【作者】 刘亚波；

【导师】 姚敏； 吴朝晖； 杨建华；

【作者基本信息】 浙江大学 ， 计算机科学与技术， 2011， 博士

【摘要】 随着机器人技术的发展和能力的不断提高,机器人应用于更广泛的领域,特别是一些枯燥的、危险的、肮脏的或人类无法进入的特殊场合,满足核泄漏检测、火星探险、矿井作业及军事反恐等应用需求。多机器人系统在功能、结构和应对复杂、未知、危险环境和任务等方面明显优于单智能机器人,但同时由于其复杂性、动态性、时变性及耦合性的增加也给设计和分析带来一些问题,因此多机器人的相互协调和协作显得至关重要。针对异构多机器人系统的协同技术的研究问题,本文的主要工作主要集中在以下五个方面：第一,多机器人系统的基础现状与发展趋势研究。对多机器人系统的任务分配、基于网络的多机器人协作、演化发育机器人方法论、机器人基础软件平台进行了阐述,对多机器人系统在军事领域的应用作了调研和分析,特别是对本文涉及的具体内容、研究现状、应用情况、未来发展规划和热点问题做了重点论述。第二,时空受限的复杂环境中机器人协同技术研究。提出了动态熟人网络以便维护机器人间的关联关系,机器人间的熟悉度将用于任务的间接代价估计及竞标；采用时间约束的与或任务树对复杂任务进行建模,同时引入相应的任务代价评估方法；为支持该类型任务树及直接/间接任务竞拍,本文采用一种增强的拍卖定锤决标方法。在面向灾害应急的一系列仿真及实物实验中,设定不同参数以评估与比较系统性能。第三,机器人协作行为的自主演化技术研究。采用人工演化技术合成物理异构多机器人系统的控制器,使个体间表现出协作行为；研究在没有人类干预的情况下,具有不同能力的机器人演化出适应各自能力的控制策略,使协作行为效率最大化。在仿真实验中,比较了有协作与无协作之间的系统性能,研究了虚拟抓手对演化收敛速度的影响,分析了子种群代表挑选方法及采用共享或独立适应度函数对协作行为合成的影响。第四,基于服务的网络机器人系统的普适协作技术研究。首先引入语义地图和服务抽象与定义的基本概念；其次,为自动生成可行的服务构造以便执行具体不同任务,提出了服务的构造与重构算法；再次,提出服务推理与使能过程解决服务不可用问题；为方便服务构造的选择,提出了服务构造的代价评估函数。在实物及仿真环境中采用不同的参数设置,对服务框架进行了测试评估。最后,智能微小型地面机器人群体协同系统研究。以特种作业任务为背景,大规模的智能机器人组成的、通过无线组网技术连接的、具有一定作业功能的地面机器人群体协同系统通过协调和协作完成单机器人无法满足或难以完成的全天候大范围作业任务。提出了作业任务的目标导向的估价机制及快速代价评估方法,引入了一种能够对传感器的先验可信度进行实时修正的复杂环境下的多传感器目标识别方法,提供了机器人的无线组网、避障控制、编队控制等支撑技术的实现,实物演示验证了系统的协同作业能力。更多还原

【Abstract】 With the developing of the technologies, the ability of robot is significantly improved. The robots therefore become more and more common in many fields, doing dirty, dull or dangerous works, ranging from mapping and exploration to robot soccer.Researches generally agree that multi-robot systems (MRS) with inherently distributed character may behave more robustly and effectively and accomplish coordinated tasks that are not possible for single robot systems. However, suitable strategies are required to organize and control these systems to alleviate the inherent complex, dynamic, volatile, and coupling influences. This dissertation makes a number of contributions in response to the need for effective coordination approaches in heterogeneous multi-robot teams.1) We address the need for a survey of the relevant literature by providing an introduction to the multi-robot coordination, a review and analysis of the state of the art in the field, and a discussion of remaining research challenges. The main focus of the survey include multi-robot systems, task allocation, network robot systems, evolutionary development robotics, development environments and middleware for robots, and their military applications.2) We proposed the constraint based approach (CoBA) for the cooperation among a team of heterogeneous robots in the complex environment with temporal constraints and network constraints. First we put forward a dynamic acquaintance network to facilitate the maintenance of inter-robot relationship. Hence the degree of acquaintanceship can be used in the indirect bid of task. Second, in order to model the complex tasks, we introduce the AND/OR task tree with temporal constraints and corresponding evaluation method. Additionally, an auction clearing routine, which supports the AND/OR task tree with temporal constraints and direct/indirect task auction is proposed to enable effective multi-robot task allocation in spite of various constraints. The solution was validated in both simulation and physical environments by a series of experiments in disaster response domains.3) Evolutionary Robotics (ER) is a methodology that uses evolutionary computation to develop controllers for autonomous robots. The existing works focus on single robot system or physically homogenous multi-robot teams while physical heterogeneousness is more common in the real world. Hence, it is instructive to examine whether the cooperative behaviors can be synthesized using artificial evolution for a team of physically heterogeneous robots. Moreover, an important contribution of this dissertation is to answer the question:whether robots with distinct capabilities can synthesize their control strategies to accommodate their own capabilities without any human interventions? This dissertation offers an empirical analysis of the collaboration mechanisms and presents some basic advice about how to choose appropriate evolution methods. Simulated experiments, using a team of E-puck robots, show that evolution can discover effective controllers for robots with distinct capabilities.4) Network robot system (NRS) is a new concept that integrates physical autonomous robots, environmental sensors and human-robot interactions through network-based cooperation. We aim to provide a ubiquitous and cooperative service framework for NRS. We first present foundational concepts of semantic map and service definition for the framework. Then, in order to generate feasible service configurations to fulfill tasks, we propose service configuration and reconfiguration algorithms, which dynamically search the appropriate service configurations for different tasks. Additionally, we put forward a service reasoning and enabling process to tackle the service unavailable problems. A cost evaluation function for service configuration is also proposed to facilitate the selection of suitable configurations. We tested and evaluated the framework in both simulation system and physical environment.5) A multiple UGV system is constructed under the application of special tasks. This system implements unmanned mobility, reconnaissance and surveillance, tactical communications, formation control, path planning and obstacle avoidance, and mission planning and monitoring. Specially, an objective oriented task evaluation method is proposed for quick response to different types of tasks, and a method with real-time correction of the credibility of the prior information is suggested for multi-sensor target recognition. The cooperative ability of the system was validated in physical environment by a series of experiments.更多还原