摘要
按照建模原理的不同,将目前各种3维路径规划方法分为4类,阐述了各种方法的工作原理,指出了各种方法在不同应用领域的优势和劣势.分别从实时性、动态环境适应性、规划路径的光滑性、全局规划能力以及加入动力学约束的难易程度等方面进行了比较.分析结果指出,基于虚拟势场与导航函数的方法实时性最好,可在局部规划器中优先选用.基于数学优化的方法能够综合考虑各种动力学约束;而基于生物智能的方法虽然便于表达各种棘手的约束,但规划周期太长,只适用于长周期调用.
A variety of three-dimensional path planning methods are divided into four categories according to their mod-eling principles.The working principles of those methods are introduced,and the advantages and disadvantages are pointed out in various applications.All of them are compared from the viewpoints whether they can be used in real-time and dy-namic environment,whether they can achieve smooth path and global planning,and whether they can add different dynamic constraints conveniently.Conclusions are drawn from the comparison that the method based on virtual potential field and navigation function is superior to others for its real-time performance and will be a priority in local planners.The method based on mathematic optimization is capable of dealing with a variety of dynamic constraints.In contrast to mathematic op-timization,the bio-inspired one is limited in solutions of long calling cycle for its large planning period although it is efficient to describe intractable constraints.
出处
《机器人》
EI
CSCD
北大核心
2010年第4期568-576,共9页
Robot
基金
国家自然科学基金资助项目(60775056
60705028)
关键词
3维空间
避障
环境建模
动力学约束
搜索算法
实时
three dimensional space
obstacle avoidance
environment modeling
dynamic constraint
searching algorithm
real-time
