting two nodes; the nodes are used to represent a functional area of ​​the smart device terminal, and the edges are used to represent a connection path formed between two functional areas;

[0015] The delay cost objective function is obtained based on the sum of the path delay cost generated by the domain controller through each connection path and node connection and the cost generated by the unit cost difference between the electronic controller and the interface center. The delay cost objective function is expressed as:

[0016] in,

[0017]

[0018] And δ ij The following constraints are met:

[0019] V represents the location coordinate set of the node, V zoner represents the location coordinate set corresponding to the domain controller, p represents the connected path, P ij represents the set of connected paths between nodes i and j, m represents the index of the domain controller, M represents the set of domain controllers, delay p represents the connectivity delay cost, The path selection variable representing the connection between the controller of node i and the domain controller by path p, expn ECU represents the unit cost of the electronic controller, expn Hubrepresents the unit cost of the interface center, price represents the unit delay cost, and δ ij represents the auxiliary decision variable, represents the indicator matrix of path p passing through edge e, E represents the set of undirected edges connecting all feasible location nodes, x i,Dum The controller deployment location variable representing the empty controller deployed on node i, x j,ECU The controller deployment location variable representing the electronic controller deployed at node i, Represents the direct connection path variable between the functional area and the controller.

[0020] In some embodiments, the wire harness length objective function is obtained based on the connectivity distance and the wire harness connection variable, and the wire harness length objective function is expressed as:

[0021]

[0022] Where e represents the edge index, E represents the set of undirected edges connecting all feasible location nodes, m represents the index of the domain controller, M represents the set of domain controllers, and dist e represents the connectivity distance of edge e, Represents a harness connection variable.

[0023] In some embodiments, the equipment cost objective function is obtained based on controller type, deployment location, unit cost and controller deployment location variables, and the equipment cost objective function is expressed as:

[0024] ∑ k∈CU ∑ i∈V expn k ·x ik

[0025] Among them, CU represents the controller type set, V represents the node position coordinate set, expn k represents the unit cost of type k controller, x ik A controller deployment location variable representing the deployment of a type k controller on node i.

[0026] In some embodiments, obtaining the topological structure parameters of the smart device includes:

[0027] Based on the multiple functional areas of the smart device and the connection paths between the multiple functional areas, a topological structure diagram of the smart device is constructed by using a graph theory method; the topological structure diagram includes multiple nodes and edges connecting two nodes, the nodes are used to represent the functional areas of the smart device, and the edges are used to represent the connection paths formed between the two functional areas;

[0028] Obtain node information of each node and edge information of each edge in the topological structure diagram; wherein the node information includes at least one of a node identifier, a node name, an activation state, a node position, a number of input interfaces, a number of output interfaces, and an identifier of a controller type that is allowed to be placed; and the edge information includes at least one of an edge identifier, an edge start node identifier, an edge end node identifier, and an edge length.

[0029] In some embodiments, obtaining the model configuration parameters further includes:

[0030] Acquire controller information, harness information, and input / output device information of each functional area at the smart device end; wherein the controller information includes at least one of a controller identification, a controller name, a controller type, a controller input / output capacity, a controller energy consumption, a controller size, and a unit cost of the controller; the harness information includes at least one of a unit harness cost, a harness type, a harness power, and a unit delay cost; and the input / output device information includes at least one of an interface type and a number of interfaces of the input / output device.

[0031] In some embodiments, the converting the multi-objective optimization model into a single-objective optimization model by weighting based on the delay cost objective function, the harness length objective function, and the equipment cost objective function comprises:

[0032] Summing the product of the sum of the harness length objective function and the equipment cost objective function and a first weight, and the product of the delay cost objective function and a second weight;

[0033] The single-objective optimization model is obtained with the goal of minimizing the sum; wherein the sum of the first weight and the second weight is 1, and the preset objective function weight is the first weight or the second weight.

[0034] In some embodiments, for different numbers of controller settings, the single-objective optimization model is solved respectively, and the configuration parameters corresponding to the number of controller settings are obtained, including:

[0035] According to different controller settings, the single-objective optimization model is divided into the main problem of optimal address allocation of controllers and the sub-problem of optimal connection paths between controllers and nodes by using Benders algorithm.

[0036] The main problem and the sub-problems are iteratively solved until a preset convergence condition is met, and configuration parameters corresponding to the controller setting quantity are obtained.

[0037] In some embodiments, the master problem is obtained by summing the product of the first weight and the equipment cost objective function and the product of the second weight and the cost resulting from the unit cost difference between the electronic controller and the interface center;

[0038] The sub-problem is obtained by summing the product of the first weight and the bundle length objective function and the product of the second weight and the path delay cost generated by the domain controller through each feasible connection path and node connection.

[0039] In some embodiments, iteratively solving the sub-problem includes: solving the sub-problem using a column generation algorithm.

[0040] In some embodiments, obtaining the target configuration parameters of the controller based on a plurality of configuration parameters of the controller setting quantity includes:

[0041] For different controller setting numbers, obtaining the total objective function value of the single-objective optimization model based on the configuration parameters corresponding to the current controller setting number;

[0042] Comparing the total objective function values ​​of a plurality of controller settings;

[0043] According to the comparison results, the number of controller settings and configuration parameters corresponding to the minimum total objective function value are used as the target configuration parameters of the controller; wherein the target configuration parameters include the number of controllers, the type of controller, the location of the controller and the connection path of the controller.

[0044] In a second aspect, the present application provides a smart device, comprising:

[0045] at least one processor;

[0046] and, a memory communicatively coupled to the at least one processor;

[0047] Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, any of the controller configuration optimization methods described above is implemented.

[0048] In a third aspect, the present application provides a computer-readable storage medium storing a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute any of the controller configuration optimization methods described above.

[0049] Solution 1. A controller configuration optimization method, characterized by comprising:

[0050] Based on decision variables, constraints and objective functions, a multi-objective optimization model for configuring a controller on the smart device side is constructed; wherein the objective function includes a delay cost objective function, a harness length objective function and an equipment cost objective function;

[0051] Based on the delay cost objective function, the harness length objective function, and the equipment cost objective function, converting the multi-objective optimization model into a single-objective optimization model by weighting;

[0052] Obtaining topological structure parameters and model configuration parameters of the smart device end, and inputting them into the single-objective optimization model, wherein the model configuration parameters at least include a maximum number of controller settings and a preset objective function weight;

[0053] For different numbers of controller settings, the single-objective optimization model is solved respectively to obtain configuration parameters corresponding to the number of controller settings; the target configuration parameters of the controller are obtained based on multiple configuration parameters of the number of controller settings; the number of controller settings is less than or equal to the maximum number of controller settings.

[0054] Solution 2. The controller configuration optimization method according to Solution 1 is characterized in that the smart device terminal includes a plurality of functional areas, and the functional areas are provided with input and output devices to be connected to the controller;

[0055] The decision variables include at least one of a controller deployment position variable, a controller connection path variable, a direct connection path variable between the functional area of ​​the smart device and the controller, a wiring harness connection variable, a computing resource occupancy variable of the input and output devices corresponding to the functional area, a path selection variable, and a controller deployment type variable;

[0056] The constraint conditions include at least one of the node controller quantity and type allocation constraints of each functional area of ​​the smart device, the communication connectivity rule constraints between different empty controllers, the connectivity extension relationship constraints of the functional area without controller, the controller input and output capacity constraints, the chip computing power constraints and the connectivity rule constraints.

