CAREER: Cooperative Payload Transport by Robot Collectives

在这份职业发展提案中，首席调查员(PI)审查了框架的开发，以创建、分析、实施和验证分布式机器人集体中的合作有效载荷运输能力。主要的研究问题涉及对基础物理和信息基础设施、系统架构和合作机制的性质和选择的调查，目的是实现具有可量化的绩效效益的实施。这种研究是在具有可变数量移动机械手的行进编队的创建、控制和主动重构的特定背景下提出的，用于协作有效载荷运输任务。在共同理论但易于计算的框架下确定要审查的三个方面的研究包括：(1)为预先规定的任务设计和优化编队的方法；(2)针对变化的任务的编队控制和适应的可扩展计划；以及(3)对模块化、感知、通信、建模和权力下放对合作绩效的作用的定量审查。提出了一种广泛的基于模拟的研究和严格的实验验证的双管齐下的方法来评估候选系统的性能。这项研究的主要贡献是为此类协作系统的设计、分析、控制和验证开发了一个灵活的统一框架，并讨论了统一处理框架的广泛适用性：由分散的操纵代理机队进行协作材料处理；具有增强运动能力的混合关节轮式系统；以及在共享虚拟环境中与常见虚拟对象进行分布式触觉协作。总体教育目标是开发课程、工具和人力资源基础，以构思、研究、实施和验证此类系统。具体的教育举措包括：(1)开发新的机器人和机电一体化本科和研究生课程并将其纳入课程；(2)加强现有的设计和系统本科课程并使之现代化；(3)创建教育技术课程单元，以概括知识，并在大学内部提供一个与当地高中、工业界和广大教育界进行外联的机制。

In this career development proposal, the Principal Investigator (PI) examines the development of frameworks to create, analyze, implement and validate cooperative payload transport capabilities in distributed robot collectives. The principal research questions pertain to the investigation of the nature and selection of the underlying physical and informational infrastructure, system architecture, and mechanisms of cooperation, with the goal of realizing implementations with quantifiable performance benefits. Such investigation is proposed in the specific context of creation, control and active reconfiguration of marching formations with variable numbers of mobile manipulators for cooperative payload transportation tasks. The three aspects of research identified for examination under the common theoretical but computationally tractable framework, include: (i) methodologies for design and optimization of formations for pre-specified tasks; (ii) scalable schemes for control and adaptation of formations for changing tasks; and (iii) quantitative examination of the role of modularity, sensing, communication, modeling and decentralization on cooperative performance. A two-pronged approach of extensive simulation-based study and rigorous experimental ratification is proposed to evaluate the performance of candidate systems. The key contribution from this investigation is the development of a flexible unified framework for design, analysis, control and validation of such cooperating systems and the broad applicability of the framework for uniformly treating: cooperative material-handling by a fleet of decentralized manipulation agents; hybrid articulated-wheeled systems with enhanced locomotion capabilities; and distributed haptic collaboration with a common virtual object within a shared virtual environment; is discussed.The overall educational goals are to develop the courses, tools and the human resource base to conceive, investigate, implement and validate such systems-of-systems. Specific educational initiatives include: (i) developing and integrating new undergraduate and graduate courses in robotics and mechatronics into the curriculum; (ii) strengthening and modernizing existing undergraduate courses in design and systems; (iii) creating educational technology curricular modules for encapsulating knowledge and providing a mechanism for outreach within the university, with local high schools and industry and the educational community at large.