Carmelo Di Franco

Energy-Aware Coverage Path Planning of UAVs

Coverage path planning is the operation of finding a path that covers all the points of a specific area. Thanks to the recent advances of hardware technology, Unmanned Aerial Vehicles (UAVs) are starting to be used for photogrammetric sensing of large areas in several application domains, such as agriculture, rescuing, and surveillance. However, most of the research focused on finding the optimal path taking only geometrical constraints into account, without considering the peculiar features of the robot, like available energy, weight, maximum speed, sensor resolution, etc. This paper proposes an energy-aware path planning algorithm that minimizes energy consumption while satisfying a set of other requirements, such as coverage and resolution. The algorithm is based on an energy model derived from real measurements. Finally, the proposed approach is validated through a set of experiments.

Coverage Path Planning for UAVs Photogrammetry with Energy and Resolution Constraints

Unmanned Aerial Vehicles (UAVs) are starting to be used for photogrammetric sensing of large areas in several application domains, such as agriculture, rescuing, and surveillance. In this context, the problem of finding a path that covers the entire area of interest is known as Coverage Path Planning (CPP). Although this problem has been addressed by several authors from a geometrical point of view, other issues such as energy, speed, acceleration, and image resolution are not often taken into account. To fill this gap, this paper first proposes an energy model derived from real measurements, and then uses this model to implement a coverage path planning algorithm for reducing energy consumption, as well as guaranteeing a desired image resolution. In addition, two safety mechanisms are presented: the first, executed off-line, checks whether the energy stored in the battery is sufficient to perform the planned path; the second, performed online, triggers a safe return-to-launch (RTL) operation when the actual available energy is equal to the energy required by the UAV to go back to the starting point.

Data fusion for relative localization of wireless mobile nodes

Monitoring teams of mobile nodes is becoming crucial in a growing number of activities. When it is not possible to use fix references or external measurements, a practicable solution is to derive relative positions from local communication. In this work, we propose an anchor-free Received Signal Strength Indicator (RSSI) method aimed at small multi-robot teams. Information from Inertial Measurement Unit (IMU) mounted on the nodes and processed with a Kalman Filter are used to estimate the robot dynamics, thus increasing the quality of RSSI measurements. A Multidimensional Scaling algorithm is then used to compute the network topology from improved RSSI data provided by all nodes. A set of experiments performed on data acquired from a real scenario show the improvements over RSSI-only localization methods. With respect to previous work only an extra IMU is required, and no constraints are imposed on its placement, like with camera-based approaches. Moreover, no a-priori knowledge of the environment is required and no fixed anchor nodes are needed.

Solving ambiguities in MDS relative localization

Monitoring teams of mobile nodes is becoming crucial in a growing number of activities. Where it is not possible to use fixed references or external measurements, one of the possible solutions involves deriving relative positions from local communication. Well-known techniques such as trilateration and multilateration exist to locate a single node although such methods are not designed to locate entire teams. The technique of Multidimensional Scaling (MDS), however, allow us to find the relative coordinates of entire teams starting from the knowledge of the inter-node distances. However, like every relative-localization technique, it suffers from geometrical ambiguities including rotation, translation, and flip. In this work, we address such ambiguities by exploiting the node velocities to correlate the relative maps at two consecutive instants. In particular, we introduce a new version of MDS, called enhanced Multidimensional Scaling (eMDS), which is able to handle these types of ambiguities. The effectiveness of our localization technique is then validated by a set of simulation experiments and our results are compared against existing approaches.

Fusing Time-of-Flight and Received Signal Strength for adaptive radio-frequency ranging

Teams of mobile cooperative robots are ideal candidates for applications where the presence of humans is impossible or should be avoided. Knowing the positions of the robots in crucial in such scenarios. A possible solution is to derive relative positions from local communication. In this work, we propose an anchor-free online channel estimation method aimed at small multi-robot teams. By combining both the Time-of-Flight (ToF) and Received Signal Strength Indicator (RSSI) ranging, provided by the nanoLoc devices, we perform an online estimation of the indoor log-distance path loss model. This model will then be used together with an Extended Kalman Filter to track distance between every pair of units. The advantages compared to previous work are: 1) we do not use any extra sensors, since all the data is captured from the transceiver module; 2) we do not use any a priori knowledge, the channel model is estimated online, without the need of fixed anchor nodes; 3) we support the high dynamics of RSSI with the improved accuracy of ToF.