[0057] Solution 3. The controller configuration optimization method according to Solution 2 is characterized in that the model configuration parameters also include controller types, and the controller types include domain controllers, electronic controllers and interface centers.

[0058] Solution 4. The controller configuration optimization method according to Solution 3 is characterized in that the method further comprises using a graph theory method to construct a topological structure diagram of the smart device end, wherein the topological structure diagram comprises a plurality of nodes and an edge connecting two nodes; the nodes are used to represent the functional areas of the smart device end, and the edges are used to represent the connection path formed between the two functional areas;

[0059] The delay cost objective function is obtained based on the sum of the path delay cost generated by the domain controller through each connection path and node connection and the cost generated by the unit cost difference between the electronic controller and the interface center. The delay cost objective function is expressed as:

[0060] in,

[0061]

[0062] And δ ij The following constraints are met:

[0063] V represents the location coordinate set of the node, V zoner represents the location coordinate set corresponding to the domain controller, p represents the connected path, P ij represents the set of connected paths between nodes i and j, m represents the index of the domain controller, M represents the set of domain controllers, delay p represents the connectivity delay cost, The path selection variable representing the connection between the controller of node i and the domain controller by path p, expn ECU represents the unit cost of the electronic controller, expn Hub represents the unit cost of the interface center, price represents the unit delay cost, and δ ij represents the auxiliary decision variable, represents the indicator matrix of path p passing through edge e, E represents the set of undirected edges connecting all feasible location nodes, x i,Dum The controller deployment location variable representing the empty controller deployed on node i, x j,ECU The controller deployment location variable representing the electronic controller deployed at node i, Represents the direct connection path variable between the functional area and the controller.

[0064] Solution 5. The controller configuration optimization method according to Solution 3 is characterized in that the harness length objective function is obtained based on the connectivity distance and the harness connection variable, and the harness length objective function is expressed as:

[0065]

[0066] Where e represents the edge index, E represents the set of undirected edges connecting all feasible location nodes, m represents the index of the domain controller, M represents the set of domain controllers, and dist e represents the connectivity distance of edge e, Represents a harness connection variable.

[0067] Solution 6. The controller configuration optimization method according to Solution 3 is characterized in that the equipment cost objective function is obtained based on the controller type, deployment location, unit cost and controller deployment location variables, and the equipment cost objective function is expressed as:

[0068] ∑ k∈CU ∑ i∈V expn k ·x ik

[0069] Among them, CU represents the controller type set, V represents the node position coordinate set, expn k represents the unit cost of type k controller, x ik A controller deployment location variable representing the deployment of a type k controller on node i.

[0070] Solution 7. The controller configuration optimization method according to Solution 1 is characterized in that obtaining the topological structure parameters of the smart device terminal includes:

[0071] Based on the multiple functional areas of the smart device and the connection paths between the multiple functional areas, a topological structure diagram of the smart device is constructed by using a graph theory method; the topological structure diagram includes multiple nodes and edges connecting two nodes, the nodes are used to represent the functional areas of the smart device, and the edges are used to represent the connection paths formed between the two functional areas;

[0072] Obtain node information of each node and edge information of each edge in the topological structure diagram; wherein the node information includes at least one of a node identifier, a node name, an activation state, a node position, a number of input interfaces, a number of output interfaces, and an identifier of a controller type that is allowed to be placed; and the edge information includes at least one of an edge identifier, an edge start node identifier, an edge end node identifier, and an edge length.

[0073] Solution 8. The controller configuration optimization method according to Solution 1 is characterized in that obtaining the model configuration parameters further comprises:

[0074] Acquire controller information, harness information, and input / output device information of each functional area at the smart device end; wherein the controller information includes at least one of a controller identification, a controller name, a controller type, a controller input / output capacity, a controller energy consumption, a controller size, and a unit cost of the controller; the harness information includes at least one of a unit harness cost, a harness type, a harness power, and a unit delay cost; and the input / output device information includes at least one of an interface type and a number of interfaces of the input / output device.

[0075] Solution 9. The controller configuration optimization method according to any one of Solutions 1 to 8, characterized in that converting the multi-objective optimization model into a single-objective optimization model by weighting based on the delay cost objective function, the harness length objective function, and the equipment cost objective function comprises:

[0076] Summing the product of the sum of the harness length objective function and the equipment cost objective function and a first weight, and the product of the delay cost objective function and a second weight;

[0077] The single-objective optimization model is obtained with the goal of minimizing the sum; wherein the sum of the first weight and the second weight is 1, and the preset objective function weight is the first weight or the second weight.

[0078] Solution 10. The controller configuration optimization method according to any one of solutions 1 to 8 is characterized in that, for different controller setting numbers, the single-objective optimization model is solved respectively, and the configuration parameters corresponding to the controller setting numbers are obtained, including:

[0079] According to different controller settings, the single-objective optimization model is divided into the main problem of optimal address allocation of controllers and the sub-problem of optimal connection paths between controllers and nodes by using Benders algorithm.

[0080] The main problem and the sub-problems are iteratively solved until a preset convergence condition is met, and configuration parameters corresponding to the controller setting quantity are obtained.

[0081] Solution 11. The controller configuration optimization method according to Solution 10 is characterized in that the main problem is obtained by the sum of the product of the first weight and the equipment cost objective function and the product of the second weight and the cost generated by the unit cost difference between the electronic controller and the interface center;

[0082] The sub-problem is obtained by summing the product of the first weight and the bundle length objective function and the product of the second weight and the path delay cost generated by the domain controller through each feasible connection path and node connection.

[0083] Solution 12. The controller configuration optimization method according to Solution 10 is characterized in that iteratively solving the sub-problem includes: using a column generation algorithm to solve the sub-problem.

[0084] Solution 13. The controller configuration optimization method according to Solution 1 is characterized in that obtaining the target configuration parameters of the controller based on a plurality of configuration parameters of the controller setting quantity comprises:

[0085] For different controller setting numbers, obtaining the total objective function value of the single-objective optimization model based on the configuration parameters corresponding to the current controller setting number;

[0086] Comparing the total objective function values ​​of a plurality of controller settings;

[0087] According to the comparison results, the number of controller settings and configuration parameters corresponding to the minimum total objective function value are used as the target configuration parameters of the controller; wherein the target configuration parameters include the number of controllers, the type of controller, the location of the controller and the connection path of the controller.

[0088] Solution 14. A smart device, comprising:

[0089] at least one processor;

[0090] and, a memory communicatively coupled to the at least one processor;

[0091] Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the controller configuration optimization method described in any one of Schemes 1 to 13 is implemented.

[0092] Solution 15. A computer-readable storage medium storing a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute the controller configuration optimization method described in any one of Solutions 1 to 13.

