Performing Active Search to Locate Indication of Ancient Water on Mars: An Online, Probabilistic Approach

Unsupervised Active Visual Search with Monte Carlo planning under Uncertain Detections

“…A MONG the most interesting areas of robotics is the problem of Active Visual Search (AVS) [1], in which an intelligent robotic agent must autonomously find an object located far from it, moving and exploring its surroundings through egocentric visual sensors. AVS applies in many different contexts, such as the domotic field [2], [3], personal assistance [4], search and rescue [5], [6], and the very intriguing Mars exploration [7]. In this paper we focus on the AVS problem in an indoor environment [8], where the only available knowledge is its 2D map.…”

Similarity and Consistency in Algorithm-Guided Exploration

“…Algorithm-based decision support systems are becoming increasingly influential in the decisionmaking practices of individuals and organizations, including product searches, portfolio choices, and human resource procurement (e.g., Adomavicius and Tuzhilin, 2005b;Panniello et al, 2016;Tauchert and Mesbah, 2019;Zhou et al, 2021;Wang et al, 2022;Uotila et al, 2009). They also affect search and rescue missions and production settings, as well as geological and space exploration, in which teams of humans and algorithm-driven robots jointly perform search tasks (e.g., Orth et al, 2021;Schoonderwoerd et al, 2022;Pouya and Madni, 2021). Such decision problems are typically characterized by sequential choices and learning over time.…”