基于改进分布估计算法的能耗最优AUV路径规划算法与仿真研究
Research on optimal AUV path planning algorithm and simulation of energy consumption based on improved estimation of distribution algorithm
戴晓强 1许赫威 1王莹 2孙啸天 1尚乐1
作者信息
- 1. 江苏科技大学自动化学院,镇江 212100
- 2. 江苏科技大学计算机学院,镇江 212100
- 折叠
摘要
针对经典基于等宽度直方图的分布估计算法(FWH)在解决自主水下机器人(AUV)路径规划问题中易陷入局部极值、计算精度不高等问题,提出一种受粒子群优化算法启发的改进分布估计算法.通过将FWH与粒子群优化算法中优势个体的部分筛选机制相结合的方法,组成双重优势个体产生方法,提高AUV路径规划最优路径的精度,增加收敛速度.在水下数字高程模型环境中对经典分布估计算法、PSO算法、A*算法以及改进后的算法进行性能评估.仿真结果表明,相比于改进前的算法,新的算法计算出的能耗值减少24.9％,有效增加计算精度,避免陷入局部极值,提高效率.
Abstract
In order to solve the problem that the classical distribution estimation algorithm based on equal width histogram(FWH)is easy to fall into local extremum and the calculation accuracy is not high in the path plan-ning problem of autonomous underwater vehicle(AUV),an improved distribution estimation algorithm inspired by particle swarm optimization(PSO)algorithm is proposed.By combining FWH with the partial screening mechanism of dominant individuals in particle swarm optimization algorithm,a dual individual screening method is formed to improve the accuracy of AUV path planning optimal path and increase the convergence speed.The performance of the classical distribution estimation algorithm,PSO algorithm,A*algorithm and the improved al-gorithm are evaluated in the environment of underwater digital elevation model.The simulation results show that compared with the previous algorithm,the energy consumption calculated by the new algorithm is reduced by 24.9％,which effectively increases the calculation accuracy,avoids falling into local extremum,and improves the efficiency.
关键词
分布估计算法/路径规划/能耗最优/粒子群优化算法/水下数字高程模型Key words
estimation of distribution algorithm/path planning/optimal energy consumption/particle swarm op-timization/underwater digital elevation model引用本文复制引用
出版年
2024
NETL
NSTL
万方数据