CLEAR: A Semantic-Geometric Terrain Abstraction for Large-Scale Unstructured Environments
🤖 AI Summary
Existing terrain abstraction methods struggle to simultaneously preserve semantic meaning, geometric structure, and computational efficiency in unstructured environments spanning tens of square kilometers, leading to unreliable long-range path planning. This work proposes CLEAR, a novel approach that uniquely integrates land-cover semantics with geometric structure through boundary-aware spatial decomposition and recursive planar fitting to generate semantically aligned convex regions. These regions form a terrain-aware graph that enables efficient navigation. Evaluated on maps ranging from 9 to 100 km², CLEAR achieves a tenfold speedup over raw grid-based planning with only a 6.7% increase in path cost. Compared to alternative abstraction baselines, it yields 6–9% shorter paths and substantially improves planning reliability.
Application Category
📝 Abstract
Long-horizon navigation in unstructured environments demands terrain abstractions that scale to tens of km$^2$ while preserving semantic and geometric structure, a combination existing methods fail to achieve. Grids scale poorly; quadtrees misalign with terrain boundaries; neither encodes landcover semantics essential for traversability-aware planning. This yields infeasible or unreliable paths for autonomous ground vehicles operating over 10+ km$^2$ under real-time constraints. CLEAR (Connected Landcover Elevation Abstract Representation) couples boundary-aware spatial decomposition with recursive plane fitting to produce convex, semantically aligned regions encoded as a terrain-aware graph. Evaluated on maps spanning 9-100~km$^2$ using a physics-based simulator, CLEAR achieves up to 10x faster planning than raw grids with only 6.7% cost overhead and delivers 6-9% shorter, more reliable paths than other abstraction baselines. These results highlight CLEAR's scalability and utility for long-range navigation in applications such as disaster response, defense, and planetary exploration.
Problem
Research questions and friction points this paper is trying to address.
terrain abstraction
large-scale navigation
unstructured environments
semantic representation
traversability-aware planning
Innovation
Methods, ideas, or system contributions that make the work stand out.
terrain abstraction
semantic-geometric representation
boundary-aware decomposition
long-horizon navigation
traversability-aware planning
🔎 Similar Papers
No similar papers found.
