This paper offers a formal framework for the rare collision risk estimation of autonomous vehicles (AVs) with multi-agent situation awareness, affected by different sources of noise in a complex dynamic environment. The estimation framework consists of two complementary parts: modeling formalism and a rare event estimation method using sequential Monte Carlo (MC) simulation instead of importance sampling. By defining incremental levels of severity that must be passed before a collision, a sequence of MC simulations can be applied from one level to the next. This particular sequential MC method consists of the simulation of an Interacting Particle System (IPS) in combination with Fixed Assignment Splitting (FAS) of particles that reach the next level. We model AVs equipped with the situation awareness as general stochastic hybrid systems (GSHS), including the IPS-FAS relevant severity levels, and assess the probability of collision in a lane-change scenario where two self-driving vehicles simultaneously intend to switch lanes into a shared one while utilizing the time-tocollision measure for decision-making as required. The IPS-FAS method is subsequently used to estimate collision risk for this GSHS model of the lane-changing scenario. The results show that in contrast to straightforward MC simulation, IPS-FAS is able to quantify the very low collision risk for the scenario of interest. @en

Advantages of commercial UAS-based services come with the disadvantage of posing third party risk (TPR) to overflown population on the ground. Especially challenging is that the imposed level of ground TPR tends to increase linearly with the density of potential customers of UAS services. This challenge asks for the development of complementary directions in reducing ground TPR. The first direction is to reduce the rate of a UAS crash to the ground. The second direction is to reduce overflying in more densely populated areas by developing risk-aware UAS path planning strategies. The third direction is to develop UAS designs that reduce the product (Formula presented.) in case of a crashing UAS, where (Formula presented.) is the size of the crash impact area on the ground, and (Formula presented.) is the probability of fatality for a person in the crash impact area. Because small UAS accident and incident data are scarce, each of these three developments is in need of predictive models regarding their contribution to ground TPR. Such models have been well developed for UAS crash event rate and risk-aware UAS path planning. The objective of this article is to develop an improved model and assessment method for the product (Formula presented.) In literature, the model development and assessment of the latter two terms is accomplished along separate routes. The objective of this article is to develop an integrated approach. The first step is the development of an integrated model for the product (Formula presented.). The second step is to show that this integrated model can be assessed by conducting dynamical simulations of Finite Element (FE) or Multi-Body System (MBS) models of collision between a UAS and a human body. Application of this novel method is illustrated and compared to existing methods for a DJI Phantom III UAS crashing to the ground.

@en

The recent surge in space launch activities, driven by the emergence of commercial space launches, has compelled the aviation and space launch sectors to collaborate for the safe and efficient integration of space launch activities. This paper introduces an agent-based modeling (ABM) and simulation framework designed to assess the impact of spacecraft launches on air traffic within an integrated air and space traffic management system. The proposed framework incorporates various agents involved in the execution phase of space launches and considers the interactions and coordination between air traffic management and space traffic management. The paper firstly provides a comprehensive overview of the current state of space launch operations and their effects. Then, a general agent-based model is developed for space launch execution phase in order to gain an understanding of various entities involved in a space launch activity as well as the interactions among these entities. Using Monte-Carlo simulations based on the ABM, the paper assesses the impact on air traffic operations in the event of a space launch failure. In each simulation, various factors are taken into account, including launch site position, launch slot, failure probability during the execution phase, debris dispersion, and time delay in Air Traffic Management (ATM)/Space Traffic Management (STM) coordination. To demonstrate the practical application of the proposed framework in an operational context, the paper presents a case study of a sea-based space launch in the Singapore FIR. The paper makes a valuable contribution to the field of air and space traffic management by addressing the need for innovative strategies to ensure the safe sharing ofairspace among different stakeholders. @en

The transition kernel of an ℝ ⁿ-valued diffusion or jump diffusion process {X _t} is known to satisfy the Feller property if {X _t} is the solution of an SDE whose coefficients are Lipschitz continuous. This Lipschitz route to Feller falls short if {X _t} is the solution of an SDE whose coefficients depend on a state-dependent regime-switching process {θ _t}. In this paper it is shown that pathwise uniqueness and the Feller property are satisfied under mild conditions for a regime-switching jump diffusion process {X _t, θ _t} with hybrid jumps, i.e. jumps in {X _t} that occur simultaneously with {θ _t} switching.

