Multiple UAV Adaptive Navigation for Three-Dimensional Scalar Fields

Abstract

Adaptive Navigation (AN) control strategies allow an agent to autonomously alter its trajectory based on realtime measurements of the environment. Compared to conventional navigation methods, these techniques can reduce required time and energy to explore scalar characteristics of unknown and dynamic regions of interest (e.g., temperature, concentration level). Multiple Uncrewed Aerial Vehicle (UAV) approaches to AN can improve performance by exploiting synchronized spatially-dispersed measurements to generate realtime information regarding the structure of the local scalar field, which is then used to inform navigation decisions. This article presents initial results of a comprehensive program to develop, verify, and experimentally implement mission-level AN capabilities in three-dimensional (3D) space using our unique multilayer control architecture for groups of vehicles. Using our flexible formation control system, we build upon our prior 2D AN work and provide new contributions to 3D scalar field AN by a) demonstrating a wide range of 3D AN capabilities using a unified, multilayer control architecture, b) extending multivehicle 2D AN control primitives to navigation in 3D scalar fields, and c) introducing statebased sequencing of these primitive AN functions to execute 3D mission-level capabilities such as isosurface mapping and plume following. We verify functionality using high-fidelity simulations of multicopter drone clusters, accounting for vehicle dynamics, outdoor wind gust disturbances, position sensor inaccuracy, and scalar field sensor noise. This paper presents the multilayer architecture for multivehicle formation control, the 3D AN control primitives, the sequencing approaches for specific mission-level capabilities, and simulation results that demonstrate these functions.

From Shadows to Light: A Swarm Robotics Approach With Onboard Control for Seeking Dynamic Sources in Constrained Environments

“…The UAVs estimate the signal strength gradient's peak by averaging each other's measurements. Lastly, the authors in [15] present centralized adaptive navigation control strategies to characterize sources in 2D and 3D scalar fields, also accounting for plume dynamics. The process incorporates the collective use of individual measurements from diverse environmental points by the cluster members.…”

“…We use fully onboard controlled Crazyflie nano-drones in an indoor environment (unlike control from a central computer as in [11] or only on simulation as in [14] and [15]), a physical custom-made on-board sensor to implement scalar source sensing (differently from [13] where the measurements are calculated by a stored function), and another physical onboard sensor for sensing and avoiding an obstacle (which is not achieved in [13], [14] and [15]). Each drone senses the gradient via its sensor and retrieves a scalar value (the intensity of the source at its location), but it does not communicate this measurement to other drones (unlike the cases where measurements are communicated to peers for the estimation in [12], [13], [14] and [15]). The drone does not store nor utilizes any historical data, and the method operates reactively unlike other studies we discuss in this section.…”

DLSMR: Deep Learning-Based Secure Multicast Routing Protocol against Wormhole Attack in Flying Ad Hoc Networks with Cell-Free Massive Multiple-Input Multiple-Output

“…These modern communications setups rely on effective routing methods which have a direct impact on performance and efficiency. Recently, security has emerged as a significant concern, warranting special attention to prevent potential breaches [ 3 , 4 , 5 , 6 , 7 , 8 ].…”