Problem
Research questions and friction points this paper is trying to address.
UAV flight
occlusion awareness
obstacle avoidance
motion planning
partially unknown environments
OA-MPPI: Occlusion-Aware Model Predictive Path Integral Control for UAV Flight
🤖 AI Summary
This study addresses the collision risks faced by unmanned aerial vehicles (UAVs) in occluded environments due to imperceptible hidden dynamic obstacles. We propose an occlusion-aware Model Predictive Path Integral (MPPI) control method that extracts three-dimensional occlusion boundaries online and innovatively converts them into dynamically reachable regions for potential moving agents. Under nonlinear dynamics and thrust constraints, the MPPI sampling process penalizes trajectories intruding into these regions, thereby explicitly handling unknown threats. Both simulation and real-world flight experiments demonstrate that the proposed approach significantly increases the clearance from occlusion boundaries and successfully avoids suddenly emerging obstacles. Furthermore, the entire pipeline supports real-time onboard execution.
📝 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.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Occlusion-Aware
Model Predictive Path Integral (MPPI)
UAV Flight
3D Occlusion Boundary
Online Occupancy Map
🔎 Similar Papers
No similar papers found.
