没有合适的资源?快使用搜索试试~ 我知道了~
首页改进的混沌自适应粒子群优化算法应用于机器人路径规划
改进的混沌自适应粒子群优化算法应用于机器人路径规划
0 下载量 145 浏览量
更新于2024-08-28
收藏 287KB PDF 举报
本文探讨了一种混沌自适应粒子群优化(Chaos-based Adaptive Particle Swarm Optimization, CAPSO)在静态环境下的移动机器人路径规划中的应用。CAPSO算法的核心在于创新地融合了两个评估函数:一是考虑路径长度,二是考虑障碍物风险。这使得算法在寻找最短且安全路径的同时,兼顾全局优化。 CAPSO算法的关键步骤包括:首先,它设计了一个综合评估函数,通过平衡路径的长度和避免障碍的策略,确保找到最优解。这表明路径规划不仅要追求效率,还要考虑到环境中的安全性,体现了实用性和智能性。 其次,引入了甲壳虫天线搜索(Beetle Antenna Search, BAS)算法对粒子位置更新方程进行改进。BAS算法有助于增强算法的全局搜索能力,使其在广阔的搜索空间中更有效地探索可能的路径,提高了算法在复杂环境中的导航性能。 为了进一步提升算法的性能,CAPSO采用了三角函数对控制参数进行自适应调整。这种动态调整策略确保了在每个算法执行阶段都能实现最佳的合作状态,增强了算法在局部搜索阶段的精度和收敛速度。这体现了算法的灵活性和自我优化能力。 最后,混沌映射被用来替代随机参数,利用混沌系统的特性,如非线性和复杂性,来增加算法的随机性和多样性,从而避免陷入局部最优,提高全局搜索的效率和稳定性。 总结来说,这篇文章介绍的混沌自适应粒子群优化算法是一种结合了混沌理论、自适应机制和协作搜索策略的创新路径规划方法,旨在解决移动机器人在复杂环境中的高效、安全路径规划问题。通过这些技术的集成,CAPSO展示了在处理此类问题时的优越性能和适应性。
资源详情
资源推荐
A Chaotic Adaptive Particle Swarm Optimization for Robot Path
Planning
Jianfang Lian
1
, Wentao Yu
1
, Weirong Liu
2
1. College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410004, China
E-mail: jflian_csuft@126.com
2. School of Computer Science and Engineering, Central South University, Changsha 410083, China
E-mail: frat@csu.edu.cn
Abstract: In this paper, a chaotic adaptive particle swarm optimization (CAPSO) algorithm is proposed for global path planning
of mobile robots under static environment. The fitness function of CAPSO synthesizes two evaluation functions which consider
path length and obstacle risk degree separately. Then the beetle antennae search algorithm (BAS) is introduced to modify the
particle position updating equation to strengthen the global search ability of the algorithm. The trigonometric function is adopted
for the adaptive adjustment of the control parameters for CAPSO to achieve the best cooperation in each stage of the algorithm
operation and improves the local search ability of the algorithm. The chaotic map is also used to replace the random parameters
of the basic particle swarm optimization. This procedure can improve the diversity of particle swarm and maintain the original
random characteristics. By CAPSO the global search ability and search speed are improved and optimal robot path planning
under static environment is realized. The simulation results verify the effectiveness of the proposed algorithm.
Key Words: Path planning, PSO, Beetle Antennae Search Algorithm, chaotic map, trigonometric function
1 Introduction
Path planning is one of the most important skills of the
mobile robots [1]. It can plan a collision-free and feasible
path in many tasks such as robot rescue [2], robot
surveillance [3] and robot patrol [4]. It is hard to find a
shortest and smooth path for robots because of the complex
robot working environment.
Existing classic path planning methods are usually
variations of some general approaches, such as potential
fields and Roadmap [5, 6]. The artificial potential field
method has the advantages of small computation, high real-
time performance and smooth trajectory planning. However,
at a certain point, when the gravitational force and repulsive
force are exactly equal and the direction is opposite, the
robot will think that it has reached the target point and stop
moving or wandering in a certain area.