Calibration-free network localization using non-line-of-sight ultra-wideband measurements

We present a method for calibration-free, infrastructure-free localization in sensor networks. Our strategy is to estimate node positions and noise distributions of all links in the network simultaneously -- a strategy that has not been attempted thus far. In particular, we account for biased, NLOS range measurements from UWB devices that lead to multi-modal noise distributions, for which few solutions exist to date. Our approach circumvents cumbersome a-priori calibration, allows for rapid deployment in unknown environments, and facilitates adaptation to changing conditions. Our first contribution is a generalization of the classical multidimensional scaling algorithm to account for measurements that have multi-modal error distributions. Our second contribution is an online approach that iterates between node localization and noise parameter estimation. We validate our method in 3-dimensional networks, (i) through simulation to test the sensitivity of the algorithm on its design parameters, and (ii) through physical experimentation in a NLOS environment. Our setup uses UWB devices that provide time-of-flight measurements, which can lead to positively biased distance measurements in NLOS conditions. We show that our algorithm converges to accurate position estimates, even when initial position estimates are very uncertain, initial error models are unknown, and a significant proportion of the network links are in NLOS.

Multidimensional scaling localization with anchors

Multidimensional Scaling (MDS) is a widely used technique for visualizing a set of objects in an n-dimensional space. It has been extensively applied in wireless sensor networks for deriving the coordinates of a set of nodes in distance-based Localization. Many variants of MDS have been proposed to overcome issues such as partial connectivity and different types of noise in the measurements. In particular, some works adapted and modified the MDS technique to include the notion of anchors. However, in order to maintain the original formulation of MDS, the algorithm was twisted by adding constraints to the minimization function or adapting the final result through roto-translations. Unfortunately, however, these adaptions do not fully solve the problem, because they try to align the relative positions of the nodes to the global reference system provided by the anchors only after the MDS algorithm. This paper provides a theoretical generalization of the classical MDS algorithm when some of the coordinates of some elements (e.g., anchors in the case of localization) are known. The proposed generalization can be applied to any of the many MDS variants (e.g., classical MDS, ordinal MDS, MDS-MAP, GM-MDS) that minimize the stress function with the SMACOF technique. The formulation is proved to be correct and does not add any constraints to MDS.

Design and analysis of target-sensitive real-time systems

A significant number of real‐time control applications include computational activities where the results have to be delivered at precise instants, rather than within a deadline. The performance of such systems significantly degrades if outputs are generated before or after the desired target time. This work presents a general methodology that can be used to design and analyze target‐sensitive applications in which the timing parameters of the computational activities are tightly coupled with the physical characteristics of the system to be controlled. For the sake of clarity, the proposed methodology is illustrated through a sample case study used to show how to derive and verify real‐time constraints from the mission requirements. Software implementation issues necessary to map the computational activities into tasks running on a real‐time kernel are also discussed to identify the kernel mechanisms necessary to enforce timing constraints and analyze the feasibility of the application. A set of experiments are finally presented with the purpose of validating the proposed methodology. Copyright

Target-sensitive systems: Analysis and implementation issues

Several real-time applications include tasks in which the output must be produced at precise time instants, rather than “within” a deadline, and the overall system performance significantly degrades when the task is executed too late or too early with respect to the desired time. This paper illustrates one of such applications and takes it as a reference case study to propose a general approach to show how to derive the timing constraints from the application requirements, how to implement the application on top of a realtime kernel, identifying the operating system features necessary to enforce such constraints, and how to analyze the schedulability of the task set. A set of experimental results are also presented to validate the proposed approach.

Dynamic Multidimensional Scaling with anchors and height constraints for indoor localization of mobile nodes

Abstract In distance-based localization, estimating the position of a network of wireless sensors is not an easy task. The problem increases when dealing with moving nodes and cluttered indoor environments. Many algorithms have been proposed in the literature and, among them, the Multidimensional Scaling (MDS) technique gained a lot of interest due to its resilience to flips ambiguities and easiness of use. Many variants of MDS have been proposed to overcome issues such as partial connectivity or distributed computation. In this context, it is common to place some anchors nodes to help in estimating the coordinates of the network correctly. However, instead of using the anchor’s positions directly during the minimization of the MDS cost function, most approaches act on the estimated coordinates at the end of the MDS computation without fully utilizing the knowledge about anchors. In this work, the classic MDS and Dynamic MDS have been reformulated to utilize the anchor’s position inside the minimization function. A set of real experiments in 3D with Ultrawide-band devices show that our approach considerably improves the accuracy of localization with respect to the usual MDS techniques.