UWB Aided Mobile Robot Localization with Neural Networks and the EKF

Abstract

This paper exploits the use of Ultra Wide Band (UWB) technology to improve the localization of robots in both indoor and outdoor environments. In order to efficiently integrate the UWB technology in existing multi-sensor architectures, such as Kalman-based, we propose two approaches to estimate the UWB position covariance values. The first approach uses statistical methods to estimate static covariance values based on data acquired a priori. The second approach adopts a neural network (NN) to capture the relationship between the positional error of the UWB data and the signal quality information, such as the Estimate Of Precision (EOP) and Received Signal Strength Indicator (RSSI). The GPS-RTK is used as ground truth and RGB-D odometry is adopted for both benchmarking and integration purposes. Position sources are fused by means of an Extended Kalman Filter (EKF). Real world experiments are conducted with a tracked mobile robot driving outdoors in a closed-loop trajectory. Results show that the NN is able to efficiently model the sensor covariances and adapt the trustworthiness of the EKF estimation, overcoming data loss by relying on the other available estimation source.

NR5G-SAM: A SLAM Framework for Field Robot Applications Based on 5G New Radio

“…However, only a small number of 5G SLAM works have been proposed so far. Given a set of gNB stations placed over large kilometric distances, position tracking can be seen as a non-linear combinatorial problem [16]. A robot carrying UE may use 5G NR to track its position over time by comparing the ranges from each base station and infer its position, as shown in Figure 1.…”

“…An optimization of the ToA ranges is performed using a Genetic Algorithm (GA) to minimize the horizontal error during multilateration. Our previous developments [16] involved benchmarking three meta-heuristics for 5G positioning. The GA implementation provided superior results in terms of the mean execution time and average Euclidean error, thus becoming a natural design choice.…”

“…In a previous work [16], we developed a realistic simulation testbed that includes an agricultural environment, as shown in Figure 5. We use the Robot Operating System (ROS) (https://www.ros.org, accessed on 4 March 2023) for perception and actuation of the field robot, and the Gazebo simulator (https://gazebosim.org, accessed on 4 March 2023) with integrated physics engine that generates the ground truth pose and the required raw data from the robot's sensors, including the IMU and wheel odometry.…”

“…In terms of positioning, the GA optimization of the UEB follows the setup presented in [16]. For the BEB, the GA has been set with a population size of 200 and an elite count of 0.05.…”

“…In order to demonstrate its significance, we compare NR5G-SAM with LIO-SAM [15], a state-ofthe-art LiDAR-inertial SLAM framework to highlight the advantages and disadvantages it can potentially provide to the mobile robotics domain. This work is an extension of our previous work on 5G Positioning [16] and serves as an important development step towards an innovative and robust cellular SLAM framework for field robotics.…”