The roadmap planning method can decompose the
forward space of the robot according to the shape of the
obstacle. By connecting the decomposed path, an optimal
collision-free smooth curve can be obtained. However, with
the increase of obstacles, the number of connections
between its vertices will inevitably increase, which will lead
to the increase of planning complexity and planning time.
Another major path planning method is heuristic method.
When using heuristic algorithms, it is not guaranteed to find
a solution, but if a solution is found, it will be done much
faster than deterministic methods. The main metaheuristic
approaches employed in robot path planning are simulated
annealing (SA) [7], genetic algorithms (GA) [8], particle
swarm optimization (PSO) [9], and ant colony (ACO) [10].
Simulated annealing (SA) is a stochastic global
optimization algorithm simulated by Kirkpatrick [11]. The
convergence probability of simulated annealing algorithm is
1. This characteristic can guarantee the achievement of
**
This work is supported by National Nature Science Foundation (Grant
Nos. 61602529, 61672539) and Planned Science and Technology Project
of Hunan Province (Grant Nos. 603286580059).
global optimization when it is used for robot global path
planning. While it has the disadvantages of slow
convergence speed. Similarly, genetic algorithms (GA) has
good optimization ability and slow convergence speed [8].
The ant colony algorithm (ACO) is a random search
optimization algorithm with characteristics of positive
feedback and parallel computing, which applies to various
problems such as traveling salesman problem, quadratic
programming problem and production scheduling problem.
However, the optimal path of ant colony algorithm planning
is time-consuming.
Compared the above algorithms used for path planning,
PSO is easy to implement and can quickly approach the
optimal solution. However, it may lead to local optima
problem. In order to prevent PSO from falling into local
optima due to the premature convergence in the search
process, domestic and foreign scholars have proposed a
number of methods to improve the PSO algorithm. In [12],
a ‘stepwise’ adjustment strategy is proposed for the inertia
weights in the basic PSO algorithm, and different sizes of
inertia weights are set for the population particles in
different periods. To some extent, this method better solves
the premature convergence problem of PSO algorithm, but
the selection of parameters requires a large number of data
experiments, and the parameters are also susceptible to the
size of the environment, so the algorithm flexibility is not
high.
Particle depletion often occurs in the running process of
PSO algorithm, which seriously affects the optimization
accuracy. For this problem, the literature [13] and the
literature [14] added the bounded random disturbance
variable and the adjacent particle information respectively
on the basis of the speed update formula. These two methods
are to maintain the diversity of particle optimization by
introducing a mutation mechanism to the particle velocity
Proceedings of the 38th Chinese Control Conference
Jul
y
27-30, 2019, Guan
g
zhou, China
4751
下载后可阅读完整内容,剩余5页未读,立即下载
weixin_38506713
- 粉丝: 4
- 资源: 907
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
最新资源
- 李兴华Java基础教程:从入门到精通
- U盘与硬盘启动安装教程:从菜鸟到专家
- C++面试宝典:动态内存管理与继承解析
- C++ STL源码深度解析:专家级剖析与关键技术
- C/C++调用DOS命令实战指南
- 神经网络补偿的多传感器航迹融合技术
- GIS中的大地坐标系与椭球体解析
- 海思Hi3515 H.264编解码处理器用户手册
- Oracle基础练习题与解答
- 谷歌地球3D建筑筛选新流程详解
- CFO与CIO携手:数据管理与企业增值的战略
- Eclipse IDE基础教程:从入门到精通
- Shell脚本专家宝典:全面学习与资源指南
- Tomcat安装指南:附带JDK配置步骤
- NA3003A电子水准仪数据格式解析与转换研究
- 自动化专业英语词汇精华:必备术语集锦
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功