Indoor localization of mobile agents using wireless technologies is becoming very important in military and civil applications. This paper introduces an approach for the indoor localization of a mini UAV based on Ultra-WideBand technology, low cost IMU and vision based sensors. In this work an Extended Kalman Filter (EKF) is introduced as a possible technique to improve the localization. The proposed approach allows to use a low-cost IMU (inertial measurement unit) in the prediction step and the integration of vision-odometry for the detection of markers nearness the touchdown area. The ranging measurements allows to reduce the error of inertial measurement due to the limited performance of accelerometers and gyros. The obtained results show that an accuracy of 10 cm can be achieved.
The capability to instantiate a cooperation among heterogeneous agents is a fundamental feature in mobile robotics. In this paper we focus on the interaction between Unmanned Ground Vehicle (UGV) and Unmanned Aerial Vehicle (UAV) to extend the endurance of UAV, thanks to a novel landing/recharging platform. The UGV acts as a docking station and hosts the UAV during the indoor/outdoor transition and vice-versa. We designed a platform and a robust landing target to automate the fast recharge of UAV. The synchronization and coordination of cooperation is managed by a Ground Control Station (GCS) developed using a versatile software toolchain based on the integration of Stateflow, auto-generation of Ccode and ROS. All the software components of UAV, UGV and GCS have been developed using ROS. The obtained results show that the UAV is able to land over the UGV with high accuracy (<5cm for both x and y axis) thanks to a visual position estimation algorithm, also in presence of wind (with gust up to 20-25km/h), recharging its batteries in a short time to extend its endurance.
Determining the residual life time of systems is a determinant factor for machinery and environment safety. In this paper the problem of estimate the residual useful life (RUL) of turbo-fan engines is addressed. The adopted approach is especially suitable for situations in which a large amount of data is available offline, by allowing the processing of such data for the determination of RUL. The procedure allows to calculate the RUL through the following steps: features extraction by Artificial Neural Networks (ANN) and determination of remaining life time by-prediction models based on a Hidden Markov Model (HMM). Simulations confirm the effectiveness of the proposed approach and the promising power of Bayesian methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.