[0093] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:

[0094] In implementing the controller configuration optimization method provided by the present application, a multi-objective optimization model for configuring the controller on the smart device side is constructed based on decision variables, constraints, delay cost objective function, harness length objective function and equipment cost objective function; the multi-objective optimization model is converted into a single-objective optimization model; the topological structure parameters and model configuration parameters of the smart device side are obtained, and the single-objective optimization model is input, and the model configuration parameters at least include the maximum number of controller settings and the preset objective function; for different controller setting numbers, the single-objective optimization model is solved respectively to obtain the configuration parameters corresponding to the controller setting number; the target configuration parameters of the controller are obtained based on the configuration parameters of multiple controller setting numbers. This method can effectively quantify the controller configuration, compare the advantages and disadvantages of multiple configuration schemes, determine the optimal configuration scheme, and improve the efficiency, flexibility and reusability of controller optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] The disclosure of the present application will become easier to understand with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present application. Among them:

[0096] Figure 1 It is a flow chart of a controller configuration optimization method provided in an application embodiment;

[0097] Figure 2 is a flowchart of the execution method of step S4 provided in the application embodiment;

[0098] Figure 3 It is a schematic diagram of the architecture of the vehicle end to be configured provided by a specific embodiment of the present application;

[0099] Figure 4 It is a node information table corresponding to multiple functional areas of the vehicle end to be configured provided in an embodiment of the present application;

[0100] Figure 5 It is a side information table corresponding to multiple functional areas of the vehicle to be configured provided in an embodiment of the present application;

[0101] Figure 6 It is a controller information table to be configured provided in the embodiment of the present application;

[0102] Figure 7 It is a cost distribution diagram corresponding to configuring different numbers of controllers when the preset function target weight is 0.1 provided in a specific embodiment of the present application;

[0103] Figure 8 It is a schematic diagram of the vehicle-side topology structure corresponding to the configuration of two Zoner controllers provided in a specific embodiment of the present application;

[0104] Fig. 9 Another specific embodiment of the present application provides a cost distribution diagram corresponding to configuring different numbers of controllers when the preset function target weight is 0.9;

[0105] Fig.10 is a schematic diagram of a vehicle-side topological structure provided by another specific embodiment of the present application;

[0106] Fig.11 It is a schematic diagram of the structure of the smart device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0107] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the protection scope of the present application.

[0108] Based on the description of the above background technology, it can be seen that the design of the vehicle's electronic and electrical architecture in the relevant technology mainly relies on the design experience of engineers, so it is impossible to effectively quantify the design scheme, and it is easy to have problems such as inconsistent standards, poor reusability of the evaluation system, and inability to compare the advantages and disadvantages of multiple design schemes. In view of this, the present application provides a controller configuration optimization method to achieve multi-objective automated optimization and improve optimization efficiency, flexibility and reusability.

[0109] First, the setting idea of ​​the controller configuration optimization method provided by the present application is described. The controller configuration optimization method provided by the embodiment of the present application can be applied to the scenario of controller configuration for each functional area of ​​the smart device end. The configuration goal is to assign a controller to each functional area, determine the type, quantity, and setting location of the controller, and further set the connection path between the controllers to realize the functions in each functional area, while meeting the optimization goals of minimizing costs and maximizing connectivity of each functional area. Different objective functions can be set based on the optimization goal. In addition, the configuration of the controller also needs to meet the architecture design rules of the current smart device end. Different constraints can be set based on the architecture design rules. By combining the objective function and the constraints, the configuration scheme under different controller setting numbers can be determined, and the configuration optimization scheme that meets the constraints and optimization goals is determined by comparison, and the target configuration parameters of the controller are obtained.

[0110] In some embodiments, the smart device end may be the vehicle end, and the functional area may be divided based on the type and / or location of the implemented function. As an example, the functional area may include the right front door area, the left front door area, the right front seat area, and the left front seat area, etc. Each functional area may be provided with input and output devices for implementing the corresponding functions of the current area, such as at least one of a sensor and an actuator. Different functional areas may be connected by means of a physical harness or network for signal communication and functional interaction. The connectivity between different functional areas may be evaluated based on at least one of communication efficiency, topological complexity, connection distance, energy consumption, and delay cost. In an embodiment of the present application, when maximizing the connectivity of each functional area is the optimization goal, the connectivity evaluation is performed based on the delay cost, and the delay cost objective function may be set accordingly. When minimizing the cost is the optimization goal, the harness cost and the controller cost are evaluated, and the harness length objective function and the device cost objective function may be set accordingly.

[0111] Based on this, the present application provides a controller configuration optimization method, which can construct an optimization model based on multiple optimization objectives and constraints, realize automatic and effective quantitative evaluation based on the optimization model, and can compare configuration schemes with different controller setting quantities to determine the optimal configuration scheme. Figure 1 As shown, Figure 1 : is a flow chart of a controller configuration optimization method provided in an embodiment of the present application, which may include:

[0112] Step S1: Based on decision variables, constraints and objective functions, a multi-objective optimization model for configuring a controller on the smart device side is constructed; wherein the objective functions include a delay cost objective function, a harness length objective function and an equipment cost objective function;

[0113] Step S2: based on the delay cost objective function, the harness length objective function and the equipment cost objective function, the multi-objective optimization model is converted into a single-objective optimization model by weighting;

[0114] Step S3: obtaining topological structure parameters and model configuration parameters of the smart device, and inputting the single-objective optimization model, wherein the model configuration parameters at least include the maximum number of controller settings and the preset objective function weight;

[0115] Step S4: for different numbers of controller settings, solving the single-objective optimization model respectively to obtain configuration parameters corresponding to the number of controller settings; and obtaining target configuration parameters of the controller based on the configuration parameters of multiple numbers of controller settings.

[0116] The controller setting quantity is less than or equal to the maximum setting quantity of the controller.

[0117] In an embodiment of the present application, decision variables and objective functions can be set in advance based on the optimization objectives of the controller configuration, and constraints can be set according to the architecture design rules of the smart device side.

[0118] Among them, decision variables are key variables that affect the realization of optimization goals. For example, when optimizing controller configuration, the deployment location, quantity, deployment type, and connection paths between controllers directly affect the cost and connectivity, thereby affecting the optimization results.

[0119] The architecture design rules may include at least one of the controller types that can be set in the functional area, the connectivity constraints between different types of controllers, the connectivity constraints for the functional area without a controller, the computing power required by the controller, the interface capacity that the controller can achieve, and the connectivity between different functional areas through the wiring harness. The constraints can be set based on the architecture design rules. It should be noted that the architecture design rules can be selected according to the actual needs of the vehicle-side electronic and electrical architecture.

[0120] In some embodiments, the delay cost objective function can be constructed based on the connectivity paths between different functional areas and the unit delay cost. The harness length objective function can be constructed based on the connectivity distance and harness connection variables. The device cost objective function can be constructed based on the controller type, deployment location, unit cost, and controller deployment location variables.

[0121] In some embodiments, step S1 can be specifically based on the optimization goals of minimizing cost and maximizing connectivity of each functional area, using decision variables, constraints, and delay cost objective function, harness length objective function, and equipment cost objective function to build a multi-objective optimization model. Among them, the delay cost objective function is minimized to determine whether the optimization goal of maximizing connectivity of each functional area is achieved; the sum of the harness length objective function and the equipment cost objective function is minimized to determine whether the optimization goal of minimizing cost is achieved.

