首页> 外文会议>Third annual symposium on logic in computer science >Learning to acquire and select useful landmarks for route following
【24h】

Learning to acquire and select useful landmarks for route following

机译:学习获取并选择有用的地标以进行路线跟踪

获取原文
获取原文并翻译 | 示例

摘要

The paper describes a prototypical system for optimal landmarknacquisition and selection. Our landmark-learning approach does notnrequire any type of environment model to be supplied to the robot innadvance, and represents a step towards robots interacting with realnenvironments. The approach is fitted and tested for the TMGA system (P.nZingaretti et al., 1998), which the authors developed for landmarkntracking by adaptive, stereo template matching. Two complementarynstrategies, properly managed, are followed to construct a suitablensubset of landmarks: the selection of the more discriminant landmarksnand the selection of the landmarks that are more invariant in anneighbourhood. The robustness of the TMGA system in analysing thendiscriminant power of each landmark and the analysis of the disparitynmap and of the spatial activity maps of the stereo images are used fornidentifying discriminant and invariant landmarks. The experimentalnresults show that the number of matching failures, and consequentnlandmark changes during the following of a route is comparable with (notnmuch greater than) those obtained using an a-priori subset
机译:本文描述了一种用于最佳地标获取和选择的原型系统。我们的地标性学习方法不需要向机器人创新提供任何类型的环境模型,它代表了机器人与环境互动的一步。该方法已针对TMGA系统进行了安装和测试(P.nZingaretti等,1998),作者通过自适应的立体模板匹配为地标跟踪开发了该方法。接下来是经过适当管理的两个互补策略,以构造一个合适的地标子集:选择更具区别性的地标n和选择在附近地区不变的地标。然后,TMGA系统在分析每个界标的判别力以及对立体图像的视差图和空间活动图进行分析时的鲁棒性可用于识别判别和不变的界标。实验结果表明,匹配失败的次数以及随之而来的地标变化与使用先验子集获得的结果相当(不多)

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