本文为英国伦敦玛丽女王大学（作者：Asif Khan）的博士论文，共123页。

本文研究了多个移动的自主机器人，特别是资源有限的小型无人机在给定环境下对预定地面目标的协同监视方法。监控任务中的一个关键问题是机器人之间的协调以确定目标的时变位置。本文主要研究了两种监视应用：（1）静止目标的协同搜索；（2）运动目标的协同观测。协同搜索的目标是使固定目标定位的时间和误差最小化。协同观测的目的是最大限度地提高运动目标的观测时间和质量。

本文综述了在给定环境下，多机器人系统在更大范围内对预定目标进行监视的方法。这项研究调查确定了影响多机器人监控系统性能的各种因素和应用场景。本文提出了一种分布式融合延迟和不完全信息的策略，这种策略是由于感知和通信的限制，输入的融合信息由不同的无人机收集。提供了所需观测值的解析推导，以表明目标在区域中是否存在。将所需的观测值集成到旅行推销员问题（TSP）和多旅行推销员问题（MTSP）的迭代中，用于无人机的自主路径规划。此外，还对算法设计空间进行了探索，分析了在感知和通信受限的情况下，集中式和分布式协同对固定目标协同搜索的影响。本文还提出了无人机在多种不同分辨率条件下多个运动目标观测的应用。一个关键的贡献是使用四叉树数据结构来模拟无人机的环境和运动，这种建模有助于动态传感器在无人机上的配置，以最大限度地观察移动目标的数量以及分辨率。

This thesis investigates the coordinationapproaches of multiple mobile and autonomous robots, especiallyresource-limited small-scale UAVs, for the surveillance of pre-defined groundtargets in a given environment. A key research issue in surveillance task isthe coordination among the robots to determine the target’s time varyinglocations. The research focuses on two applications of surveillance: (i)cooperative search of stationary targets, and (ii) cooperative observation ofmoving targets. The objective in cooperative search is to minimize the time anderrors in finding the locations of stationary targets. The objective ofcooperative observation is to maximize the collective time and quality ofobservation of moving targets. The thesis presents a survey of the approachesin a larger domain of multi-robot systems for the surveillance of pre-definedtargets in a given environment. This survey identifies various factors andapplication scenarios that affect theperformance of multi-robot surveillance systems. The thesis proposes adistributed strategy for merging delayed and incomplete information, which is aresult of sensing and communication limitations, collected by different UAVs. An analytic derivation of the number of requiredobservations is provided to declare the absence or existence of a target in aregion. This number of required observations is integrated into an iterativeuse of Travelling Salesman Problem (TSP) and Multiple Travelling SalesmenProblem (MTSP) for autonomous path planning of UAVs. Additionally, it performsan exploration of the algorithmic design space and analyzes the effects of centralized and distributed coordination on the cooperativesearch of stationary targets in the presence of sensing and communicationlimitations. The thesis also proposes the application of UAVs for observingmultiple moving targets with differentresolutions. A key contribution is to use the quad-tree data-structure formodelling the environment and movement of UAVs. This modelling has helped inthe dynamic sensor placement of UAVs to maximize the observation of the numberof moving targets as well as the resolution of observation.

1. 引言

2. 相关工作

3. 多无人机搜索与观测的问题建模

4. 多无人机协同搜索中的信息融合

5. 多无人机协同搜索中的决策

6. 多无人机协作观测

7. 结论与未来工作展望

更多精彩文章请关注公众号：