【24h】

PATH PLANNING IN PARTIALLY KNOWN ENVIRONMENTS

机译:部分已知环境中的路径规划

获取原文
获取原文并翻译 | 示例

摘要

Many path planning algorithms assume that thernenvironment is either perfectly known or perfectly unknownrn(in which case the environment is assumed to be empty). Wernconsider the intermediate case in which partial priorrninformation, in the form of a probabilistic occupancy cell map,rnis available. Using heuristics for clustering the spatialrndistribution of paths based on most common routes, we derivernan algorithm, called the PD planner, which exploits thisrnprobabilistic information. Unlike local entropy-basedrnplanners, it can account for global effects such as the need forrna robot to “Back Up” if it becomes stuck in a blind alley. Thernperformance of the algorithm is assessed in a simulatedrnhighly damaged indoor scenario, where we show thatrnexploiting the global impacts of uncertainty has the potentialrnto significantly reduce both travel time and travel distance.
机译:许多路径规划算法都假定环境是完全已知的或完全未知的(在这种情况下,假定环境为空)。考虑到中间情况,其中部分先验信息以概率占用细胞图的形式存在。使用启发式方法基于最常见的路线对路径的空间分布进行聚类,我们推导了称为PD规划器的算法,该算法利用了这种概率信息。与基于局部熵的规划师不同,它可以解决全局影响,例如,如果卡在盲区中的机器人需要“备份”,那么它就可以解决。在模拟的室内严重破坏场景中评估了算法的性能,在该场景中,我们证明了利用不确定性的全局影响有可能显着减少旅行时间和旅行距离。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号