首页> 外文学位 >Multiple vehicle mission planning (navigation, scheduling).
【24h】

Multiple vehicle mission planning (navigation, scheduling).

机译:多车任务计划(导航,计划)。

获取原文
获取原文并翻译 | 示例

摘要

Autonomous mobile robots are being used in many applications. A commonality between all of these applications is the need for efficient and robust navigation of the robots. In many cases, multiple robots are used cooperatively to achieve multiple goals. This allows for more efficient and robust ways to achieve the goals of the tasks at hand. To further complicate the task, many of the environments are only partially known or dynamic in nature. This thesis presents a complete software system that has been built and tested in the real world with actual vehicles that can effectively navigate multiple vehicles to multiple goals cooperatively.; The system is described from a high-level perspective describing the system architecture, environment representation, and communication protocol used. The algorithms used to compute and maintain the optimal paths to goals are presented in detail. Tour assignment, the assignment of goals to vehicles, is discussed and five heuristics are presented to solve this problem in polynomial time. An empirical study of the tour assignment algorithms is presented to evaluate these heuristics and a summary of the results is presented.; The problem of navigating multiple goals cooperatively can also be modeled as a multidepot, multisalesmen problem. A transformation to reduce the multidepot, multisalesmen problem to a standard travelling salesman problem so that it can then be solved by any of various heuristic techniques is presented and discussed in detail, with a focus on allowing the reduction to be applicable in real-world situations. A practical application of the system to precision agriculture is described.
机译:自主移动机器人正在许多应用中使用。所有这些应用程序之间的共同点是需要高效,强大的机器人导航。在许多情况下,可以协同使用多个机器人来实现多个目标。这允许采用更有效,更可靠的方法来实现当前任务的目标。为了使任务进一步复杂化,许多环境本质上仅是部分已知的或动态的。本文提出了一个完整的软件系统,该系统已经在现实世界中使用实际车辆构建并经过测试,可以有效地将多个车辆协同导航到多个目标。从高层角度描述了系统,描述了系统体系结构,环境表示形式和所使用的通信协议。详细介绍了用于计算和维护到达目标的最佳路径的算法。讨论了旅行分配,即目标分配给车辆,并提出了五种启发式方法以在多项式时间内解决此问题。对巡回分配算法进行了实证研究,以评估这些启发式方法,并对结果进行了总结。合作导航多个目标的问题也可以建模为多仓库,多销售员的问题。提出并详细讨论了将多仓库,多销售商问题简化为标准旅行推销员问题的转换,以便可以通过各种启发式技术解决该问题,重点是使这种减少适用于实际情况。描述了该系统在精确农业中的实际应用。

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