Designing automotive embedded systems with adaptive genetic algorithms

One of the most common problems faced by planners, whether in industry or government, is optimisation—finding the optimal solution to a problem. Even a one percent improvement in a solution can make a difference of millions of dollars in some cases. Traditionally optimisation problems are solved by analytic means or exact optimisation methods. Today, however, many optimisation problems in the design of embedded architectures involve complex combinatorial systems that make such traditional approaches unsuitable or intractable. Genetic algorithms, instead, tackle these kind of problems by finding good solutions in a reasonable amount of time. Their successful application, however, relies on algorithm parameters which are problem dependent, and usually even depend on the problem instance at hand. To address this issue, we propose an adaptive parameter control method for genetic algorithms, which adjusts parameters during the optimisation process. The central aim of this work is to assist practitioners in solving complex combinatorial optimisation problems by adapting the optimisation strategy to the problem being solved. We present a case study from the automotive industry, which shows the efficiency and applicability of the proposed adaptive optimisation approach. The experimental evaluation indicates that the proposed approach outperforms optimisation methods with pre-tuned parameter values and three prominent adaptive parameter control techniques.