Track genomics, neuroscience and molecular biology.
Search papers, labs, and topics across Lattice.
This paper introduces a vision-guided optimal control framework for autonomous propulsive drone navigation on the Moon, enabling safe and fuel-efficient hops between waypoints. The system integrates a semantic segmentation model for obstacle detection and a rapid terrain reachability calculation to inform model predictive control (MPC) for robust trajectory tracking. Simulation results confirm the framework's ability to generate safe, energy-efficient trajectories that can be executed with high fidelity.
Lunar drones can now autonomously navigate complex terrain, safely landing even when initial waypoints become unreachable, by leveraging vision-guided optimal control and real-time obstacle avoidance.
Propulsive drones are a promising new technology for generating high-resolution terrain models of the Moon. In future lunar missions, propulsive drones will complete a series of hops across the lunar surface while capturing images with downward-facing cameras. We present an autonomous motion planning framework that uses a vision-guided optimal control strategy for performing safe, efficient hops between specified waypoints. The drone plans parabolic hops between waypoints generated by a global-level planner; if a nominal waypoint is unsafe or unreachable, the drone then completes a diversion maneuver to safely land. To evaluate the validity of potential landing sites, we develop a custom semantic segmentation model for identifying obstacles from aerial images, and we introduce an approach for rapidly calculating reachable terrain. We then use model predictive control (MPC) to robustly track hop trajectories that guide the drone towards safe landing sites. We test the optimal hop planner in a high-fidelity lunar simulation environment. Our results demonstrate that our unified motion planning framework yields safe, fuel-efficient hop trajectories which can be tracked rapidly and with low error via MPC.