Citation: Zhang, B.; Li, G.; Zheng, Q.;
Bai, X.; Ding, Y.; Khan, A. Path
Planning for Wheeled Mobile Robot
in Partially Known Uneven Terrain.
Sensors 2022, 22, 5217. https://
doi.org/10.3390/s22145217
Academic Editors: Shuai Li, Dechao
Chen, Vasilios N. Katsikis, Predrag
Stanimirovi´c, Dunhui Xiao and
Mohammed Aquil Mirza
Received: 2 June 2022
Accepted: 9 July 2022
Published: 12 July 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Path Planning for Wheeled Mobile Robot in Partially Known
Uneven Terrain
Bo Zhang
1,2
, Guobin Li
1,2
, Qixin Zheng
1,2
, Xiaoshan Bai
1,2,
* , Yu Ding
1,2
and Awais Khan
1,2
1
College of Mechatronics and Control Engineering, Shenzhen University, Shenzhen 518060, China;
zhangbo@szu.edu.cn (B.Z.); 2100292009@email.szu.edu.cn (G.L.); 1810294017@email.szu.edu.cn (Q.Z.);
1910294029@email.szu.edu.cn (Y.D.); awaiskhan@szu.edu.cn (A.K.)
2
Shenzhen City Joint Laboratory of Autonomous Unmanned Systems and Intelligent Manipulation,
Shenzhen University, Shenzhen 518060, China
* Correspondence: baixiaoshan@szu.edu.cn
Abstract:
Path planning for wheeled mobile robots on partially known uneven terrain is an open
challenge since robot motions can be strongly influenced by terrain with incomplete environmental
information such as locally detected obstacles and impassable terrain areas. This paper proposes
a hierarchical path planning approach for a wheeled robot to move in a partially known uneven
terrain. We first model the partially known uneven terrain environment respecting the terrain
features, including the slope, step, and unevenness. Second, facilitated by the terrain model, we
use A
?
algorithm to plan a global path for the robot based on the partially known map. Finally, the
Q-learning method is employed for local path planning to avoid locally detected obstacles in close
range as well as impassable terrain areas when the robot tracks the global path. The simulation and
experimental results show that the designed path planning approach provides satisfying paths that
avoid locally detected obstacles and impassable areas in a partially known uneven terrain compared
with the classical A
?
algorithm and the artificial potential field method.
Keywords: hierarchical path planning; uneven terrain; A
?
algorithm; Q-learning algorithm
1. Introduction
Mobile robots that are deployed for rescue missions in uneven cluttered terrains
generally need to have the ability of autonomous navigation and path planning. However,
it is challenging to plan a feasible path efficiently for a robot to move in uneven terrains due
to the terrains’ slope, step, and unevenness. In [
1
], a real-time obstacle avoidance method
is proposed based on trajectory space, which considers the mobile robot’s uncertainty.
However, the uneven terrain modeled in [
1
] does not reflect a realistic terrain environment
well. Some research has been conducted for robotic path planning on uneven terrains, such
as the path planning for the Chang’e-4 lunar exploration rover Yutu-2 to move through
uneven rough terrain [
2
]. The Yutu-2 lunar rover can passively adapt to the uneven
terrain on the moon’s far side by using its differential mechanism and rocker arm. This
configuration enables the rover to reduce its pitch angle by half compared with other
vehicles when clearing an obstacle. The Curiosity rover used in the US Mars exploration
mission is a six-wheeled vehicle [
3
]. It uses a rocker arm steering structure in which
two front wheels and two rear wheels are independent such that the rover can pivot
steering. The Curiosity rover, meanwhile, relies on a six-wheeled primary and secondary
joystick system to navigate over the uneven rocks of Mars. The above work focuses on
vehicles’ adaptation and safe travel over rough terrain through sophisticated mechanical
mechanisms. However, it must be acknowledged that this is a complex and costly approach.
The application of path planning techniques would be much more beneficial if they could
be used to continually identify whether they can be safely navigated to plan safe and
feasible paths in rough environments.
Sensors 2022, 22, 5217. https://doi.org/10.3390/s22145217 https://www.mdpi.com/journal/sensors
