Voxfield: Non-Projective Signed Distance Fields for Online Planning and 3D Reconstruction

Abstract

Creating accurate maps of complex, unknown environments is of utmost importance for truly autonomous navigation robot. However, building these maps online is far from trivial, especially when dealing with large amounts of raw sensor readings on a computation and energy constrained mobile system, such as a small drone. While numerous approaches tackling this problem have emerged in recent years, the mapping accuracy is often sacrificed as systematic approximation errors are tolerated for efficiency's sake. Motivated by these challenges, we propose Voxfield, a mapping framework that can generate maps online with higher accuracy and lower computational burden than the state of the art. Built upon the novel formulation of non-projective truncated signed distance fields (TSDFs), our approach produces more accurate and complete maps, suitable for surface reconstruction. Additionally, it enables efficient generation of Euclidean signed distance fields (ESDFs), useful e.g., for path planning, that does not suffer from typical approximation errors. Through a series of experiments with public datasets, both real-world and synthetic, we demonstrate that our method beats the state of the art in map coverage, accuracy and computational time. Moreover, we show that Voxfield can be utilized as a back-end in recent multiresolution mapping frameworks, producing high quality maps even in large-scale experiments. Finally, we validate our method by running it onboard a quadrotor, showing it can generate accurate ESDF maps usable for real-time path planning and obstacle avoidance.

From SLAM to Situational Awareness: Challenges and Survey

“…This helps improve computational efficiency and reduce storage requirements while accurately reconstructing the observed scene. TSDFs have been demonstrated in multiple works such as Voxgraph [ 224 ], Freetures [ 225 ], Voxblox++ [ 226 ], or the more recent Voxblox-Field [ 227 ]. They can create and maintain globally consistent volumetric maps that are lightweight enough to run on computationally constrained platforms and demonstrate that the resulting representation can navigate unknown environments.…”

Leveraging 3-D Data for Whole Object Shape and Reflection Aware 2-D Map Building

“…While the 2D sensor directly detects the background wall, since it is placed almost at ground height, the 3D sensor is able to capture all object information. Using 3D maps instead could solve the collision problem, as shown in works such as [5]- [7]. However, their construction and use is computationally expensive and memory requirements are very high for their storage.…”