@en

For diffusions, a well-developed approach in rare event estimation is to introduce a suitable factorization of the reach probability and then to estimate these factors through simulation of an Interacting Particle System (IPS). This paper studies IPS based reach probability estimation for General Stochastic Hybrid Systems (GSHS). The continuous-time executions of a GSHS evolve in a hybrid state space under influence of combinations of diffusions, spontaneous jumps and forced jumps. In applying IPS to a GSHS, simulation of the GSHS execution plays a central role. From literature, two basic approaches in simulating GSHS execution are known. One approach is direct simulation of a GSHS execution. An alternative is to first transform the spontaneous jumps of a GSHS to forced transitions, and then to simulate executions of this transformed version. This paper will show that the latter transformation yields an extra Markov state component that should be treated as being unobservable for the IPS process. To formally make this state component unobservable for IPS, this paper also develops an enriched GSHS transformation prior to transforming spontaneous jumps to forced jumps. The expected improvements in IPS reach probability estimation are also illustrated through simulation results for a simple GSHS example.

@en

The increasing use of on-board sensor monitoring and data-driven algorithms has stimulated the recent shift to data-driven predictive maintenance for aircraft. This paper discusses emerging challenges for data-driven predictive aircraft maintenance. We identify new hazards associated with the introduction of data-driven technologies into aircraft maintenance using a structured brainstorming conducted with a panel of maintenance experts. This brainstorming is facilitated by a prior modeling of the aircraft maintenance process as an agent-based model. As a result, we identify 20 hazards associated with data-driven predictive aircraft maintenance. We validate these hazards in the context of maintenance-related aircraft incidents that occurred between 2008 and 2013. Based on our findings, the main challenges identified for data-driven predictive maintenance are: (i) improving the reliability of the condition monitoring systems and diagnostics/prognostics algorithms, (ii) ensuring timely and accurate communication between the agents, and (iii) building the stakeholders’ trust in the new data-driven technologies.
@en

Abstract This report is concerned with a friendly competition for formal verification and policy synthesis of stochastic models. The main goal of the report is to introduce new benchmarks and their properties within this category and recommend next steps toward next year’s edition of the competition. Given that the tools for stochastic models are at their early stages of development compared to those of non-probabilistic models, the main focus is to report on an initiative to collect a set of minimal benchmarks that all such tools can run, thus facilitating the comparison between the efficiency of the implemented techniques. This friendly competition took place as part of the workshop Applied Verification for Continuous and Hybrid Systems (ARCH) in Summer 2023.@en

This report presents the results of a friendly competition for formal verification and policy synthesis of stochastic models. It also introduces new benchmarks and their properties within this category and recommends next steps for this category towards next year’s edition of the competition. In comparison with tools on non-probabilistic models, the tools for stochastic models are at the early stages of development that do not allow full competition on a standard set of benchmarks. We report on an initiative to collect a set of minimal benchmarks that all such tools can run, thus facilitating the comparison between efficiency of the implemented techniques. The friendly competition took place as part of the workshop Applied Verification for Continuous and Hybrid Systems (ARCH) in Summer 2022.@en

UAS-based commercial services such as urban parcel delivery are expected to grow in the upcoming years and may lead to a large volume of UAS operations in urban areas. These flights may pose safety risks to persons and property on the ground, which are referred to as third-party risks. Path-planning methods have been developed to generate a nominal flight path for each UAS flight that poses relative low third-party risks by passing over less risky areas, e.g., areas with low-density unsheltered populations. However, it is not clear if risk minimization per flight works well in a commercial UAS operation that involves a large number of annual flights in an urban area. Recently, it has been shown that when using shortest flight path planning, a UAS-based parcel delivery service in an urban area can lead to society-critical third-party risk levels. The aim of this paper is to evaluate the mitigating effect of state-of-the-art risk-aware path planning on these society-critical third-party risk levels. To accomplish this, a third-party risk simulation using the shortest paths is extended with a state-of-the-art risk-aware path-planning method, and the societal effects on third-party risk levels have been assessed and compared to those obtained using shortest paths. The results show that state-of-the-art risk-aware path planning can reduce the total number of fatalities in an area, but at the cost of a critical increase in safety risks for persons living in areas that are favored by a state-of-the-art risk-aware path-planning method.

