Integrated Data-Driven and Model-Based Trade-off Analysis of Sensor Suite System for Autonomous Ground Vehicle Navigation

In recent years, the deployment of robots for commercial applications has surged, emphasizing the need to incorporate autonomy into these systems. The widespread adoption of autonomous robots across diverse fields thus necessitates a rigorous design framework. To address this, we recently proposed an Integrated Data-Driven and Model-Based Systems Engineering (IDDMBSE) framework—an extension of our traditional Model-Based Systems Engineering (MBSE). Our IDDMBSE framework aims to enable trusted autonomy in autonomous systems. In this work, we outline a methodology for trade-off analysis within the IDDMBSE framework, supporting the development of autonomous robots based on functional, performance, and programmatic requirements. The method comprises two key components: 1) Model-Based Optimization (MBO) - Leveraging domain-specific knowledge, MBO facilitates efficient simulation-based design, and 2) Data-Driven Evaluation (DDE) - Decoupling DDE from MBO allows effective exploration and statistical analysis of the design space. We demonstrate the utility of the proposed framework by using it to perform design space exploration and trade-off analysis for the design of the sensor suite of an autonomous ground robot. Our results show how a classical design problem that is often deemed intractable using model-based methods and unscalable using data-driven methods can benefit through the use of our hybrid framework.