TY - JOUR
T1 - Optimal Path Planning Generation for Mobile Robots using Parallel Evolutionary Artificial Potential Field
AU - Montiel, Oscar
AU - Sepúlveda, Roberto
AU - Orozco-Rosas, Ulises
N1 - Publisher Copyright:
© 2014, Springer Science+Business Media Dordrecht.
PY - 2015/9/20
Y1 - 2015/9/20
N2 - In this paper, we introduce the concept of Parallel Evolutionary Artificial Potential Field (PEAPF) as a new method for path planning in mobile robot navigation. The main contribution of this proposal is that it makes possible controllability in complex real-world sceneries with dynamic obstacles if a reachable configuration set exists. The PEAPF outperforms the Evolutionary Artificial Potential Field (EAPF) proposal, which can also obtain optimal solutions but its processing times might be prohibitive in complex real-world situations. Contrary to the original Artificial Potential Field (APF) method, which cannot guarantee controllability in dynamic environments, this innovative proposal integrates the original APF, evolutionary computation and parallel computation for taking advantages of novel processors architectures, to obtain a flexible path planning navigation method that takes all the advantages of using the APF and the EAPF, strongly reducing their disadvantages. We show comparative experiments of the PEAPF against the APF and the EAPF original methods. The results demonstrate that this proposal overcomes both methods of implementation; making the PEAPF suitable to be used in real-time applications.
AB - In this paper, we introduce the concept of Parallel Evolutionary Artificial Potential Field (PEAPF) as a new method for path planning in mobile robot navigation. The main contribution of this proposal is that it makes possible controllability in complex real-world sceneries with dynamic obstacles if a reachable configuration set exists. The PEAPF outperforms the Evolutionary Artificial Potential Field (EAPF) proposal, which can also obtain optimal solutions but its processing times might be prohibitive in complex real-world situations. Contrary to the original Artificial Potential Field (APF) method, which cannot guarantee controllability in dynamic environments, this innovative proposal integrates the original APF, evolutionary computation and parallel computation for taking advantages of novel processors architectures, to obtain a flexible path planning navigation method that takes all the advantages of using the APF and the EAPF, strongly reducing their disadvantages. We show comparative experiments of the PEAPF against the APF and the EAPF original methods. The results demonstrate that this proposal overcomes both methods of implementation; making the PEAPF suitable to be used in real-time applications.
KW - APF
KW - Artificial potential field
KW - Mobile robot navigation
KW - PEAPF
KW - Parallel evolutionary computation
KW - Path planning
UR - http://www.scopus.com/inward/record.url?scp=84932195890&partnerID=8YFLogxK
U2 - 10.1007/s10846-014-0124-8
DO - 10.1007/s10846-014-0124-8
M3 - Artículo
SN - 0921-0296
VL - 79
SP - 237
EP - 257
JO - Journal of Intelligent and Robotic Systems: Theory and Applications
JF - Journal of Intelligent and Robotic Systems: Theory and Applications
IS - 2
ER -