Showing papers by "Roberto Sabatini published in 2021"

A High-Integrity and Low-Cost Navigation System for Autonomous Vehicles

Abstract: The introduction of autonomous vehicles can potentially lead to enhanced situational awareness and safety for road transport. However, the performance required for autonomous operations places stringent requirements on the vehicle navigation systems design. Relevant performance measures include not only the accuracy of the system but also its ability to detect sensor faults within a specified Time-to-Alert (TTA) and without generating an excessive number of false alarms. A new integrated navigation system architecture is proposed which utilizes Global Navigation Satellite Systems (GNSS), low-cost Inertial Navigation Systems (INS), visual odometry and Vehicle Dynamic Models (VDM). The system design is based on various navigation modes, each with independent failure mechanisms and fault-detection capabilities. A two-step data fusion approach is adopted to optimize the system accuracy and integrity performance. This includes a Knowledge-Based Module (KBM) performing a detailed sensor integrity analysis followed by a conventional Extended Kalman Filer (EKF). CAV navigation integrity requirement (i.e., alert limits and time-to-alert) are considered in the KBM where fault detection probabilities are calculated for each mode and translated to protection levels. A simulation case study is executed to verify the performance of each navigation mode in the presence of faults affecting the individual navigation sensors. Results confirm that the required CAV integrity performances are met, while the inclusion of visual odometry and VDM data provides significant performance enhancements both in terms of accuracy and integrity over existing INS/GNSS systems.

Explanation of Machine-Learning Solutions in Air-Traffic Management

Abstract: Advances in the trusted autonomy of air-traffic management (ATM) systems are currently being pursued to cope with the predicted growth in air-traffic densities in all classes of airspace. Highly automated ATM systems relying on artificial intelligence (AI) algorithms for anomaly detection, pattern identification, accurate inference, and optimal conflict resolution are technically feasible and demonstrably able to take on a wide variety of tasks currently accomplished by humans. However, the opaqueness and inexplicability of most intelligent algorithms restrict the usability of such technology. Consequently, AI-based ATM decision-support systems (DSS) are foreseen to integrate eXplainable AI (XAI) in order to increase interpretability and transparency of the system reasoning and, consequently, build the human operators’ trust in these systems. This research presents a viable solution to implement XAI in ATM DSS, providing explanations that can be appraised and analysed by the human air-traffic control operator (ATCO). The maturity of XAI approaches and their application in ATM operational risk prediction is investigated in this paper, which can support both existing ATM advisory services in uncontrolled airspace (Classes E and F) and also drive the inflation of avoidance volumes in emerging performance-driven autonomy concepts. In particular, aviation occurrences and meteorological databases are exploited to train a machine learning (ML)-based risk-prediction tool capable of real-time situation analysis and operational risk monitoring. The proposed approach is based on the XGBoost library, which is a gradient-boost decision tree algorithm for which post-hoc explanations are produced by SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME). Results are presented and discussed, and considerations are made on the most promising strategies for evolving the human–machine interactions (HMI) to strengthen the mutual trust between ATCO and systems. The presented approach is not limited only to conventional applications but also suitable for UAS-traffic management (UTM) and other emerging applications.

Laser Beam Atmospheric Propagation Modelling for Aerospace LIDAR Applications

Abstract: Atmospheric effects have a significant impact on the performance of airborne and space laser systems. Traditional models used to predict propagation effects rely heavily on simplified assumptions of the atmospheric properties and their interactions with laser systems. In the engineering domain, these models need to be continually improved in order to develop tools that can predict laser beam propagation with high accuracy and for a wide range of practical applications such as LIDAR (light detection and ranging), free-space optical communications, remote sensing, etc. The underlying causes of laser beam attenuation in the atmosphere are examined in this paper, with a focus on the dominant linear effects: absorption, scattering, turbulence, and non-linear thermal effects such as blooming, kinetic cooling, and bleaching. These phenomena are quantitatively analyzed, highlighting the implications of the various assumptions made in current modeling approaches. Absorption and scattering, as the dominant causes of attenuation, are generally well captured in existing models and tools, but the impacts of non-linear phenomena are typically not well described as they tend to be application specific. Atmospheric radiative transfer codes, such as MODTRAN, ARTS, etc., and the associated spectral databases, such as HITRAN, are the existing tools that implement state-of-the-art models to quantify the total propagative effects on laser systems. These tools are widely used to analyze system performance, both for design and test/evaluation purposes. However, present day atmospheric radiative transfer codes make several assumptions that reduce accuracy in favor of faster processing. In this paper, the atmospheric radiative transfer models are reviewed highlighting the associated methodologies, assumptions, and limitations. Empirical models are found to offer a robust analysis of atmospheric propagation, which is particularly well-suited for design, development, test and evaluation (DDT&E) purposes. As such, empirical, semi-empirical, and ensemble methodologies are recommended to complement and augment the existing atmospheric radiative transfer codes. There is scope to evolve the numerical codes and empirical approaches to better suit aerospace applications, where fast analysis is required over a range of slant paths, incidence angles, altitudes, and atmospheric conditions, which are not exhaustively captured in current performance assessment methods.

