🤖 AI Summary
Non-convex trajectory optimization for vision-based autonomous flight in unknown environments suffers from high computational cost and susceptibility to local minima.
Method: We propose a “learning-based initialization” planning framework: a neural network directly regresses high-quality trajectory initializations from visual-inertial observations, followed by spatial-temporal joint non-convex optimization for refinement—balancing interpretability and reliability. We introduce a novel batch-sampling-driven multimodal data generation paradigm to enable end-to-end initialization learning, and design a latency-tolerant online replanning mechanism to ensure robust real-time execution.
Results: Simulations demonstrate several-fold speedup over pure optimization methods while maintaining trajectory quality. Real-world UAV experiments validate real-time performance and navigation robustness in complex, unstructured environments.
📝 Abstract
Autonomous flight in unknown environments requires precise planning for both the spatial and temporal profiles of trajectories, which generally involves nonconvex optimization, leading to high time costs and susceptibility to local optima. To address these limitations, we introduce the Learning-Initialized Trajectory Planner (LIT-Planner), a novel approach that guides optimization using a Neural Network (NN) Planner to provide initial values. We first leverage the spatial-temporal optimization with batch sampling to generate training cases, aiming to capture multimodality in trajectories. Based on these data, the NN-Planner maps visual and inertial observations to trajectory parameters for handling unknown environments. The network outputs are then optimized to enhance both reliability and explainability, ensuring robust performance. Furthermore, we propose a framework that supports robust online replanning with tolerance to planning latency. Comprehensive simulations validate the LIT-Planner's time efficiency without compromising trajectory quality compared to optimization-based methods. Real-world experiments further demonstrate its practical suitability for autonomous drone navigation.
