Urban Metric Maps for Small Unmanned Aircraft Systems Motion Planning

TL;DR

This paper introduces a set of motion planning metrics tailored for small UAS urban flight, evaluating planner performance and map properties in complex city environments using Monte Carlo simulations.

Contribution

It presents new map-based and path-based metrics for assessing motion plan quality in urban UAS flight planning, including a novel multi-objective heuristic for various planners.

Findings

Metrics effectively characterize motion plan quality.

Planner performance varies with location, range, and altitude.

Urban environment impacts planning outcomes.

Abstract

Low-altitude urban flight planning for small Unmanned Aircraft Systems (UAS) requires accurate vehicle, environment maps, and risk models to assure flight plans consider the urban landscape as well as airspace constraints. This paper presents a suite of motion planning metrics designed for small UAS urban flight. We define map-based and path-based metrics to holistically characterize motion plan quality. Proposed metrics are examined in the context of representative geometric, graph-based, and sampling-based motion planners applied to a multicopter small UAS. A novel multi-objective heuristic is proposed and applied for graph-based and sampling motion planners at four urban UAS flight altitude layers. Monte Carlo case studies in a New York City urban environment illustrate metric map properties and planner performance. Motion plans are evaluated as a function of planning algorithm, location, range, and flight altitude.