A Single-Copter UWB-Ranging-Based Localization System Extendable to a Swarm of Drones

Abstract

This paper presents a single-copter localization system as a first step towards a scalable multihop drone swarm localization system. The drone was equipped with ultrawideband (UWB) transceiver modules, which can be used for communication, as well as distance measurement. The location of the drone was detected based on fixed anchor points using a single type of UWB transceiver. Our aim is to create a swarm localization system that enables drones to switch their role between an active swarm member and an anchor node to enhance the localization of the whole swarm. To this end, this paper presents our current baseline localization system and its performance regarding single-drone localization with fixed anchors and its integration into our current modular quadcopters, which was designed to be easily extendable to a swarm localization system. The distance between each drone and the anchors was measured periodically, and a specially tailored gradient descent algorithm was used to solve the resulting nonlinear optimization problem. Additional copter and wireless-specific adaptations were performed to enhance the robustness. The system was tested with a Vicon system as a position reference and showed a high precision of 0.2 m with an update rate of <10 Hz. Additionally, the system was integrated into the FINken copters of the SwarmLab and evaluated in multiple outdoor scenarios. These scenarios showed the generic usability of the approach, even though no accurate precision measurement was possible.

Two-Dimensional Cartesian Coordinate System Educational Toolkit: 2D-CACSET

“…For example, in a scenario described in this paper, the participant can manipulate the Tag to correlate placement and coordinates. In contrast, for a more advanced scenario, the Tag can be placed on a drone [39] or another mobile robot to carry out trajectory tracking tests (see Figure 20). Some alternative configurations that come from rearranging the RTLS elements are listed next.…”

Improving UWB Ranging Accuracy via Multiple Network Model with Second Order Motion Prediction

“…2 ○Ridge regression is a vector auto-regressive (VAR) model with L2 regularization, which has been popular in time series prediction problems (LRidge) [19]. 3 ○Vector autoregressive (VAR) model with support vector regression objective function (LSVR) [32]. 4 ○Locally Weighted Linear Regression (LWLR).…”

“…It has broad application prospects in wireless communication, radar tracking, and precise positioning [2]. In addition, ranging technology has also played a great role in the aerospace field, such as the space exploration of China's Chang'e-3 and Chang'e-4, the auxiliary navigation and obstacle avoidance during the landing of the probe, and the fuel level measurement of the carrier rocket [3].…”

A Survey of Robot Swarms’ Relative Localization Method

“…UAVs are mostly used in outdoor environments, and many studies have researched relative localization in outdoor environments. Steup et al [81] set UWB anchors in an outdoor environment to deal with the problem of GPS signal loss, providing UAVs with a positioning system with an accuracy of about 0.2 m. Lazzari et al [43] deployed a UWB relative localization system containing four movable anchors on a vehicle to meet the tracking and control of UAVs in the following convoys. The system had a horizontal error of 30 cm, an altitude error of 50 cm, and a positioning refresh interval of 6.5 ms to 12.5 ms. Cao et al [70] addressed the challenge of close formation of multiple fixed-wing UAVs in lead-follow mode by loading a UWB positioning system with three anchors on a pilot UAV.…”