Core Concepts
This paper presents a novel control strategy for drone networks that simultaneously controls both the camera orientation and drone translational motion to improve the quality of 3D structures reconstructed from aerial images.
Abstract
The paper presents a novel control strategy for drone networks to improve the quality of 3D structures reconstructed from aerial images. Unlike existing coverage control strategies, the proposed approach simultaneously controls both the camera orientation and drone translational motion, enabling more comprehensive perspectives and enhancing the overall map quality.
The key highlights are:
Novel problem formulation including a new performance function to evaluate the drone positions and camera orientations.
Design of a QP-based controller with a control barrier-like function for a constraint on the decay rate of the objective function.
Technological approach to address the increased computational complexity by introducing JAX, utilizing just-in-time (JIT) compilation and Graphical Processing Unit (GPU) acceleration.
Extensive verifications through simulation in ROS (Robot Operating System) demonstrating the real-time feasibility of the controller and its superiority over the conventional method.
Stats
The paper presents the following key metrics and figures:
The target field is modeled as a cube with a range of [-1, 1]m x [-1, 1]m x [0, 0.5]m.
The viewing angle space is set to θh ∈ [-π, π) and θv ∈ [π/6, π/2].
The number of drones n is set to 3, with their initial positions uniformly distributed.
The field is divided into m = 1.5 x 10^7 small cells of size 0.02m x 0.02m x 0.1m x π/30rad x π/30rad.
The controller parameters are set as: a1 = 5.0, a2 = 1.0, σ1 = 0.13, σ2 = 0.18, ϵ = 0.0001, γ = 0.05, δ = 5.0.