[0122] In some embodiments, step S2 can specifically assign a weight coefficient to each objective function, perform a weighted sum based on each objective function and its corresponding weight coefficient, and obtain a single-objective optimization model with the goal of minimizing the sum; wherein the sum of multiple weight coefficients is equal to 1, and the weight coefficients corresponding to different objective functions can be flexibly set according to needs.

[0123] In some embodiments, the weight coefficients corresponding to the harness length objective function and the equipment cost objective function may be the same, and accordingly, step S2 may be specifically as follows:

[0124] Summing the product of the sum of the harness length objective function and the equipment cost objective function and the first weight, and the product of the delay cost objective function and the second weight;

[0125] A single-objective optimization model is obtained with the goal of minimizing the sum; wherein the sum of the first weight and the second weight is 1, and the preset objective function weight is the first weight or the second weight.

[0126] The topology parameters may be directly obtained according to the architecture of the smart device to be configured. In some embodiments, obtaining the topology parameters of the smart device in step S3 may include:

[0127] Based on multiple functional areas of the smart device end and the connection paths between the multiple functional areas, a topological structure diagram of the smart device end is constructed using graph theory methods; the topological structure diagram may include multiple nodes and edges connecting two nodes, the nodes are used to represent the functional areas of the smart device end, and the edges are used to represent the connection paths formed between the two functional areas; the node information of each node and the edge information of each edge in the topological structure diagram are obtained.

[0128] The node information may include a node identification, such as a node ID (Identification), a node name, an activation state, a node position (such as at least one of a node X-axis position coordinate, a node Y-axis position coordinate, and a node Z-axis offset), the number of input interfaces, the number of output interfaces, and at least one of an identification of a controller type that is allowed to be placed. The edge information may include an identification of an edge, such as an edge ID, an identification of a starting node of an edge, an end node representation of an edge, and at least one of an edge length. In other embodiments, the edge information may also include a multi-dimensional weight of a wire bundle corresponding to the edge, such as at least one of gravity, cost, damping, and wire diameter.

[0129] The model configuration parameters can be configured based on demand, and the preset objective function weight can be the value of the first weight determined according to demand. In some embodiments, obtaining the model configuration parameters in step S3 can also include obtaining controller information, wiring harness information, and input and output device information of each functional area on the vehicle side.

[0130] The controller information may include at least one of the controller identification, controller name, controller type, controller input / output capacity, controller power consumption, controller size, and controller unit cost, wherein the controller input / output capacity may be characterized by the maximum number of input interfaces and the maximum number of output interfaces. The wiring harness information may include at least one of unit wiring harness cost, wiring harness type, wiring harness power, and unit delay cost, and the input / output device information may include at least one of the interface type and the number of interfaces of the input / output device.

[0131] In some embodiments, in step S4, the single-objective optimization model is solved for different numbers of controller settings, and the configuration parameters corresponding to the number of controller settings are obtained. Specifically, the single-objective model X2 is solved using the GUROBI algorithm for different numbers of controller settings to obtain the configuration parameters.

[0132] The number of controller settings may correspond one-to-one to configuration parameters, and the configuration parameters may include controller types, controller positions, and controller connectivity paths when constraints and objective functions are satisfied under the current number of controller settings.

[0133] In some embodiments, the target configuration parameters of the controller obtained based on the configuration parameters of the plurality of controller setting quantities in step S4 may be specifically:

[0134] For different controller setting numbers, the total objective function value of the single-objective optimization model is obtained based on the configuration parameters corresponding to the current controller setting number;

[0135] Compare the total objective function values ​​for multiple controller setting quantities;

[0136] According to the comparison results, the number of controller settings and configuration parameters corresponding to the minimum total objective function value are used as the target configuration parameters of the controller; wherein the target configuration parameters include the number of controllers, the type of controller, the location of the controller and the connection path of the controller.

[0137] Among them, the optimal configuration optimization scheme F corresponds to the minimum total objective function value, and the configuration parameters of the controller corresponding to the minimum total objective function value can be used as the target configuration parameters of the controller.

[0138] In other embodiments, in order to improve the effectiveness of the configuration, in step S4, the number of controller settings may be different, and the single objective model X2 may be solved based on the Benders algorithm. For details, please refer to the description below.

[0139] In some embodiments, the controller configuration method provided in the embodiments of the present application can be applied to the vehicle side, and will be described below by taking the application to the vehicle side as an example. The vehicle side may be pre-set with a preset master controller CCC, and the controller types to be configured may include a domain controller Zoner, an electronic controller (ECU, Electronic Control Unit) and an interface center (IO Hub, Input / Output Hub). When configuring the controller to the vehicle side, the CCC, Zoner, ECU or IO hub, and the input and output devices in the functional area, such as actuators or sensors, can be connected in sequence to obtain a tree-like topology. In the following, the controller configuration optimization method provided by the present application will be described by taking the realization of a tree-like topology on the vehicle side as an example.

[0140] Among them, in the first level of this topology, the CCC controller is directly connected to the Zoner, and the Zoners are interconnected through the CCC controller, and there is no connection between the Zoners at the same level; at the second level, the Zoner is connected to the ECU or IO Hub; at the last level, various controllers, namely any of the Zoner, ECU and IOHub, are connected to input and output devices (IO devices) such as actuators or sensors. If the IO devices of all required functions are to be connected, it must be ensured that there is at least one connection path between the device as the input end and the device as the output end. At the second level, the Zoner can be connected to the ECU, and the IO devices of the functional area are connected through the ECU, or the IO hub is used as an extension of the Zoner to connect the IO devices of the functional area. In addition, the Zoner can also be directly connected to the IO device.

[0141] The tree-like topological structure diagram is constructed by graph theory. Accordingly, the architectural design rules that the above topological structure needs to meet may include at least: 1) Each ECU or IOHub can only be connected to one Zoner; 2) Direct physical connection is not allowed between ECUs, and each ECU can only be directly physically connected to Zoner; 3) IOHub can only be used as an extension of Zoner, directly physically connected to Zoner, and is not allowed to be connected through other nodes; 4) Dummy type nodes, that is, the functional area corresponding to the node has no controller deployed, can be directly connected to other non-empty controller type nodes as an extension of other nodes. Unlike the connection using a controller, Dummy type nodes cannot be used as a transfer for other nodes and can only be a terminal node. In addition, it is also necessary to consider the requirements of the setting location for the controller type, quantity, chip computing power and interface. In some embodiments, the computing power of the Zoner chip is divided into three levels: A, B, and C, and each level corresponds to a certain amount of IO capacity and cost. The computing power and cost of all controllers can be ranked as: CCC>Zoner_A>Zoner_B>Zoner_C>ECU>IOHub.

[0142] In some embodiments, decision variables and constraints may be converted into mathematical expressions.

[0143] In some embodiments, the decision variables may include at least one of a controller deployment location variable, a controller connection path variable, a direct connection path variable between the functional area on the vehicle side and the controller variable, a wiring harness connection variable, a computing power resource occupancy variable of the input and output devices corresponding to the functional area, a path selection variable, and a controller deployment type variable.

