AirSplan improves safe quadrotor flight in complex 3D environments
AirSplan: Risk-Aware Motion Planning for Quadrotors in Cluttered 3D Gaussian Splats
Robotics
Summary
Flying small drones near obstacles is risky because even tiny collisions can cause crashes. The authors created AirSplan, a system that builds detailed 3D maps from uncertain data and then plans safe flight paths to avoid collisions. Their method uses a special way to represent the environment and optimizes the drone's path considering its flying dynamics. Tests show AirSplan finds collision-free routes much more often than previous methods.
What this means in practice
- •For agriculture drone operators: Plan safe flight routes around complex crops and structures using detailed 3D maps when GPS or ground-truth data is unavailable.
- •For infrastructure inspection teams: Automatically generate collision-free paths for drones inspecting cluttered structures like bridges or antennas using uncertain environment data.
Authors
Seth Isaacson, William Hong, Katherine A. Skinner, Ram Vasudevan
Abstract
Quadrotors are increasingly deployed in applications such as agriculture, infrastructure inspection, and maintenance. In each of these applications, the robot must navigate complex scene geometry while remaining strictly collision-free. Unlike in ground domains, even minor collisions for aerial vehicles can result in the loss of the robot. This safety requirement induces a pair of technical challenges. First, the environment must be represented with sufficient fidelity to encode complex structure, even when no ground-truth obstacle data is available. Second, a motion planner must leverage this representation to determine a collision-free path to the goal. This paper proposes a system that addresses these complementary challenges. The proposed method, AirSplan, adopts a normalized variant of 3D Gaussian Splatting that encodes high-fidelity scene geometry. It then applies a novel reachability-based motion planner that leverages the differential flatness of quadrotors to compute continuous-time collision constraints that tightly overapproximate the robot's occupancy. Experiments demonstrate that AirSplan successfully finds a path in 81.2% of challenging test cases, a significant improvement over the nearest baseline method's 51.2%.