@en

This paper focuses on estimating reach probability of a closed unsafe set by a stochastic process. A well-developed approach is to make use of multi-level MC simulation, which consists of encapsulating the unsafe set by a sequence of increasing closed sets and conducting a sequence of MC simulations to estimate the reach probability of each inner set from the previous set. An essential step is to copy (split) particles that have reached the next level (inner set) prior to conducting a MC simulation to the next level. The aim of this paper is to prove that the variance of the multi-level MC estimated reach probability under fixed assignment splitting is smaller or equal than under random assignment splitting methods. The approaches are illustrated for a geometric Brownian motion example.

@en

Commercial aviation distinguishes three indicators for third party risk (TPR): i) Expected number of ground fatalities per aircraft flight hour; ii) Individual risk; and iii) Societal risk. The latter two indicators stem from TPR posed to population by operation of hazardous installations. Literature on TPR of Unmanned Aircraft System (UAS) operations have focused on the development of the first TPR indicator. However the expected increase of commercial UAS operations requires an improved understanding of third party risk (TPR). To support such improvement, this paper extends the existing TPR model for UAS operations with societal and individual risk indicators. The extension is developed both at modelling level and at assessment level. Subsequently the extended approach is applied to a hypothetical UAS based parcel delivery service in the city of Delft. The results obtained for the novel UAS TPR indicators show that this aligns commercial UAS operations with land use policies and standing TPR regulation for airports and hazardous facilities.

@en

The unique capabilities of an Unmanned Aircraft System (UAS) creates opportunities for commercial services. The key question is what is an acceptable level of risk posed to third parties on the ground that have no direct benefit from commercial UAS flights. In literature the common view is that an acceptable level of Third Party Risk (TPR) posed by UAS operations follows from an Equivalent Level Of Safety (ELOS) criterion, which means that per flight hour a UAS should not pose more safety risk to persons on the ground than a commercial aircraft does. However in commercial aviation there are also TPR indicators in use that are directed to accident risk posed by all annual commercial flights to the population around an airport. These population directed indicators find their origin in TPR posed by hazardous installations to its environment. The aim of this paper is to improve the understanding of risk posed to the population by annual UAS-based services through learning from TPR knowledge and regulation for airports and hazardous installations. As main result this paper develops an analytical approach to evaluate the annual risk posed by a commercial UAS-based parcel delivery service in urban and metropolitan areas. The obtained results show that the TPR indicators that stem from hazardous installations and airports provide novel insight regarding TPR of commercial UAS-based service. @en

This report presents the results of a friendly competition for formal verification and policy synthesis of stochastic models. It also introduces new benchmarks within this category, and recommends next steps for this category towards next year's edition of the competition. The friendly competition took place as part of the workshop Applied Verification for Continuous and Hybrid Systems (ARCH) in Spring/Summer 2021.@en

Motivated by the need to understand and further optimize AOC decision making processes under uncertainty, this paper implements and evaluates the effects of operational uncertainties using Agent-Based Modelling and Simulation. The specific application concerns a challenging scenario composed of two consecutive disruptions. To evaluate the effects of uncertainties, an agent-based model of AOC processes has been developed using a logic-based ontology. Subsequently, this agent-based model is used to analyze the sensitivities of different model parameters. The simulation results provide novel insights into the effects of operational uncertainties on AOC decision-making and consequently airline performance. For the aircraft breakdown scenario considered, it is shown that adding buffers into the schedule promote a degree of self-recovery. The sensitivity analysis also reveals that transit buffer time and crew duty slack time act as tipping points for the airline operating costs. This demonstrates that ABMS allows to analyze and bring into light various sensitivities, which can be used in the early design phase to increase airline resilience, and train airline controllers for different environment states. The paper concludes that ABMS is a valuable approach that can enable a paradigm shift from reactive recovery to proactive recovery.@en

Recent developments in the concept of UAS operations in urban areas have led to risk concerns of UAS collision with human. To better understand this risk, head and neck injuries due to UAS collisions have been investigated by different research teams using crash dummies. Because of the limitations in biofidelity of a crash dummy, head injury level for a crash dummy impact may differ from the human body impact. Therefore, the aim of this paper is to investigate differences in head and neck injuries subject to UAS collision between an often-used Hybrid III crash dummy and a human body. To perform such investigation, multibody system (MBS) impact models have been used to simulate UAS impacts on validated models of the Hybrid III crash dummy and the human body at various impact conditions. The findings show that the Hybrid III predicts similar head and neck injury compared to the human body when UAS collides horizontally from front and rear. However, the Hybrid III over-predicts head injury due to horizontal side impact. Moreover, under vertical drop and 45 degree elevated impact of UAS, the Hybrid III under-predicts head injury, and over-predicts neck injury.