[0144] Taking the tree-like topology above as an example, the controller deployment location variable can be represented by xi,k It means that when a controller of type k is deployed on node i, x i,k =1; otherwise 0. The controller connection path variable can be used It means that when the controller of location node i is associated with Zoner m, Otherwise, it is 0, and m represents the index of the Zoner type controller. The direct connection path variable between the functional area on the vehicle side and the controller can be used Indicates that when the controller where node j is located is associated with Zonerm, node i is of type Dummy and is directly connected to the controller where node j is located (type is IO Hub, ECU, or Zoner). Otherwise it is 0. Harness connection variables can be used The directed edge from node i to j is represented by e = (i, j). When the position nodes i and j associated with Zoner m are connected by a line harness, Otherwise, it is 0. The computing resource occupancy variables of the input and output devices corresponding to the functional area may include variables occupying Zoner m resources. and variables occupying ECU resources When the input and output of node i requires the computing resources of Zoner m, Otherwise, it is 0; when the input and output of node i need to occupy the computing resources of ECU, Otherwise it is 0. The path selection variable can be used It means that when the controller of node i and the associated Zoner m select path p for connection, Otherwise it is 0. The controller deployment type variable can be used It means that when the controller Zoner m is deployed on node i and the selected type is k∈ZU, Otherwise it is 0, where ZU represents the type set of the controller Zoner, ZU = {Zoner_A, Zoner_B, Zoner_C}.

[0145] In some embodiments, the constraints may include at least one of the controller quantity and type allocation constraints for each functional area on the vehicle side, the communication connectivity rule constraints between different controllers, the connectivity extension relationship constraints for functional areas without controllers, the controller input and output capacity constraints, the chip computing power constraints, and the connectivity rule constraints.

[0146] Among them, taking the above tree-like topology as an example, the number and type allocation constraints of controllers in each functional area of ​​the vehicle can be expressed as:

[0147] Among them, constraint (1) means that each node can only deploy one type of controller; constraint (2) means that each zoner must specify a location and type; constraint (3) means that a zoner of type k is located at node i, and node i is only deployed if and only if a zoner of type k is deployed. V represents the location coordinate set of the node, M represents the zoner set, and there can be multiple zones. The maximum number of zoners, that is, the maximum number of controllers, can be determined by configuration. jk Represents the controller deployment location variable corresponding to a controller of type k deployed at location node j.

[0148]

[0149] CU represents the controller type set, CU = {A-Zoner, B-Zoner, C-Zoner, ECU, IO hub, Dummy}.

[0150] The communication connectivity rules and constraints between different controllers can be expressed as:

[0151] Among them, constraint (4) means that each ECU / IO Hub controller deployed on node i can only be associated with one Zoner; constraint (5) means that if the device deployed on node j is a Zoner itself, it can only be associated with itself, not with other Zoners (that is, if Zoner m is deployed on node j, then Indicates association with oneself).

[0152] The connectivity extension relation constraint of the functional area without controller can be expressed as:

[0153] Among them, constraint (6) indicates that when node i is an extension of j, i and j must be associated with the same Zoner; constraint (7) indicates that node i needs to be extended only when its deployment type is Dummy, that is, when no controller is deployed; constraint (8) indicates that node i is an extension of j only when the controller type of j is non-empty; constraint (9) indicates that each extension can select at most one Zoner. i,Dum Indicates the deployment location variable corresponding to when a dummy type controller is deployed on node i, i.e., no controller is deployed. j,Dum Indicates the deployment location variable corresponding to when a dummy type controller is deployed on node j, that is, when no controller is deployed.

[0154] The controller input and output capacity constraints can be expressed as:

[0155] Constraints (10) and (11) indicate that the input and output devices inside each node must be connected to the deployed controller, and the capacity of the controller must meet the total input and output interface requirements of the node, all extended nodes, and sub-node controllers. i represents the total input interface requirements of the electronic equipment configured at node i, DO i Indicates the total output interface requirements of the electronic devices configured in the node, NI k Indicates the upper limit of the number of input interfaces of type k controller, NO k Indicates the upper limit of the number of output interfaces of type k controller, Δ - (j) = {e∈E:e = (i, j)}, which represents the set of incoming edges pointing to node j. E represents the set of undirected edges connecting all feasible location nodes.

[0156] When k = Dum, that is, when it is a Dummy type controller, the interface capacity is NI k =NO k = 0. When node j is of Dummy type, that is, x j,Dum =1, constraints (10) and (11) can be transformed into: e∈Δ + (j),m∈M, where Δ + (j) = {e∈E:e = (j, i)}, which represents the outgoing edge set of node j. Through the above constraints (10) and (11), it can be ensured that the Dummy type node is an end node and will not become the parent node of any other node, satisfying the setting of the topological network structure of this application.

[0157] The chip computing power constraint can be expressed as:

[0158] Constraints (12)-(13) indicate that for each selected Zoner, its chip computing power should be able to meet the computing requirements of the total number of connected electronic devices after excluding the input and output requirements of the ECU; Constraint (14) indicates that only when node i is associated with Zoner m and the deployed controller type is non-ECU, the node i needs to occupy the computing power of Zoner m; Constraint (15) indicates that when Zoner m is located at node j, the computing power requirements of the node must be met; Constraint (16) indicates that IO Hub type nodes must occupy the computing power of a certain Zoner; Constraint (17) indicates that when node i is an extension of j, if node j does not occupy the computing power of Zoner m, that is, j is an ECU type, then node i will not occupy the computing power of Zoner m, that is, the local ECU has met the computing power requirements of the extended node i; Constraint (18) indicates that when node i is an extension of j, if node j occupies the computing power of Zoner m, then node i will also occupy the computing power of Zoner m. m computing power; constraint (19) means that each node can only occupy the computing power of one Zoner or local ECU.

[0159] The connectivity rule constraint can be expressed as:

[0160] Among them, constraints (20)-(21) indicate that each node i must construct only one connected path p∈P between the node j where the zoner m associated with it is located. ij , this constraint is applicable to the case where safety redundancy constraints are not considered. Constraint (22) indicates that node i can select an empty path only when the controller Zoner m is deployed on node i. Connect Zoner m. Constraint (23) indicates that IO Hub type nodes must establish a first-level connection with Zoner m, and select a direct connection path of p = {i, j}, while other ECUs allow multi-level connections; Constraint (24) indicates that a wire harness connection is constructed between nodes i and j only when there is a selected path p that needs to use this edge (i, j); Constraint (25) indicates that a wire harness e = (i, j) is constructed only when i and j are associated with the same Zoner m, that is, each constructed edge can only belong to one Zoner m; Constraint (26) indicates that physical connection is not allowed between two ECU type nodes; Constraint (27) indicates that each selected extension relationship must construct a wire harness connection accordingly; Constraint (28) indicates that each node can only construct one outgoing edge; Constraint (29) indicates that any edge can only be used by one Zoner at most.

[0161] Constraint (22) ensures that ECU, IO Hub and Dummy type nodes will not choose empty paths. In constraint (28), Δ + (i) = {e∈E:e = (i, j)} represents the set of outgoing edges from node i. This constraint is applicable to the case where safety redundancy constraints are not considered. In other embodiments, this constraint may not be set but added to the target optimization model as a valid inequality.

[0162] In addition, the domains of the above decision variables satisfy:

[0163] Where P represents the set of all connected paths. Setting decision variables and constraints through the above method is conducive to the subsequent systematic optimization of controller configuration.

