Scribd
Upload
21 views2 pages

A Dijkstra Algorithm For Fixed-Wing UAV Motion Planning Based On Terrain Elevation

This document presents a Dijkstra algorithm tailored for fixed-wing UAV trajectory planning, utilizing terrain elevation data represented by digital elevation models. The proposed Elevation-based Dijkstra Algorithm (EDA) improves upon existing methods by efficiently planning dynamically feasible, collision-free trajectories through visibility graphs. Additionally, a new grid-based local search heuristic is introduced for collision verification, significantly reducing computational time in complex navigation environments.

Uploaded by

nabin.bhandari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
21 views2 pages

A Dijkstra Algorithm For Fixed-Wing UAV Motion Planning Based On Terrain Elevation

This document presents a Dijkstra algorithm tailored for fixed-wing UAV trajectory planning, utilizing terrain elevation data represented by digital elevation models. The proposed Elevation-based Dijkstra Algorithm (EDA) improves upon existing methods by efficiently planning dynamically feasible, collision-free trajectories through visibility graphs. Additionally, a new grid-based local search heuristic is introduced for collision verification, significantly reducing computational time in complex navigation environments.

Uploaded by

nabin.bhandari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

A Dijkstra Algorithm for Fixed-Wing UAV Motion

Planning Based on Terrain Elevation

Felipe Leonardo Lôbo Medeiros and José Demisio Simões da Silva

Instituto de Estudos Avançados, Rod. dos Tamoios, km 5,5, Putim, CEP 12.228-001,
São José dos Campos, São Paulo, Brasil
[email protected]
Instituto Nacional de Pesquisas Espaciais, Av. dos Astronautas 1.758, Jardim da Granja,
CEP 12227-010, São José dos Campos, São Paulo, Brasil
[email protected]

Abstract. Abstract. The automatic motion or trajectory planning is essential for


several tasks that lead to the autonomy increase of Unmanned Aerial Vehicles
(UAVs). This work proposes a Dijkstra algorithm for fixed-wing UAVs trajec-
tory planning. The navigation environments are represented by sets of visibility
graphs constructed through the terrain elevations of these environments. Digital
elevation models are used to represent the terrain elevations. A heuristics to ver-
ify if a trajectory is collision-free is also proposed in this work. This heuristics is
a method of grid-based local search which presents linear computational time
O(np), where np is the number of verification steps. This heuristics is compared
with another method for collision verification. Results are presented in this work.

Keywords: fixed-wing UAV; motion planning; Dijkstra algorithm; digital ele-


vation model; grid-based local search.

1 Introduction
The automatic motion or trajectory planning is essential for several tasks that lead to
the autonomy increase of Unmanned Aerial Vehicles (UAVs). This work approaches
the problem of trajectory planning for fixed-wing UAVs navigating to constant alti-
tudes. This problem is similar to the trajectory planning for nonholonomic wheeled
mobile robots [1]. A trajectory is a sequence of waypoints connected by straight line
segments and arcs that allow to the fixed-wing UAV navigates between an initial and
a final waypoint of the navigation environment. A trajectory must be collision-free
and dynamically feasible [1], i. e., a trajectory must allow the UAV navigate from a
waypoint to the next waypoint without violating the UAV dynamics and kinematic
constraints. A formal definition of dynamically feasible trajectory is presented in [1].
Fig. 1 presents an example of a dynamically feasible collision-free trajectory planned
for a fixed-wing UAV.
A recent revision of UAV motion planning methods is presented in [2]. As men-
tioned in this revision, Dijkstra algorithms assure optimality when used to plan navi-
gation paths, which are sequences of waypoints connected by straight line segments.
However, the paths planned by these algorithms must be transformed in trajectories

A.C. da Rocha Costa, R.M. Vicari, F. Tonidandel (Eds.): SBIA 2010, LNAI 6404, pp. 213–222, 2010.
© Springer-Verlag Berlin Heidelberg 2010
214 F.L.L. Medeiros and J.D.S. da Silva

through the application of smoothing methods. A Dijkstra algorithm is proposed in [3]


to solve the single-destination UAV shortest trajectory problem based on visibility
graphs, without the necessity of application of smoothing methods. This algorithm
is a modification of the original Dijkstra algorithm [4] used for the single-source
shortest path problem. The Modified Dijkstra Algorithm (MDA) uses rectangular
representations of the non-navigable regions of the navigation environment to create
the collision-free nodes and edges of the visibility graphs. The rectangular representa-
tion is also used by a method to check a trajectory is collision-free. In the worst case,
the computational time of this method is O(no), where no is the number of obstacles or
non-navigation regions of the navigation environment. Due to this computational
time, the MDA is not efficient for navigation environments with a large number of
obstacles as, for example, the navigation environments with surfaces represented by
digital elevation models [5]. A digital elevation model is a computational representa-
tion of the terrain elevation of a navigation environment.
Therefore, this work proposes an Elevation-based Dijkstra Algorithm (EDA) that is
an adaptation of the MDA [3] aiming at the planning of dynamically feasible colli-
sion-free trajectories through visibility graphs and digital elevation models.

Fig. 1. Example of a planned trajectory. The planned trajectory is represented by circles,


straight line segments and arcs. The circles indicate the waypoints of the trajectory. The black
polygons represent the non-navigable regions of the navigation environment. The stippled gray
circles indicate the circles defined by the UAV turning radius.

EDA presents three main distinctions regarding the MDA. The first distinction is
the elaboration of the proposed algorithm to solve the single-pair shortest trajectory
problem. As second distinction, EDA plans trajectories of the class k-trajectories
proposed in [1]. A k-trajectory is a Dubins curve whose length and form can be ad-
justed by the parameter k. The last distinction is that this work also proposes a grid-
based local search heuristics to verify if a trajectory is collision-free. This heuristics is
invariant to the number of obstacles and presents a computational time of O(np), in the
worst case, where np is the number of verifications. Considering navigation environ-
ments with a large number of obstacles, in comparison with the method used in the
modified Dijkstra algorithm, this heuristics allows a significant decrease of the com-
putational time, becoming the main contribution of this work.
The remainder of this paper is organized as follows. Section 2 presents the con-
struction of visibility graphs which represent the navigation environment used in this
work. Section 3 presents the Dijkstra algorithm proposed in this work and presents the
results of the application of this algorithm. Final conclusions based on the obtained
results are presented in Section 4.

You might also like