Understanding U-AGREE: EuroUSC Italia Insights on Ground Risks of UAS Operations

Aurora Forcella is a Safety and Compliance Engineer at EuroUSC Italia. In U‑AGREE her work focuses on improving the SORA ground risk model, with particular contributions to the sheltering assessment. She has proposed a probabilistic, physics‑based methodology to evaluate sheltering effectiveness, based on Monte Carlo simulations of building penetration and accounting for aircraft characteristics, impact dynamics, and representative building typologies.

In this interview, she shares her insights on how ground risks are defined, assessed, and integrated within the U-AGREE framework.

How is “ground risk” defined within the context of UAS operations?
In UAS operations, “ground risk” refers to the risk of injury or fatality to third parties on the ground in the event that an unmanned aircraft unintentionally impacts the surface (for example, due to a loss of control, a system failure, or a structural breakup).  
In simple terms, we could say that the concept of “ground risk” can help us answer the question:   

“If the drone crashes, how likely is that people on the ground will be harmed, and how severe could the consequences be?”.  

 Ground risk is influenced both by exposure of people on the ground and by the likelihood of a drone failure. Conceptually, it depends on several factors: 

•  Population density in the operational area  
• Likelihood of losing control of the operation 
• Impact characteristics (such as aircraft Maximum Take Off Mass, Speed, Kinetic Energy, Critical Area) 
• Exposure and sheltering of people on the ground 

 These concepts make clear what is shown in the card: 

• In contexts where there are assemblies of people, who cannot easily escape in the event of a drone crash, the ground risk is higher. This is because the probability of several fatalities or serious injuries is higher, and there is little effective sheltering that could protect people on the ground. 

• In populated areas the ground risk could be considered “medium”. People are more dispersed compared to assemblies, and many are inside buildings, vehicles, or other sheltered structures. This might significantly reduces the probability of a person being directly hit.  These risks are therefore more manageable and could eventually be addressed through appropriate operational and technical mitigations. 

• Finally, rural areas can be classified as “low ground risk areas”, since the population density is very low, and the probability that anyone is within the critical area of a crash is much smaller.  

 In summary, the key takeaway is that the concept of “Ground risk” depends on several aspects. The proposed card clearly illustrates, graphically, how one key aspect (namely the population density and the exposure of people in a given operational area) plays a significant role in defining the overall ground risk level. 

 

What is the current state of the art in assessing ground risk? Please refer to existing methodologies, models, or regulatory approaches currently in use.  
The current state of the art for ground risk assessment in UAS operations is certainly the SORA methodology in its 2.5 version. Published in 2024 by JARUS, this version places a strong emphasis on strengthening the ground risk model by introducing a more quantitative and physics-based approach, compared to the previous version. This evolution is particularly reflected in SORA Annex F, which provides the theoretical and methodological foundation for quantitative ground risk assessment, which introduces structured methods to estimate aspects like population density, or critical area, enabling a more consistent and defensible evaluation of ground risk. The SORA 2.5 methodology has been adopted at European level as part of the EASA regulatory framework, and Member States are progressively transitioning from SORA 2.0 (largely based on qualitative ground risk assessments) to SORA 2.5. As a result, the regulatory ecosystem is currently in a transition phase, during which both approaches may still be encountered in practice.

 

Which factors or variables are considered most critical in evaluating ground risk?  
Ground risk is primarily driven by a combination of exposure, impact severity, and environmental context. These factors can be grouped into the following core variables: 

• Population density, which directly influences the probability that people are present within the area potentially affected by a crash. 
• Likelihood of loss-of-control events, which is mainly related to UAS design characteristics, system reliability, and operational robustness.  
• The size of the critical area, which represents the ground surface over which a person could be lethally injured following a loss‑of‑control event. It depends mainly on the UAS impact dynamics and post-impact behavior. 
• The impact of kinetic energy, whose value largely determines the potential lethality of a crash. 
• The sheltering effectiveness, which depends on building typologies and material properties, as well as the probability of impacting roofs, walls, or windows. 
• The obstacle density and urban morphology, since the presence, density and geometry of obstacles influence the effective size of the critical area. 

Moreover, the concept of ground risk may also be extended beyond immediate injuries, to include topics such as environmental, economic, and reputational impacts, which can significantly affect the overall severity and societal consequences of UAS ground impact events

 

How does the U-AGREE project approach and implement these risks? 
The U-AGREE project implements the ground risk model by extending the SORA v2.5 framework specifically on three topics: 

• By refining the treatment of sheltering, with particular attention to sheltering effectiveness assessment for UAS with MTOM above 25 kg 
• By evaluating obstacle effect and the effective critical area reduction in two scenarios with different buildings density, explicitly distinguishing between rotorcraft and fixed-wing behaviour 
• By extending the concept of ground risk beyond direct fatalities, introducing a structured severity assessment of environmental, economic, and reputational impacts.  

 

What results or preliminary findings have U-AGREE achieved so far regarding privacy and security risks? 
The U-AGREE Project has produced concrete results in the field of ground risk assessment, with its most significant findings related to sheltering effectiveness and obstacle effects in urban environments. These results are based on physics‑based modelling and Monte Carlo simulations, and represent a refinement of the current SORA v2.5 ground risk framework. Specifically, U-AGREE developed a probabilistic Monte Carlo-based framework to estimate the likelihood that a UAS impact results in building penetration, explicitly accounting for uncertainties in UAS characteristics, impact dynamics, and impacted building material properties. 
The model evaluates penetration probability distributions across multiple simulated impact scenarios. One of the most relevant findings for typical Italian residential buildings is that penetration likelihood is governed by the most fragile components, most notably windows. 
Another contribution of U-AGREE concerns the quantification of obstacle effects; that is, the extent to which buildings and structures reduce the effective critical area following a ground impact. Here too, U-AGREE adopted a Monte Carlo–based approach, generating thousands of random crash scenarios in synthetic urban and metropolitan environments. To make the analysis more consistent, a distinction was introduced between fixed-wing and rotary-wing UAS, in order to account for their different impact dynamics and post-impact behavior, which vary considerably between the two categories.  
Implementing the UAS–buildings crash dynamics led U-AGREE to also develop a novel method for correlating population density with building density in specific urban contexts. The proposed approach is fully consistent with the definitions established by JARUS, ensuring alignment with the current regulatory framework, while at the same time offering a perspective that could prove valuable for future refinements of the SORA methodology. 

Overall, U‑AGREE has demonstrated—through large‑scale Monte Carlo simulations—that both sheltering effectiveness and obstacle effects can be quantified in a robust and uncertainty‑aware manner, providing more realistic and defensible evidence to support ground risk assessment in complex urban contexts.