Autonomous Exploration
Autonomous exploration focuses on enabling robots to independently map and navigate unknown environments, optimizing efficiency and completeness of coverage. Current research emphasizes efficient mapping techniques using various sensor modalities (LiDAR, RGB-D cameras), incorporating semantic information for improved understanding, and leveraging advanced algorithms like deep reinforcement learning, active inference, and graph-based methods for planning optimal exploration trajectories. This field is crucial for advancing robotics in diverse applications, including search and rescue, planetary exploration, and industrial inspection, by enabling robots to operate effectively in unstructured and dynamic settings.
Papers
Active Inference in Contextual Multi-Armed Bandits for Autonomous Robotic Exploration
Shohei Wakayama, Alberto Candela, Paul Hayne, Nisar Ahmed
HDPlanner: Advancing Autonomous Deployments in Unknown Environments through Hierarchical Decision Networks
Jingsong Liang, Yuhong Cao, Yixiao Ma, Hanqi Zhao, Guillaume Sartoretti