As the design of electrical/electronic (E/E)-architectures is becoming more complex, multi-objective optimization algorithms such as evolutionary algorithms (EAs) have been proposed for generating resource optimized architectures. In this paper we extend existing approaches by excluding infeasible solutions from the search space and thereby enhance the quality and runtime behavior of the optimization.
Due to the introduction of new safety and comfort functions related to advanced driver assistance, highly automated driving, and car-to-x connectivity, the number and interconnections of functions in vehicles has been growing over the last decades and is going to grow further over the next years. As a result, distributed vehicle system architectures consisting of heterogeneous computing resources become more complex and power consuming. Thereby, also the complexity of the allocation task problem is growing, which is de ned as assigning functions to Electronic Control Units (ECUs) while ful lling various design constraints, such as safety requirements and resource restrictions (see Figure 1).
In recent years, several optimization methods with the aim of supporting
engineers in power and resource aware design of E/E-architectures have been
proposed. While Walla [
Models and Representation
In our approach, we optimize three objective functions: network communication,
safety, and the collection of functions with certain properties on a minimum
number of ECUs. As the search space reduction is independent of the objective
functions, we will concentrate on the constraints in the following:
{ Location constraints either de ne a set of ECUs CI;i - where function fi will
be mapped to one element of this set - or exclude a set of ECUs CE;i.
{ Colocation constraints either forbid two functions to be allocated to the same
ECU (Cban : e(fi) 6= e(fj )) or force two functions to reside on the same ECU
(Cforce : e(fi) = e(fj )).
{ Some functions demand certain components or functional requirements reqf,
which have to be provided by the corresponding ECU (reqe). Therefore,
function fi can only be allocated to one ECU of the set Creq;i = fejreqe
reqf g.
{ Some resources, such as the memory and computing units, are shared among
all functions allocated to the ECU. Therefore, re;j P rf;i with (e(fi) = ej )
has to hold for each ECU.
For the optimization, we use the NSGA-II algorithm presented by Deb [
New solutions are generated by variation operators such as mutation, which assigns an element of set M to a random number of functions. Thereby, a huge amount of solutions violating the constraints de ned in chapter 2.2 might be generated that would have to be either repaired or discarded. As an approach to reduce this number, we reduce the sets Mi for each function fi in advance of the optimization by analyzing the design constraints and use those sets during initialization and mutation to generate new individuals. 4
5
Furthermore, let CI;2 = fe1; e2; e3g, CE;5 = fe1g, and consider f orce(f5; f6) and ban(f1; f2). Applying SSR to the optimization problem results in M1 = fe1g, M2 = fe2; e3g, M3 = fe1; e2; e3g, M4 = fe1; e2; e3g, M5 = fe2; e3; e4g and M6 = fe2; e3; e4g. As shown in Table 3, the search space could be pruned to about 1 % of its original size. This has also an impact on the number of generations for nding the optimal solutions. With a population size of 10, we commonly nd the ve Pareto-points after 15 generations, whereas an EA with repair mechanism needs 150 generation and an EA without repair function and mapping sets more than 5000 generations. For a larger use case with 32 functions, 10 ECUs, location constraints on 28 functions, and a population size of 50, Figure 5 shows the mean hypervolume values over 25 optimization runs. Whereas the standalone NSGA-II nds rst feasible solutions after 60 generations and then slowly increases, the NSGA-II with SSR converges after 140 generations.