Collision hazard modeling and analysis in a multi-mobile robots system transportation task with STPA and SPN

Abstract

This paper investigates the ability to use complex multi-mobile robotic systems in risky and dynamic environments, such as industrial plants and laboratories, with the presence of the human factor. More specifically, it presents an approach to analyze the dominant risk, extract, model and quantify the hazard scenarios, then propose requirements using the combination between two methods: System Theoretic Process Analysis (STPA) and Stochastic Petri Nets (SPN). This approach is demonstrated with a case study related to a chemical transportation task within a miniature analysis laboratory in oil and gas industry. The main purposes of this article are: to investigate how the risk and safety of these systems should be managed, to create a framework for modeling collision hazard scenarios, further a Monte Carlo simulation is performed to quantify the collision frequency and unavailability. In addition, to generate the required constraints and requirements in order to improve the safe operation of robots within the laboratory. The novel contribution of this study is to provide a hazard assessment approach suited to multi-controller and autonomous systems while considering coordination between controllers. Hence, the current study proposes a set of safety requirements, which allow designers and programmers to improve autonomy features in mobile robots. The STPA-SPN combination offers better modeling and assessment of robot's performance as well as their hazard frequency.