Time-Optimized Safe Navigation in Unstructured Environments through Learning Based Depth Completion
Jeffrey Mao, Raghuram Cauligi Srinivas, Steven Nogar, Giuseppe Loianno
TL;DR
This work tackles safe, time-efficient autonomous navigation for SWaP-constrained quadrotors in unknown environments by fusing monocular and stereo depth to produce dense depth maps onboard, then building a graph of safe flight corridors via cuboid decomposition. A global planner operates on this graph with a Bernstein-based time-optimized trajectory generator that enforces safety through convex hull properties, while the map is incrementally updated and trajectories are replanned at 20 Hz. Key contributions include a real-time depth completion method without retraining, a fast shortest-path-on-convex-sets planner, and a Bernstein-based, safety-guaranteed trajectory generator, all validated onboard in indoor and outdoor tests with comparisons to MINCO. The approach yields faster planning with explicit safety guarantees, maintains robust performance under wind and sensor noise, and demonstrates real-time autonomous exploration using only lightweight sensors. This has practical impact for autonomous inspection, search-and-rescue, and other applications where SWaP constraints preclude heavy sensing and offline computation.
Abstract
Quadrotors hold significant promise for several applications such as agriculture, search and rescue, and infrastructure inspection. Achieving autonomous operation requires systems to navigate safely through complex and unfamiliar environments. This level of autonomy is particularly challenging due to the complexity of such environments and the need for real-time decision making especially for platforms constrained by size, weight, and power (SWaP), which limits flight time and precludes the use of bulky sensors like Light Detection and Ranging (LiDAR) for mapping. Furthermore, computing globally optimal, collision-free paths and translating them into time-optimized, safe trajectories in real time adds significant computational complexity. To address these challenges, we present a fully onboard, real-time navigation system that relies solely on lightweight onboard sensors. Our system constructs a dense 3D map of the environment using a novel visual depth estimation approach that fuses stereo and monocular learning-based depth, yielding longer-range, denser, and less noisy depth maps than conventional stereo methods. Building on this map, we introduce a novel planning and trajectory generation framework capable of rapidly computing time-optimal global trajectories. As the map is incrementally updated with new depth information, our system continuously refines the trajectory to maintain safety and optimality. Both our planner and trajectory generator outperforms state-of-the-art methods in terms of computational efficiency and guarantee obstacle-free trajectories. We validate our system through robust autonomous flight experiments in diverse indoor and outdoor environments, demonstrating its effectiveness for safe navigation in previously unknown settings.