Bayesian Models for Science-Driven Robotic Exploration

Abstract

Planetary rovers allow for science investigations in remote environments. They have traversed many kilometers and made major scientific discoveries. However, rovers spend a considerable amount of time awaiting instructions from mission control. The reason is that they are designed for highly supervised data collection, not for autonomous exploration. The exploration of farther worlds will face increasing challenges and constraints. Such missions will demand a new approach.

This work advocates Bayesian models as powerful tools for a new paradigm for robotic explorers. In this approach, the explorer's description of where to go is not prescribed by a fixed set of instructions, but instead by a model of what the explorer believes. This formulation has several benefits. Bayesian models provide a mathematically grounded framework to reason about uncertainty. They can allow robots to gain a deeper understanding of the evolving scientific goals guiding the mission. Furthermore, they can empower scientists by providing explainable results.

To this end, this research develops models that allow for data interpretation by learning and exploiting structure in the data and the environment. It shows how these models enable robotic explorers to make intelligent decisions based on instantaneous information. Ultimately, it demonstrates how science productivity is improved by measuring science value with information-theoretic variables and by formulating the exploration problem in terms of Bayesian experimental design.

This work makes several contributions to the field of science-driven robotic exploration. First, it introduces three different deep generative models for the analysis of data that enable scientists to quantify and interpret learned statistical dependencies. Then, it presents an adaptive exploration model that leverages contextual information from remote data to efficiently extrapolate features from in situ observations, as well as a corresponding strategy for improving science productivity. Afterward, it establishes a hierarchical probabilistic structure in which scientists initially describe their abstract beliefs and hypotheses, and then this belief evolves as the robot makes raw measurements; additionally, science information gain is efficiently computed and maximized. Finally, it proposes a comprehensive model for planetary rover exploration that considers both science productivity and risk.

The presented Bayesian models are validated and evaluated in various science investigation scenarios that can provably benefit from autonomous robotic exploration. Such scenarios include terrestrial, airless, and Martian surface surveys, as well as marine biology studies. Emphasis is placed on spectroscopic data, which is widely used in the natural sciences for composition analysis. Promising results are shown in simulations and field experiments using the autonomous rover Zoe.