Sound Propagation Modelling for Manned and Unmanned Aircraft Noise Assessment and Mitigation: A Review

Abstract: This paper addresses one of the recognized barriers to the unrestricted adoption of Unmanned Aircraft (UA) in mainstream urban use—noise—and reviews existing approaches for estimating and mitigating this problem. The aircraft noise problem is discussed upfront in general terms by introducing the sound emission, propagation, and psychoacoustic effects. The propagation of sound in the atmosphere, which is the focus of this paper, is then analysed in detail to isolate the environmental and operational factors that predominantly influence the perceived noise on the ground, especially looking at large-scale low-altitude UA operations, such as in the envisioned Urban Air Mobility (UAM) concepts. The physics of sound propagation are presented, considering all attenuation effects and the anomalies due to Doppler and atmospheric effects, such as wind, thermal inversion, and turbulence. The analysis allows to highlight the limitations of current mainstream aircraft noise modelling and certification approaches and, in particular, their inadequacy in addressing the noise of UA and, more generally, UAM vehicles. This finding is important considering that, although reducing noise at the source has remained a priority for manufacturers to enable the scaling up of UAM and drone delivery operations in the near future, the impact of poorly considered propagation and psychoacoustic effects on the actual perceived noise on the ground is equally important for the same objective. For instance, optimizing the flight paths as a function of local weather conditions can significantly contribute to minimizing the impact of noise on communities, thus paving the way for the introduction of full-scale UAM operations. A more reliable and accurate modelling of noise ground signatures for both manned and unmanned low-flying aircraft will aid in identifying the real-time data stream requirements from distributed sensors on the ground. New developments in surrogate sound propagation models, more pervasive real-time sensor data, and suitable computing resources are expected to both yield more reliable and effective estimates of noise reaching the ground listeners and support a dynamic planning of flight paths.

Reinforcement Learning-Based Flow Management Techniques for Urban Air Mobility and Dense Low-Altitude Air Traffic Operations

Abstract: As Unmanned Aircraft Systems (UAS) technology matures, and the demand for UAS commercial operations is gradually increasing, a widespread proliferation of UAS operations may lead saturation of the airspace resources. Such congestion instances would increase the time-criticality of UAS Traffic Management (UTM) interventions and likely reduce operational efficiency and safety. Therefore, innovative tools and services are needed to deliver Demand and Capacity Balancing (DCB) services in a range of airspace regions, thus increasing operational efficiency and safety while also reducing the time-criticality of UTM operator’s duties. The research presented in this paper aims to develop an efficient and uncertainty-resilient DCB process and solution framework based on hybrid learning algorithms, which allows UTM systems to satisfy the operational requirements of UAS in dense metropolitan regions. The focus of this particular paper is on the analysis of uncertainty factors affecting UAS trajectory conformance in the urban and suburban low-altitude airspace and on the requirements which these factors pose on the determination of recommended DCB processes and techniques. Capitalising on these findings, this research will try to improve the safety, efficiency and uncertainty-resilience of UAS traffic in low-altitude urban airspace operations.