[0164] In some embodiments, the delay cost objective function may represent the total delay cost of node i connecting to zoner m through path p. The delay cost objective function may be expressed as:

[0165] Among them, delay p represents the connectivity delay cost coefficient, which varies with the configuration location decision.

[0166]

[0167] Considering the constraint condition, only one Dummy node can select a path passing through the ECU node. At this time, this Dummy node belongs to the extension of the ECU. Therefore, by introducing the auxiliary decision variable δ ij The delay cost objective function can be obtained by summing the path delay cost generated by the domain controller through each connection path and node connection and the cost generated by the unit cost difference between the electronic controller and the interface center, which can be specifically expressed as:

[0168] The right side of the plus sign can be a positive number only when i is of Dummy type, j is of ECU type and extends to i.

[0169] Defined as:

[0170] The parameter price represents the unit delay cost and can be parameterized according to the actual situation. ECU Represents the unit cost of ECU, expn Hub Represents the unit cost of the IO Hub.

[0171] The auxiliary decision variable δ ij Satisfy the constraints:

[0172] Each edge e that is physically connected to a harness will increase the harness, and the harness length will increase accordingly. In some embodiments, the harness length objective function is obtained based on the connectivity distance and the harness connection variable, and can be expressed as:

[0173]

[0174] Among them, dist e Represents the connectivity distance of edge e.

[0175] In some embodiments, the equipment cost objective function is obtained based on the controller type, deployment location, unit cost and controller deployment location variables, and can be expressed as:

[0176] ∑ k∈CU ∑ i∈V expn k ·x ik .

[0177] Among them, expn k Represents the unit cost of type k controller.

[0178] In some embodiments, the multi-objective optimization model constructed in step S1 can be expressed as:

[0179] Cost indicators

[0180] Connectivity indicators

[0181] Where ρ represents the normalization coefficient of the unit harness cost. The multi-objective optimization model must meet the above conditions (1)-(34). A cost optimization sub-model with the goal of minimizing cost can be constructed based on the equipment cost objective function and the harness length objective function, and a connectivity optimization sub-model with the goal of maximizing the connectivity of each functional area can be constructed based on the delay cost objective function. The lower the delay cost, the better the connectivity. Therefore, the connectivity optimization sub-model can be constructed by minimizing the delay cost objective function.

[0182] By introducing the normalization coefficient ρ of the unit harness cost, the dimensions can be unified and the units of multiple objective functions can be consistent.

[0183] In some embodiments, the single-objective optimization model obtained in step S2 can be expressed as: Constraints (1)-(34).

[0184] Among them, α in formula (35) represents the first weight, which corresponds to the weight of the cost-related objective function, and 1-α is the second weight, which corresponds to the weight of the connectivity-related objective function.

[0185] In some embodiments, when the single-objective model is solved based on the Benders algorithm, see Figure 2 As shown, in step S4, the single-objective optimization model is solved for different controller setting numbers, and the configuration parameters corresponding to the controller setting numbers can be specifically:

[0186] Step S41: for different controller setting numbers, the single-objective optimization model is divided into a main problem of optimal address allocation of the controller and a sub-problem of the optimal connection path between the controller and the node using the Benders algorithm;

[0187] Step S42: Iteratively solve the main problem and the sub-problems until the preset convergence conditions are met, and obtain the configuration parameters corresponding to the controller setting quantity.

[0188] In some embodiments, step S41 can be specifically divided into a main problem of optimal controller location allocation and a sub-problem of optimal connection path between controller and node using Benders algorithm, wherein the main problem MP solves the location allocation decision (x, y, z, u, δ, γ), and the sub-problem SP determines the connection decision (w, π) under the pre-fixed location allocation decision. The location allocation decision is related to the controller deployment location variable x, controller connection path variable y, direct connection path variable z, computing power resource occupancy variable u, and controller deployment type variable γ in the decision variables, as well as the auxiliary decision variable δ. The connectivity decision is related to the wiring harness connection variable w and the path selection variable π in the decision variables.

[0189] In some embodiments, step S42 may specifically include:

[0190] Set the convergence accuracy of the algorithm ∈ = 1e-3, initialize an empty set Ω, which is used to store Benders cuts, and set an infinite upper bound UB and an infinite lower bound LB;

[0191] When Ω is an empty set, no Benders cut is considered. First, solve the main problem MP(Ω) to obtain the initial location allocation decision.

[0192] Enter the iteration until any of the following convergence conditions is met: (UB-LB) / LB≥∈, and the current running time exceeds the preset time limit.

[0193] In each iteration, the following steps are performed:

[0194] Based on the current main problem solution Solving subproblems To determine the path connection decision

[0195] Solve the linear relaxed dual problem of the subproblem, generate a Benders optimal cut based on the dual solution, and add it to the set Ω; the Benders optimal cut is used to restrict the solution space in the subsequent solution of the main problem to help converge to the optimal solution faster;

[0196] Calculate the total cost of the current solution, that is, the total objective function value of the objective function f(x, y, z, u, δ, γ) of the main problem plus the objective function g(w, π) of the subproblem. If the total cost is less than the current upper bound UB, update the upper bound UB.

[0197] Solve the updated master problem: Use the updated Benders cut set Ω to re-solve the master problem MP(Ω) and obtain a new location allocation decision. and possible additional variables

[0198] According to the new solution of the main problem, the lower bound LB is calculated and updated; the function value of the objective function f(x, y, z, u, δ, γ) of the main problem and the additional variables The sum is compared with the lower bound LB. When the function value of the objective function f(x, y, z, u, δ, γ) of the main problem and the additional variable When the sum is greater than the lower bound LB, the lower bound LB is updated.

[0199] When the algorithm meets any convergence condition, the iteration ends. At this time, the location allocation decision and path connectivity decisions That is the approximate optimal solution to the problem, which can be used to determine the target configuration parameters of the controller.

[0200] Among them, the main problem can be obtained by the product of the first weight and the equipment cost objective function, and the product of the second weight and the delay cost generated by the unit cost difference between the electronic controller and the interface center. The objective function of the main problem can be expressed as:

[0201]

[0202] The sub-problem can be obtained by summing the product of the first weight and the objective function of the harness length, and the product of the second weight and the path delay cost generated by the domain controller through each connection path and node connection. The objective function of the sub-problem can be expressed as:

[0203] Further, sub-question It can be written as follows:

[0204] st constraints (20)-(29);

[0205] Next, solve the subproblem The linear relaxation dual problem of are relaxed to continuous variables. dM i ,dN e denote the dual variables corresponding to the constraints (20)-(29) in the subproblem respectively. The dual problem of the linear relaxation of the subproblem can be constructed as follows:

[0206] The indicator matrix is ​​defined as:

[0207] Introducing a new variable ξ, during the ωth iteration, the following benders optimal cut can be constructed from the solution of the dual subproblem:

[0208] During the entire Benders algorithm iteration process, the Benders optimal cut generated in each iteration is added to the set Ω.

[0209] The master problem MP(Ω) is a mixed integer programming problem, which can be solved directly using a solver such as (GUROBI algorithm or CPLEX algorithm, etc.). Based on the objective function of the master problem, the master problem MP(Ω) can be expressed as: st Constraints (1)-(19), (32)-(33); ξ∈R.

