Uav flight control accounts for hidden obstacles and moving agents
OA-MPPI: Occlusion-Aware Model Predictive Path Integral Control for UAV Flight
Robotics
Summary
Flying drones safely through places with many obstacles is tricky, especially when some objects are hidden from the drone's sensors. This paper presents a method where the drone predicts where unseen obstacles or moving things might appear from hidden spaces. By considering these risks during its flight planning, the drone can choose safer paths. The researchers tested their approach both in computer simulations and real drone flights, showing improved safety around hidden areas compared to older methods.
What this means in practice
- •For drone navigation engineers: Improve drone flight safety by predicting and avoiding threats emerging from hidden spaces during autonomous navigation through cluttered environments.
- •For robotics software developers: Incorporate occlusion-aware control into robotic platforms to better plan paths in spaces with limited sensor visibility and potential moving obstacles.
Authors
Vittorio Palladino, Teaya Yang, Ruiqi Zhang, Mark W. Mueller
Abstract
Autonomous UAV flight through cluttered and partially unknown environments requires reasoning not only about observed obstacles but also about occluded regions that the sensor cannot observe. We present OA-MPPI, an obstacle- and occlusion-aware extension of Model Predictive Path Integral (MPPI) control for quadrotor flight that accounts for potential moving agents emerging from these regions into the vehicle's path. At every planning step, we extract a 3D occlusion boundary from the online occupancy map and use it to model the regions that hidden agents could reach over the prediction horizon. We penalize trajectories that enter these expanding regions within MPPI rollouts generated using nonlinear quadrotor dynamics and accounting for individual rotor thrust limits. We validate the proposed approach in simulation and hardware flight experiments, with the complete pipeline running onboard the vehicle in real time. Results show increased clearance from occlusion boundaries compared to baseline MPPI in both settings, as well as avoidance of an agent emerging from occlusion in simulation.