基于贝叶斯优化的无人机路径规划算法  

UAV Path Planning Based on Bayesian Optimization Algorithm

作　　者：符小卫[1] 高晓光[1] 

机构地区：[1]西北工业大学电子信息学院,西安710072

出　　处：《宇航学报》2006年第3期422-425,共4页Journal of Astronautics

基　　金：国家自然科学基金重大研究计划资助项目(90205019);航空科学基金资助项目(05D53022)

摘　　要：提出了一种基于贝叶斯优化算法的无人机路径规划方法。把无人机路径编码为离散时间上的速度和航向变化序列,每一步的速度和航向变化量都限制在无人机相应最大变化量之内,所以这种编码方法对应的物理轨迹是无人机可飞的。利用每代种群中的优良解集构造贝叶斯网络,用贝叶斯网络的结构体现染色体基因位之间的联系,用贝叶斯网络参数体现染色体基因位之间的联系程度。设计了一个多变量K2度量评价网络的优劣。用贝叶斯网络产生新的染色体以体现种群的进化,这取代了传统遗传算法的交叉和变异过程。如果不满足终止条件,则用新一代种群的优良解集构造贝叶斯网络,直到满足终止条件。仿真结果验证了算法的有效性。A Bayesian optimization algorithm （BOA） for UAV path planning problem is presented, which involves choosing path representation and designing appropriate metric to measure the quality of the constructed network. Unlike our previous work that used genetic algorithm to implement implicit learning, the learning in the proposed algorithm is explicit, i.e. eventually, we will be able to identify and mix building blocks directly. The Bayesian optimization algorithm is applied to implement such explicit learning by building a Bayesian network of the joint distribution of solutions. The conditional probability of each variable in the network is computed according to an initial set of promising solutions. Subsequently, each new instance for each variable is generated by using the corresponding conditional probabilities, until all variables have been generated, i.e. in our case, a new path genotype string has been obtained. Another set of path genotype strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the conditional probabilities for all nodes in the Bayesian network are updated again using the current set of promising path genotype strings. Experimental results demonstrate that this approach can overcome drawbacks of other path planning algorithm. It is also suggested that the learning mechanism in the proposed approach might be suitable for other multivariate encoding problems.

关 键 词：无人机 路径规划 遗传算法 贝叶斯网络 贝叶斯优化 

分 类 号：V218[航空宇航科学与技术—航空宇航推进理论与工程]