Information fusion as an autonomy enabler for uas traffic management (Utm)

Abstract: Autonomy proliferates air and space traffic management with the National Aeronautics and Space Administration (NASA) initiative on Unmanned Aircraft System Traffic Management (UTM) New endeavors such as electric vertical take-off and landing (eVTOL) and COVID19 are challenging every aspect of the NextGEN rollout Hence, to develop a safe and secure UTM, there is need for knowledge management Knowledge management comes from awareness about the environment and awareness is based on assessment The information fusion community has long developed methods for data fusion (e g , statistical analysis), sensor fusion (e g , navigation and tracking), information fusion (e g , Notice to Airman and air tracks), as well intelligence fusion (e g , response to malicious attacks) for knowledge assessment Each of these techniques has opportunities to enable UTM autonomy, joint all-domain command and control, and surveillance This paper explores the various uses of fusion available to support the autonomy For example, three types of autonomy have been proposed: autonomy at rest (e g , flight plans, radar positions), autonomy in motion (e g , dynamic tracking with automatic dependent surveillance-broadcast (ADS-B) and weather), as well as autonomy in use (e g , getting the right data at the correct time) A use case is presented for UTM that utilizes data fusion from ADS-B, radar, LiDAR, and visual data to provide effective positioning in response to various cyber attacks on the ADS-B data © 2021, American Institute of Aeronautics and Astronautics Inc, AIAA All rights reserved

Online Multimodal Inference of Mental Workload for Cognitive Human Machine Systems

Abstract: With increasingly higher levels of automation in aerospace decision support systems, it is imperative that the human operator maintains the required level of situational awareness in different operational conditions and a central role in the decision-making process. While current aerospace systems and interfaces are limited in their adaptability, a Cognitive Human Machine System (CHMS) aims to perform dynamic, real-time system adaptation by estimating the cognitive states of the human operator. Nevertheless, to reliably drive system adaptation of current and emerging aerospace systems, there is a need to accurately and repeatably estimate cognitive states, particularly for Mental Workload (MWL), in real-time. As part of this study, two sessions were performed during a Multi-Attribute Task Battery (MATB) scenario, including a session for offline calibration and validation and a session for online validation of eleven multimodal inference models of MWL. The multimodal inference model implemented included an Adaptive Neuro Fuzzy Inference System (ANFIS), which was used in different configurations to fuse data from an Electroencephalogram (EEG) model’s output, four eye activity features and a control input feature. The online validation of the ANFIS models produced good results, while the best performing model (containing all four eye activity features and the control input feature) showed an average Mean Absolute Error (MAE) = 0.67 ± 0.18 and Correlation Coefficient (CC) = 0.71 ± 0.15. The remaining six ANFIS models included data from the EEG model’s output, which had an offset discrepancy. This resulted in an equivalent offset for the online multimodal fusion. Nonetheless, the efficacy of these ANFIS models could be confirmed by the pairwise correlation with the task level, where one model demonstrated a CC = 0.77 ± 0.06, which was the highest among all of the ANFIS models tested. Hence, this study demonstrates the suitability for online multimodal fusion of features extracted from EEG signals, eye activity and control inputs to produce an accurate and repeatable inference of MWL.

A Multi-Criteria Clustering Method for UAS Traffic Management and Urban Air Mobility

Abstract: Avionics and Air Traffic Management (ATM) systems are featuring progressively higher levels of automation in order to assist human operators in increasingly more complex operational tasks. This is evident in the context of emerging UAS Traffic Management (UTM) and Urban Air Mobility (UAM) operational paradigms, where a transition from "Human-in-the-Loop" to "Human-on-the-Loop" is required. As the nature of UAS/UAM traffic is inherently heterogeneous and rapidly changing, the resulting levels of complexity are expected to be much higher than in traditional ATM operations. Such complexity can be at least in part addressed by introducing suitable traffic clustering schemes to improve operational efficiency and reduce computational overheads. Traffic clustering algorithms typically adopted in flow management problems utilise aircraft trajectory data as a primary feature variable. However, this approach is not the most adequate for UAS traffic clustering because their mission profiles are affected by various factors, such as weight class, type of mission, and Communication, Navigation and Surveillance (CNS) capabilities. Therefore, a multi-criteria traffic clustering methodology is proposed in this paper, which provides a new, versatile and computationally efficient approach suitable for traffic flow optimisation in next generation DSS for UTM/UAM operations. A preliminary verification of the proposed clustering methodology is performed based on synthetic UAS/UAM traffic data around Melbourne’s inner suburbs. Our preliminary analysis demonstrates that the proposed multi-criteria clustering methodology is numerically feasible and can efficiently support diverse operational UTM/UAM scenarios.