[0210] In some embodiments, after obtaining the target configuration parameters of the controller, the method further includes: providing a deployment topology diagram of the controller, which may include the number of controllers, controller types, controller locations, and controller connection paths.

[0211] In some specific embodiments, see Figure 3 As shown, Figure 3 This is a schematic diagram of the architecture of the vehicle to be configured provided by the specific embodiment of the present application, which may include 35 functional areas divided based on the types of functions to be implemented. It should be noted that Figure 3 When the controller configuration optimization method provided in the embodiment of the present application is applied to configure the controller on the vehicle side, the graph theory method can be used to obtain the initial topology structure diagram corresponding to the vehicle side to be configured, and the topology structure parameters of the vehicle side can be obtained based on the initial topology structure diagram to obtain the node information and edge information corresponding to each functional area, such as Figure 4 As shown, Figure 4 The embodiment of this application provides Figure 3 The node information table corresponding to the multiple functional areas of the vehicle to be configured may include the node ID, node name, number of input interfaces, number of output interfaces and the controller type ID allowed to be placed, such as Figure 5 As shown, Figure 5 The embodiment of this application provides Figure 3 The edge information table corresponding to multiple functional areas on the vehicle to be configured in the middle. The edge information may include the starting node ID of the edge, the end node ID of the edge, the length of the edge and the optional weight of the edge. When the optional weight of the edge is 1, it means that there is a connected path between the current starting node and the end node. When the optional weight of the edge is 0, it means that there is no connected path between the current starting node and the end node. In addition, you can also obtain model configuration parameters, such as the maximum number of controller settings, preset objective function weights, controller information, wiring harness information, and input and output device information. See Figure 6 As shown, Figure 6 The embodiment of the present application provides Figure 3 The controller information table of the vehicle-side configuration to be configured may include controller ID, controller name, controller type, maximum number of input interfaces, maximum number of output interfaces and unit cost.

[0212] The obtained topological structure parameters and model configuration parameters can be input into the single-objective optimization model obtained based on the above expression (35) and constraints (1)-(34). The preset objective function weight can be configured based on actual needs. When minimizing the connectivity cost is given priority, the preset objective function weight can be configured to be less than 0.5, so that the first weight in the single-objective optimization model is less than the second weight, which is conducive to obtaining a controller configuration that is more inclined to a topological structure with direct edges and improves connectivity. See Figure 8 When minimizing the cost is given priority, the preset objective function weight can be configured to be greater than 0.5, so that the first weight in the single-objective optimization model is greater than the second weight, which is beneficial to reducing equipment costs, but it is easy to obtain a more complex multi-level network, see Fig.10As shown. For different controller settings, the single-objective model is solved respectively, and the configuration optimization scheme of the controller is obtained based on multiple solution results, and the target configuration parameters of the controller can be determined. Among them, the single-objective model can be directly solved using the GUROBI algorithm, or the single-objective model can be solved using the method based on the Benders algorithm provided in any of the above embodiments of the present application. For different controller settings, the controller position, controller connection path and the number of deployments of each type of controller under the corresponding controller setting number can be obtained after solving. In addition, the deployment cost of each type of controller, the length of the wiring harness connection and the objective function value of the overall optimization target corresponding to the single-objective optimization model, that is, the total objective function value, can be further obtained. By comparing the total objective function values ​​corresponding to different controller settings, the final configuration optimization scheme of the controller can be determined. Among them, the number of controllers, controller types, controller positions and controller connection paths corresponding to the minimum total objective function value can be used as the target configuration parameters of the controller.

[0213] See also Figure 7 As shown, Figure 7 The specific embodiment of the present application provides that when the preset function target weight is 0.1, Figure 3 The cost distribution diagram corresponding to different numbers of controllers configured on the vehicle side. Zone represents the total number of deployed controllers. Taking the second column as an example, it shows the Zoner deployment cost, wiring harness connection cost, connectivity cost and total objective function value when two controllers are deployed: 1 Zoner_B and one Zoner_C. In other cases, when the configured controller types include ECU or IO Hub, the deployment cost of the corresponding controller can also be shown. By comparing the total objective function values ​​corresponding to different numbers of controller configurations, it can be determined that the total objective function value corresponding to the second column is the smallest, which can be used as the final controller configuration optimization solution. Accordingly, the corresponding vehicle-side topology structure diagram for 2 Zoner controllers can be obtained, see Figure 8 As shown, the position of the CCC controller can be predetermined according to the vehicle-side architecture, and the numbers 1 to 35 are respectively Figure 4 The node ID in Figure 8 The controller type corresponding to each node ID is also shown in the figure, and the connection path between each node and the controller is shown by a dotted line, and the connection path between the CCC and the controller is shown by a solid line. For example, the reference numeral 14 corresponds to the right rear door area, and the controller type set in this area is Dummy, that is, no controller is set in this area, and this area is connected to the node 20, that is, it is controlled by the ZonerB controller set at the node 20.

[0214] In addition, you can also set the preset function target weight to 0.9, see Fig. 9 As shown, Fig. 9Another specific embodiment of the present application provides that when the preset function target weight is 0.9, Figure 3 The cost distribution diagram corresponding to different numbers of controllers configured on the vehicle side. Taking the third column as an example, it shows the Zoner deployment cost, IOhub deployment cost, wiring harness connection cost, connectivity cost and total objective function value when 3 Zoner_Cs and 5 IO hubs are deployed. In other cases, when the configured controller type includes ECU, the deployment cost of the corresponding controller can also be shown. By comparing the objective function values ​​of the overall optimization target corresponding to different controller configuration numbers, it can be determined that the total objective function value corresponding to the third column is the smallest, which can be used as the final controller configuration optimization solution. Accordingly, the corresponding vehicle-side topology structure diagram when 3 Zoner_Cs and 5 IO hubs are deployed can be obtained, see Fig.10 As shown, the position of the CCC controller can be predetermined according to the vehicle-side architecture, and the numbers 1 to 35 are respectively Figure 4 The node ID in Fig.10 The controller type corresponding to each node ID is also shown in the figure, and the connection path between each node and the controller is shown by a dotted line, and the connection path between the CCC and the controller is shown by a solid line. For example, the number 14 corresponds to the right rear door area, and the controller type set in this area is IO hub. This area is connected to nodes 31, 35 and 20 at the same time. The controller types of nodes 31 and 35 are both Dummy, which can be connected to node 14 as an extension point. Nodes 31, 35 and 14 are all controlled by the Zoner_C controller set at node 20.

[0215] It is understood by those skilled in the art that all or part of the processes in the method for implementing the above-mentioned embodiment of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.

[0216] Another aspect of the present application also provides a computer-readable storage medium.

[0217] In an embodiment of a computer-readable storage medium according to the present application, the computer-readable storage medium may be configured to store a program for executing the controller configuration optimization method of the above-mentioned method embodiment, and the program may be loaded and run by a processor to implement the above-mentioned controller configuration optimization method. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present application. The computer-readable storage medium may be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-temporary computer-readable storage medium.

[0218] Another aspect of the present application also provides a smart device.

