Computers and Electrical Engineering 80 (2019) 106493
Contents lists available at ScienceDirect
Computers and Electrical Engineering
journal homepage: www.elsevier.com/locate/compeleceng
Evolution-algorithm-base d unmanne d aerial vehicles path
planning in complex environment
✩
Xiaolei Liu
a , ∗
, Xiaojiang Du
b
, Xiaosong Zhang
c
, Qingxin Zhu
a
, Mohsen Guizani
d
a
School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, China
b
Department of Computer and Information Sciences, Temple University, Philadelphia, PA, USA
c
Center for Cyber Security, University of Electronic Science and Technology of China, Chengdu, China
d
Department of Computer Science and Engineering, Qatar University, Doha, Qatar
a r t i c l e i n f o
Article history:
Received 4 November 2018
Revised 15 June 2019
Accepted 12 October 2019
Available online 17 October 2019
Keywords:
UAV
Dynamic planning
Path planning
Evolution algorithm
a b s t r a c t
With the wide application of Unmanned Aerial Vehicles (UAVs) in production and life,
more and more attention has been paid to the autonomous track planning of UAVs. When
UAV path planning algorithm is dealing with flying in an unknown complex environment,
there are some problems, such as inability to dynamically plan the track and slow speed
to calculate the path. This paper proposes a dynamic path planning based on an improved
evolutionary optimization algorithm. The experimental results show that the evolution-
ary optimization algorithm based on improved t-distribution can effectively deal with the
problems of high computational complexity and low search efficiency encountered in UAV
dynamic track planning. It has strong robustness and can dynamically plan the appropriate
track.
© 2019 Elsevier Ltd. All rights reserved.
1. Introduction
Unmanned Aerial Vehicle (UAVs) path planning refers to a feasible and satisfactory plan for UAVs under the premise of
considering the maneuverability, the surrounding environment threats and the mission time. The flight route can ensure the
safety of UAVs and can complete specific tasks. UAVs path planning is one of the cores of the Mission Planning System and
is widely used in control systems for robots, drones, missiles, etc. [1,2] .
Traditional route planning methods include sketch-based planning methods, cell-decomposition-based planning methods,
artificial potential-based planning methods [3–7] , etc. The planning method based on the sketch map usually first converts
the 3D scene into a 2D plan and then solves the problem by using the network map search method. This method is less
efficient when dealing with high-dimensional problems, and it is not possible to update planned routes in real time based
on environmental changes. The unit decomposition-based planning method needs to decompose the space into multiple
cells, but this decomposition process is very complicated. The advantage of the artificial potential field planning method is
that the planning speed is very fast, but there is a local minimum point where the attraction and repulsive force are equal,
which leads to the failure of the planning task. So far, intelligent evolutionary algorithm-based planning methods have been
the most effective and fastest growing. It can speed up the convergence, avoid falling into the local optimal solution, and
✩
This paper is for CAEE special section SI-umv. Note that it was originally submitted for SI-csc. Reviews processed and recommended for publication to
the Editor-in-Chief by Guest Editor Dr. Xiaokang Zhou.
∗
Corresponding author.
E-mail address: liuxiaolei@uestc.edu.cn (X. Liu).
https://doi.org/10.1016/j.compeleceng.2019.106493
0045-7906/© 2019 Elsevier Ltd. All rights reserved.