Intelligent Health and Mission Management for Multicopter UAS Integrity Assurance

Abstract: This paper presents a framework for the implementation of an Intelligent Health and Mission Management system within a multicopter Unmanned Aerial System (UAS) architecture. The proposed system comprised of several fault models capable of detecting and predicting faults that affect mission and safety critical subsystems. In doing so, it provides improved integrity assurance and augments health management capabilities that allow real-time avoidance of critical flight conditions and fast recovery, or at the very least, graceful degradation of safety-critical systems with minimal damage to equipment and risk of human injury. In particular, the faults affecting the command and communication link between the UAS and the ground control station, along with their root causes and pathways, were investigated as this is a single point of failure for any mission. Different artificial intelligence techniques were used to diagnose the faults that were simulated in this study. The benefits of the opportune implementation of this technology include increased integrity, reliability and availability of the multicopter UAS.

A Unified Collision Risk Model for Unmanned Aircraft Systems

Abstract: This paper proposes a novel risk management framework, and specifically a methodology to model UAS collision risk. The model inherently accounts for the performance of Communication, Navigation and Surveillance (CNS) systems, as well as aircraft vehicle dynamics in evaluating collision risk. The model is applicable to encounters between two or more aircraft, and to terrain collision scenarios. The methodology is underpinned by modelling of the CNS error characteristics, as well as wind uncertainty, and the translation of those characteristics to the spatial domain to form a virtual risk protection volume around each aircraft. The volume is then inflated in proportion to a Target Level of Safety (TLS). The methodology is demonstrated in a simulation case study representative of aircraft-aircraft collision encounters. A 2σ navigation accuracy bound of 185.2 m was found to incur a maximum collision risk of approximately 1.5×10-4 at a separation of 200m.

Integration of a UAV-LIDAR System for Remote Sensing of CO 2 concentrations in Smart Agriculture

Abstract: Precision agriculture is reliant on making timely, effective measurements to optimize workflow and crop yield. Manual inspection techniques are time consuming and do not support a frequent and autonomous monitoring of the field. This paper examines an innovative application of using a bistatic LIDAR sensor integrated into a UAV to monitor in-field CO 2 concentrations. Measuring variations in atmospheric CO 2 concentrations in agriculture can be used to detect plant health through correlation of photosynthesis efficiency and used to determine soil quality. LIDAR systems can maintain highly accurate measurements, particularly of small changes in CO 2 concentrations, from substantial distances, making it ideal for remote UAV monitoring. Integration of such system into the UAV requires resolving mechanical, electrical, communication computing and control aspects, which are addressed in this paper through a feasibility study of the ground station tracking the system to steer the gimbal, on which the sensor is mounted. There is substantial interest in determining the appropriate UAV platform and consequently the hardware and software architecture and integration into the UAS.

Restructuring Avionics Engineering Curricula to Meet Contemporary Requirements and Future Challenges

Abstract: A comprehensive approach is offered for avionics engineering topics and curricula for educational activities that better meet contemporary aerospace industry requirements. These topics include a growing demand for airspace mobility services, the proliferation of unmanned aircraft systems (UAS), and the required enhancements in avionics and air traffic management (ATM) design standards (hardware and software components); targeting the evolving safety, interoperability, and cybersecurity needs of the aeronautical and space sectors. Proposed topics and approaches stem from the Institute of Electrical and Electronics Engineers (IEEE) Aerospace and Electronics Systems Society (AESS) Avionics Systems Panel (ASP) discussions aimed at aligning educational approaches to relevant industry needs and technical advancements in the field of avionics engineering. Suitable curricular development approaches are proposed to address the career life cycle of avionics engineers, including undergraduate and graduate education. Collaborative exchange and discussion, including both industrial and academic perspectives, focus on informing curricular development approaches that bridge the gaps between higher education, industry practice, and public stakeholder needs toward maximizing educational outcomes and preparedness of the avionics engineering workforce to tackle some of the most important challenges and opportunities faced by the aerospace sector.