@en

This report presents the results of a friendly competition for formal verification and policy synthesis of stochastic models. It also introduces new benchmarks within this category, and recommends next steps for this category towards next year's edition of the competition. The friendly competition took place as part of the workshop Applied Verification for Continuous and Hybrid Systems (ARCH) in Spring/Summer 2020.@en

Evaluating safety risk posed to third parties on the ground due to UAS impact requires a model of probability of fatality (PoF) for human. For quadrotor UAS, the existing impact models predict remarkably different PoFs. The most pessimistic is the impact model adopted by Range Commanders Council (RCC) while the Blunt Criterion model is far more optimistic. The ASSURE study has assessed the third set of PoF values through conducting controlled drop tests of a DJI Phantom III on a crash dummy; these results differ again. To investigate these discrepancies, this paper employs a numerical impact analysis of UAS collisions on humans. The current paper is the third in a series of studies. The first study developed a MultiBody System (MBS) simulation model of a DJI Phantom III impacting the head of a crash dummy; this MBS model has been validated against the experimental drop test results of ASSURE. The second study conducted simulations with the validated MBS model to systematically show the differences in head and neck injuries if the human dummy is replaced by a validated MBS model of a human body. The aim of the current paper is threefold: i) to extend the latter MBS model to assess injury levels for DJI Phantom III impact on thorax and abdomen; ii) to transform the assessed injury levels for head, thorax and abdomen to PoFs; and iii) to compare the MBS obtained PoFs to those from RCC and Blunt Criteria models. The MBS based results show that variations in the scenario of DJI Phantom III impact on the head significantly affect PoF. These variations are not captured by the RCC or BC model, and neither in the ASSURE measurements. Both for head, thorax and abdomen, in case of comparable impact scenarios, the RCC model tends to over-predicts PoF compared to the MBS model, while the BC model tends to under-predict PoF.

@en

Use of Unmanned Aircraft Systems (UAS) is growing rapidly around the world. Very different types of UAS are used for applications such as aerial photography, inspection, emergency and Urban Air Mobility (UAM), operating in low altitude and urban environment, as well as in high altitude airspace integrated with the conventional air transportation system. As a new airspace user, UAS brings novel safety challenges to the current aviation system. For current aviation the main safety issues concern first and second parties, i.e. lives and property of crew and passengers. In contrast, the main safety concern of UAS operations is third party risk (TPR), i.e. the risk posed to people and properties that have no responsibility for the UAS operation and neither benefit in some way from the UAS operation. In order to ensure the safe operation of UAS, there is a need for an evaluation of safety regulation developments for UAS operations against relevant TPR indicators. The aim of this paper is to identify relevant TPR indicators for UAS operations and to evaluate safety regulations against these TPR indicators. The main finding is that current UAS safety regulations do not consider the accumulation of TPR contributions from many UAS flights per annum over rural or urban populations.@en

Airborne collision avoidance systems (ACASs) form a key safety barrier by providing last-moment resolution advisories (RAs) to pilots for avoiding midair collisions. Intrinsic uncertainties, such as noise in ACAS input signals and variability in pilot performance, imply that the generation of RAs and the effectuated aircraft trajectories are nondeterministic processes. Existing ACAS validation methods reflect the intrinsic uncertainties to a limited extent only. This paper develops an agent-based model, which systematically captures uncertainties in ACAS input and pilot performance for Monte Carlo (MC) simulation of encounter scenarios. The agent-based model has been integrated with industry-specific implementations of Traffic Alert and Collision Avoidance System II and ACAS Xa in a novel collision avoidance validation and evaluation tool. Through illustrative MC simulation results, it is demonstrated that the intrinsic uncertainties can have a significant effect on the variability in timing and types of RAs, and subsequently on the variability in miss distance. Even the MC simulation estimated mean miss distance can differ significantly from the deterministically simulated miss distance. Most important, the tails of miss distance probability distributions and probabilities of near-midair collisions are affected. This stipulates that addressing intrinsic uncertainties through MC simulation is essential in evaluating ACASs.

@en