GODEL(Generative Optimal Design Engine Language) is a hardware design optimization toolkit.
The general idea behind the tooling is that hardware design is a series of tradeoffs. By creating a tool which indicates those tradeoffs to the designer in a simple way, we can create a hardware design toolkit for the consumer to decide which tradeoffs they want to make.
The concept is applied generally, but in this work we implemented a component selection tool for the heating module in a Personal Food Computer.
The core of the system is a Multi-Objective Optimization algorithm which explores the pareto-optimal front of a hardware design model. A Pareto Front is a line in a multi-dimensional hyperplane which defines the most optimal points for a specific problem. When the user selects their desired tradeoffs, the algorithm selects the nearest neighbor in the pareto front.