Autonomous Vehicles (AV) are expected to bring considerable benefits to society, such as traffic optimization and accidents reduction. They rely heavily on advances in many Artificial Intelligence (AI) approaches and techniques. However, while some researchers in this field believe AI is the core element to enhance safety, others believe AI imposes new challenges to assure the safety of these new AI-based systems and applications. In this non-convergent context, this paper presents a systematic literature review to paint a clear picture of the state of the art of the literature in AI on AV safety. Based on an initial sample of 4870 retrieved papers, 59 studies were selected as the result of the selection criteria detailed in the paper. The shortlisted studies were then mapped into six categories to answer the proposed research questions. An AV system model was proposed and applied to orient the discussions about the SLR findings. As a main result, we have reinforced our preliminary observation about the necessity of considering a serious safety agenda for the future studies on AI-based AV systems.

The objective of this research is to present the state of the art of the safety assurance of Artificial Intelligence (AI)-based systems and guidelines on future correlated work. For this purpose, a Systematic Literature Review comprising 5090 peer-reviewed references relating safety to AI has been carried out, with focus on a 329-reference subset in which the safety assurance of AI-based systems is directly conveyed. From 2016 onwards, the safety assurance of AI-based systems has experienced significant effervescence and leaned towards five main approaches: performing black-box testing, using safety envelopes, designing fail-safe AI, combining white-box analyses with explainable AI, and establishing a safety assurance process throughout systems’ lifecycles. Each of these approaches has been discussed in this paper, along with their features, pros and cons. Finally, guidelines for future research topics have also been presented. They result from an analysis based on both the cross-fertilization among the reviewed references and the authors’ experience with safety and AI. Among 15 research themes, these guidelines reinforce the need for deepening guidelines for the safety assurance of AI-based systems by, e.g., analyzing datasets from a safety perspective, designing explainable AI, setting and justifying AI hyperparameters, and assuring the safety of hardware-implemented AI-based systems.

Currently, the increasing number of daily flights emphasizes the importance of air transportation. Furthermore, Air Traffic Management (ATM) enables air carriers to operate safely and efficiently through the multiple services provided. Advanced analytic solutions have demonstrated the potential to solve complex problems in several domains, and Deep Learning (DL) has attracted attention due to its impressive results and disruptive capabilities. The adoption of DL models in ATM solutions enables new cognitive services that have never been considered before. The main goal of this research is to present a comprehensive review of state-of-the-art Deep Learning (DL) solutions for Air Traffic Management (ATM). This review focuses on describing applications, identifying opportunities, and highlighting open challenges to foster the evolution of ATM systems. To accomplish this, we discuss the fundamental topics of DL and ATM and categorize the contributions based on different approaches. First, works are grouped based on the DL approach adopted. Then, future directions are identified based on the ATM solution area. Finally, open challenges are listed for both DL applications and ATM solutions. This article aims to support the community by identifying research problems to be faced in the future.

Virtual Reality (VR) has played an important role in the development of autonomous robots. From Computer Aided Design (CAD) to simulators for testing automation algorithms without risking expensive equipment, VR has been used in a wide range of applications. Most recently, Autonomous Vehicles (AV), a special application of autonomous robots, became a major focus of the scientific and practitioner community for road and vehicle safety improvements. However, recent AV accidents shed a light on the new safety challenges that need be addressed to fulfill those safety expectations. This paper presents a systematic literature mapping on the use of VR for AV safety and assimilates this literature to create a vision of how VR will play an important role in the development of safety in AV.

Nowadays, there is an increase in the demand for optimized services in urban environments. However, ground transportation in big urban centers has been facing challenges for many years (e.g., resilience and congestion) and new paradigms have been proposed, such as the Urban Air Mobility (UAM) concept. UAM aims to enhance the urban transportation system using manned and unmanned aerial vehicles (i.e., Electric Vertical Takeoff and Landing - eVTOL - vehicles). Although UAM offers many benefits (e.g., cost reduction and increase in transportation capacity), many challenges need to be faced to enable safe and efficient operations. Furthermore, trajectory planning is challenging in the National Airspace System (NAS) and UAM operations due to several factors. Finally, new initiatives concerning UAM trajectory planning can be accelerated with the support of an automatic what-if platform capable of evaluating trajectories feasibility and efficiency. This research aims to propose a simulation platform for enabling trajectory evaluation in UAM operations. This platform, named Trajectory-Based Urban Air Mobility Simulator (TUS), focuses on simulating trajectories in the urban aerial environment. TUS enables users to test new UAM algorithms (e.g., flow management strategies, real-time evaluation of maneuvers effectiveness, airspace configurations) and simulate both manned and unmanned vehicles. Furthermore, TUS operation relies on a set of inputs and evaluates the trajectories generated from efficiency and safety perspectives. This process is performed using a Discrete Event Simulation (DES) approach. The experiments performed showed that TUS can perform a realistic evaluation of UAM trajectories and can be effortlessly used to simulate hundreds of scenarios.

Despite the superior performance in modeling complex patterns to address challenging problems, the black-box nature of Deep Learning (DL) methods impose limitations to their application in real-world critical domains. The lack of a smooth manner for enabling human reasoning about the black-box decisions hinder any preventive action to unexpected events, in which may lead to catastrophic consequences. To tackle the unclearness from black-box models, interpretability became a fundamental requirement in DL-based systems, leveraging trust and knowledge by providing ways to understand the model’s behavior. Although a current hot topic, further advances are still needed to overcome the existing limitations of the current interpretability methods in unsupervised DL-based models for Anomaly Detection (AD). Autoencoders (AE) are the core of unsupervised DL-based for AD applications, achieving best-in-class performance. However, due to their hybrid aspect to obtain the results (by requiring additional calculations out of network), only agnostic interpretable methods can be applied to AE-based AD. These agnostic methods are computationally expensive to process a large number of parameters. In this paper, we present the RXP (Residual eXPlainer), a new interpretability method to deal with the limitations for AE-based AD in large-scale systems. It stands out for its implementation simplicity, low computational cost and deterministic behavior, in which explanations are obtained through the deviation analysis of reconstructed input features. In an experiment using data from a real heavy-haul railway line, the proposed method achieved superior performance compared to SHAP, demonstrating its potential to support decision making in large scale critical systems.