Improving LRTA*(k) by Eacapsing from the Trap of h-surface

摘要：

LRTA* is a real time heuristic search algorithm widely used for agent find path on unknown maps, which is special case of LRTA*(k) with k=. In each iteration, it updates the heuristic estimate of the k states. Our Adaptive LRTA*(k) (abbr. ALATA*(k))) update the k states according the history of visited states which means the special terrain. Our update-strategy maintains heuristic admissibility. ALATA*(k) update heuristic value of more states just as LRTA*(k) when the agent stuck into a local trap of heuristic surface, also keep the good theoretical properties of LRTA* when there are few obstacles. Experimentally, we show that the ALATA*(k) produces better solution in the first trial than LRTA*, While reduced the number of expanded states of LRTA*(k).

关键词： LRTA* LRTA*(k) adadptive visited history hvalue sutface heuristic search

作者: Shaohan Liu Zhiqiu Huang

作者单位: College of Information Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing, China

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Automation Engineering(ICCAE 2011)(2011年第三届IEEE计算机与自动化工程国际会议)

会议地点: 重庆

会议语种:英文

页码: 251-254

在线出版日期: 2011-01-21（万方平台首次上网日期，不代表论文的发表时间）