An Adaptive Route Plan Technique with Risk Contour for Autonomous Navigation of Surface Vehicles

Jeong, Mingi; Lee, Eun-Bang; Lee, Moonjin

doi:10.1109/oceans.2018.8604638

Cited by 4 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In previous works, the purposes of ASV path planning commonly target obstacle avoidance, energy efficiency, path length optimization, etc. [14][15][16] with fixed interaction points. However, in such a scenario with time-sensitive interaction points between ASV and AUGs, the ASV has to schedule an efficient sequence of access to the interaction points of AUGs that occur at specific times.…”

Section: Introductionmentioning

confidence: 99%

Heuristic Surface Path Planning Method for AMV-Assisted Internet of Underwater Things

et al. 2023

View full text Add to dashboard Cite

Ocean exploration is one of the fundamental issues for the sustainable development of human society, which is also the basis for realizing the concept of the Internet of Underwater Things (IoUT) applications, such as the smart ocean city. The collaboration of heterogeneous autonomous marine vehicles (AMVs) based on underwater wireless communication is known as a practical approach to ocean exploration, typically with the autonomous surface vehicle (ASV) and the autonomous underwater glider (AUG). However, the difference in their specifications and movements makes the following problems for collaborative work. First, when an AUG floats to a certain depth, and an ASV interacts via underwater wireless communication, the interaction has a certain time limit and their movements to an interaction position have to be synchronized; secondly, in the case where multiple AUGs are exploring underwater, the ASV needs to plan the sequence of surface interactions to ensure timely and efficient data collection. Accordingly, this paper proposes a heuristic surface path planning method for data collection with heterogeneous AMVs (HSPP-HA). The HSPP-HA optimizes the interaction schedule between ASV and multiple AUGs through a modified shuffled frog-leaping algorithm (SFLA). It applies a spatial-temporal k-means clustering in initializing the memeplex group of SFLA to adapt time-sensitive interactions by weighting their spatial and temporal proximities and adopts an adaptive convergence factor which varies by algorithm iterations to balance the local and global searches and to minimize the potential local optimum problem in each local search. Through simulations, the proposed HSPP-HA shows advantages in terms of access rate, path length and data collection rate compared to recent and classic path planning methods.

show abstract

Section: Introductionmentioning

confidence: 99%