Advancements in Autonomous Robot Exploration and Mapping

The field of autonomous robot exploration and mapping is rapidly advancing, with a focus on improving the efficiency and accuracy of exploration in complex environments. Recent developments have seen the integration of hierarchical representations, attention-based deep reinforcement learning, and novel reward mechanisms to enable more effective exploration. Additionally, there is a growing interest in multi-robot systems, with research focusing on task coordination, trajectory execution, and active target discovery. These advancements have the potential to significantly impact various applications, including robotics, logistics, and environmental monitoring. Noteworthy papers include: HEADER, which presents an attention-based reinforcement learning approach for efficient exploration in large-scale environments. SEA, which proposes a novel approach for active robot exploration through semantic map prediction and a reinforcement learning-based hierarchical exploration policy. IMAS$^2$, which introduces a two-layer optimization structure for joint agent selection and information-theoretic coordinated perception in Dec-POMDPs.

Sources

Traversability-aware Consistent Situational Graphs for Indoor Localization and Mapping

HEADER: Hierarchical Robot Exploration via Attention-Based Deep Reinforcement Learning with Expert-Guided Reward

Few-Shot Demonstration-Driven Task Coordination and Trajectory Execution for Multi-Robot Systems

Zero-Shot Coordination in Ad Hoc Teams with Generalized Policy Improvement and Difference Rewards

Heterogeneous Multi-Agent Task-Assignment with Uncertain Execution Times and Preferences

Active Target Discovery under Uninformative Prior: The Power of Permanent and Transient Memory

DiRAC - Distributed Robot Awareness and Consensus

Active Inference for an Intelligent Agent in Autonomous Reconnaissance Missions

SEA: Semantic Map Prediction for Active Exploration of Uncertain Areas

IMAS$^2$: Joint Agent Selection and Information-Theoretic Coordinated Perception In Dec-POMDPs

Built with on top of