Session: 06-11-03 Ocean Engineering Technology III

Paper Number: 130108

130108 - Research on Usv Smooth Path Planning Method Based on Iac-Qpso Algorithm 

In order to solve the problem of USV smooth path planning in complex obstacle environment, a USV smooth path planning method based on improved ant colony algorithm and quantum particle swarm optimization (IAC-QPSO) is proposed. The traditional ant colony algorithm is improved from pheromone renewal formula, pheromone volatility coefficient and deadlock, which avoids precocious phenomenon and speeds up the convergence speed of ant colony algorithm. The global optimal polyline path is planned by the improved ant colony algorithm in the complex grid environment, and the optimal path transition region and the optimal polyline channel are obtained by redundant optimization strategy and expansion strategy. The particle dimension is determined according to the number of path transition regions, which solves the blindness and redundancy of the particle dimension determination in the existing quantum particle swarm optimization algorithm, realizes the adaptive adjustment of particle dimension according to the environment, effectively reduces the particle dimension, and improves the convergence speed of the algorithm. The particle position initialization in the transition region of the optimal path improves the quality of the initial solution and makes it easier to find the global optimal path. The boundary constraint is determined according to the optimal polyline channel, which greatly reduces the understanding space, reduces the solving difficulty and improves the searching efficiency of the algorithm. The spiral search mechanism of whale optimization algorithm, Levi flight, and the particle resurrection mechanism are introduced into the quantum particle swarm optimization algorithm to increase the effective particle number, enrich the population diversity, improve the local mining ability of the algorithm, and avoid falling into the local optimal. The fitness function is established by weighted path length, safety degree and smoothness, and the optimal smooth path is obtained by using the improved quantum particle swarm optimization algorithm and cubic spline interpolation. Simulation results show the effectiveness of the proposed algorithm.

Presenting Author: Hongjian Wang College of Intelligent Systems Science and Engineering, Harbin Engineering University

Presenting Author Biography: N/A

Authors:

Chengfeng Li College of Intelligent Systems Science and Engineering, Harbin Engineering University
Hongjian Wang College of Intelligent Systems Science and Engineering, Harbin Engineering University
Yutong Huang College of Intelligent Systems Science and Engineering, Harbin Engineering University
Bo Zhong College of Intelligent Systems Science and Engineering, Harbin Engineering University
Zhikang Chi College of Intelligent Systems Science and Engineering, Harbin Engineering University
Jinmu Tian College of Intelligent Systems Science and Engineering, Harbin Engineering University
Naifu Luo College of Intelligent Systems Science and Engineering, Harbin Engineering University