C-3TO: Continuous 3D Trajectory Optimization on Neural Euclidean Signed Distance Fields

2025-09-24

AI Summary

This work addresses the problem of continuous 3D trajectory optimization for local replanning of unmanned aerial vehicles (UAVs) in cluttered environments. We propose a novel trajectory optimization framework based on a continuous neural Euclidean Signed Distance Field (Neural ESDF). Unlike conventional approaches relying on discrete ESDF grids and interpolation, our method jointly models quintic polynomial trajectory parameterization with an implicit Neural ESDF, enabling end-to-end, gradient-accurate nonlinear optimization directly in continuous space. We introduce an innovative two-stage optimization strategy that simultaneously ensures collision avoidance, dynamical feasibility, computational efficiency, and robustness. Experimental results demonstrate that the method consistently generates safe, smooth, and dynamically feasible trajectories across diverse complex scenarios. It supports online environmental updates and adaptive local trajectory window adjustment, significantly improving optimization accuracy, safety guarantees, and generalization capability compared to prior methods.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Intelligent Robots: Learning & Optimization for ROB
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This paper introduces a novel framework for continuous 3D trajectory optimization 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.

Problem

• Optimizing continuous 3D trajectories in cluttered environments
• Overcoming limitations of discretized ESDF grids with interpolation
• Ensuring collision-aware dynamically feasible trajectories for aerial robotics

Innovation

• Continuous neural ESDF for trajectory optimization
• Fifth-order polynomial smooth trajectory representation
• Two-stage nonlinear optimization balancing efficiency safety