A frontier-based approach for autonomous exploration

By Brian Yamauchi. 1997

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A Survey of Embodied AI: From Simulators to Research Tasks (2021, methodology): “For frontier-exploration, the agent visits the edges between free and unexplored spaces iteratively using a map [62, 47].”
Cognitive Mapping and Planning for Visual Navigation (2017, methodology): “geometric exploration along with a semantic segmentation network trained2 to identify object categories of interest. The agent systematically explores the environment using frontier-based exploration [77]3 till it detects the speciﬁed category using the semantic segmentation network. These labels ar”
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016, positive): “In practice, a small subset of locations in the map is selected, using techniques such as frontier-based exploration [127], [262].”
Coverage, Exploration and Deployment by a Mobile Robot and Communication Network (2003, methodology): “Single robot exploration of unknown environments has been studied before [6{8].”
ZSON: Zero-Shot Object-Goal Navigation using Multimodal Goal Embeddings (2022, background): “– CLIP on Wheels (CoW) [21]: builds an occupancy map by projecting depth observations, then searches the environment with frontier-based exploration [36].”
Coverage and connectivity issues in wireless sensor networks: A survey (2008, background): “Conceptually, it is similar to the frontier-based approach [68]; however, in this case, occupancy maps are built from live sensory data and are analyzed to find frontiers between the free space and the unknown space.”
The ThreeDWorld Transport Challenge: A Visually Guided Task-and-Motion Planning Benchmark Towards Physically Realistic Embodied AI (2021, unknown): “• Frontier Exploration: Similar to [49], the agent uses depth images and semantic segmentation masks to construct an occupancy map and a semantic map of the scene.”
Aligning Cyber Space With Physical World: A Comprehensive Survey on Embodied AI (2024, unknown): “The A-EQA task extended EM-EQA meth-ods with frontier-based exploration (FBE) [151] for problem-independent environment exploration.”
NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration (2023, methodology): “Some prior exploration methods use local strategies for generating control actions for the robots [4–7], while others use global strategies based on the frontier method [8–10].”
Vision-based autonomous mapping and exploration using a quadrotor MAV (2012, methodology): “We use the frontier-based exploration [25] algorithm as our exploration strategy, and we set a fixed altitude for the MAV to fly at during the exploration process.”
CLIP on Wheels: Zero-Shot Object Navigation as Object Localization and Exploration (2022, methodology): “Frontier based exploration (FBE) [73].”
Learning Exploration Policies for Navigation (2019, methodology): “Current approaches either use a human operator to control the robot for building the map ( e.g. Thrun et al. (1999)), or use heuristics such as frontier-based exploration (Yamauchi, 1997).”
Bayesian Learning for Safe High-Speed Navigation in Unknown Environments (2015, background): “In the exploration literature, the primary objective is to build a map, and actions may be taken to view unseen parts of the environment [30, 28].”
Neural Topological SLAM for Visual Navigation (2020, methodology): “Metric Spatial Map + FBE + Local.”
An Exploration of Embodied Visual Exploration (2020, unknown): “…[42, 35] samples from a uniform distribution over all actions, (2) forward-action [35] always samples the forward action, (3) forward-action+ samples the forward action unless a collision occurs, in which case, it turns left, and (4) frontier-exploration [56] uses the egocentric map from Fig.”
Receding Horizon "Next-Best-View" Planner for 3D Exploration (2016, background): “In [17] the notion of frontier regions was introduced for the ﬁrst time.”
Deep Reinforcement Learning Robot for Search and Rescue Applications: Exploration in Unknown Cluttered Environments (2019, methodology): “In this method, a robot is directed to explore the boundaries of an already-explored area, i.e., the frontiers [10].”
BeBold: Exploration Beyond the Boundary of Explored Regions (2020, unknown): “Frontier-based exploration (Yamauchi, 1997; 1998; Topiwala et al., 2018) is used to help speciﬁc robots explore the map by maximizing the information gain.”