RT Conference Proceedings
T1 C-3TO: Continuous 3D Trajectory Optimization on Neural Euclidean Signed Distance Fields
A1 Gil García, Guillermo
A1 Cobano-Suárez, José-Antonio
A1 Merino, Luis
A1 Caballero, Fernando
K1 Trajectory Planning
K1 Distance Map
K1 Local Path
K1 Signed Distance Function
K1 Safe Path
K1 3D Trajectory Planning
K1 Path Planning
K1 Trajectory Optimization
K1 Neural Network
K1 Urban Planning
K1 Pathfinding
K1 Representation Of The Environment
K1 Planning Algorithm
K1 Software Framework
K1 Efficient Path
K1 Path Computation
K1 Robot Operating System
K1 Safe Navigation
AB This paper introduces a novel framework for continuous 3D trajectory optimization (C-3TO) in cluttered environments, leveraging online neural Euclidean Signed Distance Fields (ESDFs). Unlike prior approaches that rely on discretized ESDF grids with interpolation, our method directly optimizes smooth trajectories represented by fifth-order polynomials over a continuous neural ESDF, ensuring precise gradient information throughout the entire trajectory. The framework integrates a two-stage nonlinear optimization pipeline that balances efficiency, safety and smoothness. Experimental results demonstrate that C-3TO produces collision-aware and dynamically feasible trajectories. Moreover, its flexibility in defining local window sizes and optimization parameters enables straightforward adaptation to diverse user’s needs without compromising performance. By combining continuous trajectory parameterization with a continuously updated neural ESDF, C-3TO establishes a robust and generalizable foundation for safe and efficient local replanning in aerial robotics.
PB IEEE
YR 2026
FD 2026-07-14
LK https://hdl.handle.net/10433/27332
UL https://hdl.handle.net/10433/27332
LA en
NO 2026 International Conference on Unmanned Aircraft Systems (ICUAS), Corfu, Greece, 2026, pp. 161-168
NO Departamento de Deporte e Informática
NO Service Robotics Laboratory
DS RIO
RD Sep 14, 2026