Active Exploration based on Information Gain by Particle Filter for Efficient Spatial Concept Formation

Abstract

Autonomous robots need to learn the categories of various places by exploring their environments and interacting with users. However, preparing training datasets with linguistic instructions from users is time-consuming and labor-intensive. Moreover, effective exploration is essential for appropriate concept formation and rapid environmental coverage. To address this issue, we propose an active inference method, referred to as spatial concept formation with information gain-based active exploration (SpCoAE) that combines sequential Bayesian inference using particle filters and information gain-based destination determination in a probabilistic generative model. This study interprets the robot's action as a selection of destinations to ask the user, `What kind of place is this?' in the context of active inference. This study provides insights into the technical aspects of the proposed method, including active perception and exploration by the robot, and how the method can enable mobile robots to learn spatial concepts through active exploration. Our experiment demonstrated the effectiveness of the SpCoAE in efficiently determining a destination for learning appropriate spatial concepts in home environments.

arXiv commentAccepted to Advanced Robotics

References in corpus (13)

• Volumetric Instance-Aware Semantic Mapping and 3D Object Discovery
• Rao-Blackwellised Particle Filtering for Dynamic Bayesian Networks
• Semantics for Robotic Mapping, Perception and Interaction: A Survey
• Active inference body perception and action for humanoid robots
• Online Spatial Concept and Lexical Acquisition with Simultaneous Localization and Mapping
• Improved and Scalable Online Learning of Spatial Concepts and Language Models with Mapping
• Fast Autonomous Robotic Exploration Using the Underlying Graph Structure
• Autonomous Planning Based on Spatial Concepts to Tidy Up Home Environments with Service Robots
• Spatial Concept-Based Navigation with Human Speech Instructions via Probabilistic Inference on Bayesian Generative Model
• Multimodal Hierarchical Dirichlet Process-based Active Perception
• SEAL: Self-supervised Embodied Active Learning using Exploration and 3D Consistency
• Ready Policy One: World Building Through Active Learning
• Learning to Map for Active Semantic Goal Navigation

Cited by in corpus (1)

• Hierarchical Path-planning from Speech Instructions with Spatial Concept-based Topometric Semantic Mapping