A Socio-technical Model for Analyzing Safety Risk of Unmanned Aircraft Systems (UAS): An Application to Precision Agriculture☆

Abstract There are a number of concerns about the introduction of Unmanned Aircraft Systems (UAS) into the National Airspace System (NAS) of which safety risk is of paramount importance. For UAS that typically fly low and slow, the possibility of a mid-air collision with a nearby general aviation aircraft needs to be studied by identifying possible hazards and assessing mitigations. The Aviation System Risk Model (ASRM) is a first-generation socio-technical model that uses a Bayesian Belief Network (BBN) methodology to integrate possible hazards to assess a non-linear safety risk metric. The ASRM may be used to evaluate underlying causal factors linked to the vehicle and/or to the systems and procedures that led to the unsafe state and the interactions among these factors that contributed to the safety risk. The ASRM can also assess the projected impact of mitigations. The ASRM facilitates robust inductive reasoning on the hypothesized accident scenarios, ideal for addressing emergent NAS operations where there may be obvious data and experience limitations. Recently, the ASRM has been updated with the use of the Hazard Classification and Analysis System (HCAS) that provides analytic structure for categorizing hazards related to the UAS, Airmen, Operations and the Environment. In this paper, the ASRM, together with the HCAS, is demonstrated with a precision agriculture application of a notional scenario that involves an UAS being used for crop scouting. It is conjectured that the UAS interacts with a neighboring farm where a conventional piloted cropduster is being used for pesticide spraying. The UAS being used is a rotorcraft-type where there is a failure of the separation assurance function since the UAS leaves its Area of Operation (AO) due to a Ground Control Station (GCS) transmission disruption (from faulty maintenance) and by the waypoints being incorrectly programmed. The ASRM is used to analyze such a scenario leading to a collision volume being entered by both the UAS and the piloted cropduster that possibly leads to a mid-air collision. The efficacy of a geofence or a “virtual barrier” mitigation for the UAS is also analyzed. The ASRM safety risk results for this notional scenario are presented and interpreted.