Risk-Based Path Planning Optimization Methods for UAVs Over Inhabited Areas

Abstract

Operating unmanned aerial vehicles (UAVs) over inhabited areas requires mitigating the risk to persons on the ground. Because the risk depends upon the flight path, UAV operators need approaches (techniques) that can find low-risk flight paths between the mission’s start and finish points. In some areas, the flight paths with the lowest risk are excessively long and indirect because the least-populated areas are too remote. Thus, UAV operators are concerned about the tradeoff between risk and flight time. Although there exist approaches for assessing the risks associated with UAV operations, existing risk-based path planning approaches have considered other risk measures (besides the risk to persons on the ground) or simplified the risk assessment calculation. This paper presents a risk assessment technique and bi-objective optimization methods to find low-risk and time (flight path) solutions and computational experiments to evaluate the relative performance of the methods (their computation time and solution quality). The methods were a network optimization approach that constructed a graph for the problem and used that to generate initial solutions that were then improved by a local approach and a greedy approach and a fourth method that did not use the network solutions. The approaches that improved the solutions generated by the network optimization step performed better than the optimization approach that did not use the network solutions.

Feasibility Analysis of UAV Operations for Monitoring of Infrastructure Networks: a Risk-based Approach

“…On the contrary if Σ is 'too high', it will be considered that flying over the corresponding location is not feasible without requiring major changes in the flight trajectory, i.e. at least deviation outside the ground impact footprint, which is beyond the scope of this paper (see risk aware path planning methods for example [14] [15]). Therefore it is of interest to define some threshold δ Σ for the surface Σ.…”

Multi-Objective Design and Path Planning Optimization of Unmanned Aerial Vehicles (UAVs)

“…The time objective was based off the assumption that speed of the UAV would remain constant throughout the flight, thus defining it as simply being the length of the path divided by the speed of the vehicle. The risk objective was defined using the risk metric defined in [3]. The crash location probability distribution was parametrized in terms of the design variables under considerations, which was used to optimize the risk objective.…”

“…The crash location probability distribution was parametrized in terms of the design variables under considerations, which was used to optimize the risk objective. The crash location distributions used in the model were generated by running Monte Carlo simulations of an unpowered UAV crashing such as the ones done in [3]. Multiple distributions were generated for different design variable combinations, the parametrization used these distributions to estimate the distribution for arbitrary design variable combinations within the bounds of the design variables using Delaunay triangulation.…”

“…Multiple distributions were generated for different design variable combinations, the parametrization used these distributions to estimate the distribution for arbitrary design variable combinations within the bounds of the design variables using Delaunay triangulation. For the exact formulae for the objectives used see [3].…”

“…Narayan et al [2] considered a trajectory optimization approach for the problem of optimizing multiple objectives using weighted sum methods and dynamic programming. Rudnick-Cohen et al [3] compared multiple methods for risk-based path planning for a given UAV. They used Monte Carlo simulations of a flight dynamics model to determine a crash location distribution, which was used to estimate the expected fatalities of a flight over inhabited areas.…”