[0219] In an embodiment of an intelligent device according to the present application, the intelligent device may include at least one processor; and a memory connected to the at least one processor in communication; wherein a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the method described in any of the above embodiments is implemented. The intelligent device described in the present application may include a driving device, a smart car, a robot, and other devices. Fig.11 , Fig.11 FIG. 4 exemplarily shows that the memory 11 and the processor 12 are communicatively connected via a bus.

[0220] In some embodiments of the present application, the smart device may further include at least one sensor, which is used to sense information. The sensor is connected to any type of processor mentioned in the present application. Optionally, the smart device may also include an autonomous driving system, which is used to guide the smart device to drive itself or assist in driving. The processor communicates with the sensor and / or the autonomous driving system to complete the method described in any of the above embodiments. (The electronic device described in the present application may be, but is not limited to, a mobile phone, a tablet computer, a desktop, a laptop, a handheld computer, a notebook computer, a vehicle-mounted device, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), an augmented reality (AR)\virtual reality (VR) device, etc., and the embodiments of the present application are not limited to this.)

[0221] So far, the technical solution of the present application has been described in conjunction with an embodiment shown in the accompanying drawings, but it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present application.

Claims

1. A controller configuration optimization method, characterized in that: include: Based on decision variables, constraints and objective functions, a multi-objective optimization model for configuring a controller on the smart device side is constructed; wherein the objective function includes a delay cost objective function, a harness length objective function and an equipment cost objective function; Based on the delay cost objective function, the harness length objective function, and the equipment cost objective function, converting the multi-objective optimization model into a single-objective optimization model by weighting; Obtaining topological structure parameters and model configuration parameters of the smart device end, and inputting them into the single-objective optimization model, wherein the model configuration parameters at least include a maximum number of controller settings and a preset objective function weight; For different controller setting numbers, the single-objective optimization model is solved respectively to obtain configuration parameters corresponding to the controller setting numbers; the target configuration parameters of the controller are obtained based on multiple configuration parameters of the controller setting numbers; the controller setting number is less than or equal to the maximum controller setting number.

2. The controller configuration optimization method according to claim 1, characterized in that: The smart device terminal includes a plurality of functional areas, and the functional areas are provided with input and output devices to be connected with the controller; The decision variables include at least one of a controller deployment position variable, a controller connection path variable, a direct connection path variable between the functional area of ​​the smart device and the controller, a wiring harness connection variable, a computing resource occupancy variable of the input and output devices corresponding to the functional area, a path selection variable, and a controller deployment type variable; The constraint conditions include at least one of the controller quantity and type allocation constraints of each functional area of ​​the smart device, the communication connectivity rule constraints between different controllers, the connectivity extension relationship constraints of the functional area without controller, the controller input and output capacity constraints, the chip computing power constraints and the connectivity rule constraints.

3. The controller configuration optimization method according to claim 2, characterized in that: The model configuration parameters also include controller types, which include domain controllers, electronic controllers, and interface centers.

4. The controller configuration optimization method according to claim 3, characterized in that: The method further comprises constructing a topological structure diagram of the smart device end by using a graph theory method, wherein the topological structure diagram comprises a plurality of nodes and an edge connecting two nodes; the nodes are used to represent the functional areas of the smart device end, and the edges are used to represent the connection path formed between the two functional areas; The delay cost objective function is obtained by summing the path delay cost generated by the domain controller through each connection path and node connection and the cost generated by the unit cost difference between the electronic controller and the interface center. The delay cost objective function is expressed as: in, And δ ij The following constraints are met: V represents the location coordinate set of the node, V zoner represents the location coordinate set corresponding to the domain controller, p represents the connected path, P ij represents the set of connected paths between nodes i and j, m represents the index of the domain controller, M represents the set of domain controllers, delay p represents the connectivity delay cost coefficient, The path selection variable representing the connection between the controller of node i and the domain controller by path p, expn ECU represents the unit cost of the electronic controller, expn Hub represents the unit cost of the interface center, price represents the unit delay cost, and δ ij represents the auxiliary decision variable, represents the indicator matrix of path p passing through edge e, E represents the set of undirected edges connecting all feasible location nodes, x i,Dum The controller deployment location variable representing the empty controller deployed on node i, x j,ECU The controller deployment location variable representing the electronic controller deployed at node i, Represents the direct connection path variable between the functional area and the controller.

5. The controller configuration optimization method according to claim 3, characterized in that: The harness length objective function is obtained based on the connectivity distance and the harness connection variable, and the harness length objective function is expressed as: Where e represents the edge index, E represents the set of undirected edges connecting all feasible location nodes, m represents the index of the domain controller, M represents the set of domain controllers, and dist e represents the connectivity distance of edge e, Represents a harness connection variable.

6. The controller configuration optimization method according to claim 3, characterized in that: The equipment cost objective function is obtained based on the controller type, deployment location, unit cost and controller deployment location variables, and the equipment cost objective function is expressed as: Among them, CU represents the controller type set, V represents the node position coordinate set, expn k represents the unit cost of type k controller, x ik A controller deployment location variable representing the deployment of a type k controller on node i.

7. The controller configuration optimization method according to claim 1, characterized in that: Acquiring the topological structure parameters of the smart device includes: Based on the multiple functional areas of the smart device and the connection paths between the multiple functional areas, a topological structure diagram of the smart device is constructed by using a graph theory method; the topological structure diagram includes multiple nodes and edges connecting two nodes, the nodes are used to represent the functional areas of the smart device, and the edges are used to represent the connection paths formed between the two functional areas; Obtain node information of each node and edge information of each edge in the topological structure diagram; wherein the node information includes at least one of a node identifier, a node name, an activation state, a node position, a number of input interfaces, a number of output interfaces, and an identifier of a controller type that is allowed to be placed; and the edge information includes at least one of an edge identifier, an edge start node identifier, an edge end node identifier, and an edge length.

8. The controller configuration optimization method according to claim 1, characterized in that: Obtaining model configuration parameters also includes: Acquire controller information, harness information, and input / output device information of each functional area at the smart device end; wherein the controller information includes at least one of a controller identification, a controller name, a controller type, a controller input / output capacity, a controller energy consumption, a controller size, and a unit cost of the controller; the harness information includes at least one of a unit harness cost, a harness type, a harness power, and a unit delay cost; and the input / output device information includes at least one of an interface type and a number of interfaces of the input / output device.

9. The controller configuration optimization method according to any one of claims 1 to 8, characterized in that: The converting the multi-objective optimization model into a single-objective optimization model by weighting based on the delay cost objective function, the harness length objective function, and the equipment cost objective function comprises: Summing the product of the sum of the harness length objective function and the equipment cost objective function and a first weight, and the product of the delay cost objective function and a second weight; The single-objective optimization model is obtained with the goal of minimizing the sum; wherein the sum of the first weight and the second weight is 1, and the preset objective function weight is the first weight or the second weight.

10. The controller configuration optimization method according to any one of claims 1 to 8, characterized in that: For different controller setting numbers, the single objective optimization model is solved respectively, and the configuration parameters corresponding to the controller setting numbers are obtained, including: According to different controller settings, the single-objective optimization model is divided into the main problem of optimal address allocation of controllers and the sub-problem of optimal connection paths between controllers and nodes by using Benders algorithm. The main problem and the sub-problems are iteratively solved until a preset convergence condition is met, and configuration parameters corresponding to the controller setting quantity are obtained.

Images