Affiliations: [a] College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, Guangdong, China | [b] Department of Computer Science, City University of Hong Kong, Hong Kong,
China | [c] School of Computer Science and Electronic Engineering, University of Essex,
Essex, UK
Correspondence:
[*]
Corresponding author: Zexuan Zhu/Jian-Qiang Li, College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, Guangdong, China. E-mail:[email protected]
Abstract: Global path planning is a fundamental problem of mobile
robotics. The majority of global path planning methods are designed to find
a collision-free path from a start location to a target location while
optimizing one or more objectives like path length, smoothness, and safety
at a time. It is noted that providing multiple tradeoff path solutions of
different objectives is much more beneficial to the user's choice than
giving a single optimal solution in terms of some specific criterion. This
paper proposes a global path planning of wheeled robots using
multi-objective memetic algorithms (MOMAs). Particularly, two MOMAs are
implemented based on conventional multi-objective genetic algorithms with
elitist non-dominated sorting and decomposition strategies respectively to
optimize the path length and smoothness simultaneously. Novel path encoding
scheme, path refinement, and specific evolutionary operators are designed
and introduced to the MOMAs to enhance the search ability of the algorithms
as well as guarantee the safety of the candidate paths obtained in complex
environments. Experimental results on both simulated and real environments
show that the proposed MOMAs are efficient in planning a set of valid
tradeoff paths in complex environments.
