Modeling the motion of small unmanned aerial system (sUAS) due to ground collision

Abstract

The aim of this paper is to develop a modeling method in order to predict the rebounce distance of a small size unmanned aerial system (sUAS) after collision with the ground. This distance would be useful to determine the safe range on the ground when operating the UAS. A two-step strategy is developed to model this procedure. In step one, numerical simulations are performed to model the first impact and predict the rebounce angle and impact velocity. Based on the information obtained, in step two, an analytical model is employed to calculate the rebounce distance without running numerical simulations, to save the computational time. Based on the model predictions, a non-linear function (known as a Meta model) is established to describe the rebounce behavior as a function of impact parameters, i.e. impact velocity, angle, and friction coefficient between the vehicle and ground. A correlation analysis is further performed to analyze the effect of these parameters on the response. The results show that the maximum rebounce range could be greater than 10 m. Compared to impact angle and friction coefficient, impact velocity is the most significant factor influencing the rebounce range. The larger velocity, smaller impact angle, and smaller ground friction coefficient would lead to the large rebounce range. The established Meta model would be useful to estimate safety range when operating sUAS platforms.

Application of Dempster–Shafer Networks to a Real-Time Unmanned Systems Risk Analysis Framework

“…While the previous research focused on population and crowd modeling while including basic dynamics of the vehicle upon impact, other research has been done to model the dynamics of the vehicle including bounce effects to better understand the potentially affected area [56].…”

Análisis aerodinámico de un vehículo aéreo no tripulado con forma de halcón para monitoreo de fugas de hidrocarburos

“…The aerodynamic analysis of the UAV is obtained using a computational fluid dynamic (CFD) through the CFX module of ANSYS software. Other authors have used the ANSYS software to predict aerodynamics coefficients of UAVs with satisfactory results [27], [28]. For this aerodynamic analysis, we built a control volume to simulate the air around UAV (Figure 4).…”

Understanding the Effect of Hyperparameter Optimization on Machine Learning Models for Structure Design Problems

“…In the development of complex structures, such as a vehicle or aircraft, a large number of full-scale numerical simulations are often needed. As the supplement and partial substitute of these expensive simulations, surrogate models have been widely used in engineering design and optimization to reduce the computational cost [1][2][3]. A surrogate model is established from the design or simulation datasets through regression, to approximate the real model.…”