Configuration optimization, path planning and control of serial modular and reconfigurable robots

美国国家研究委员会的《2020年制造业挑战展望》指出，在制造业的所有优先技术领域中，“可重构和适应性系统”最为重要。目前存在几种旨在提高生产灵活性的制造方法，例如精益制造和准时生产。在机器人领域，最近的努力利用模块化和可重构机器人（mrr），以实现制造的灵活性。此外，近年来，一些制造工具供应商开始提供基于mrr的解决方案，用于设计多功能设施，这些设施比特定于任务的自动化硬件提供更高的灵活性。为了满足未来几年对适应性制造系统的预计需求，对mrr的需求将继续增长。因此，申请人的研究计划将侧重于推进MRR的运动规划和控制方法，以促进这些系统在先进制造应用中的更广泛使用。首先，研究计划将侧重于开发智能路径规划方法来解决mrr的逆运动学（IK）问题。通过结合自适应小生境方法，提出的方法将能够为没有封闭形式解析解的配置找到IK问题的多个解决方案。其次，该计划将针对mrr的构型无关鲁棒控制方法的发展。最后，研究计划将集中于利用基于遗传算法的新型任务算法开发mrr的配置优化策略。提议的算法将利用申请人在研究计划期间开发的多解IK求解器来确定给定任务最合适的MRR配置。通过上述里程碑的成功实施，该研究计划将使mrr在各种制造应用中的应用成为可能。

Visionary Manufacturing Challenges for 2020 by the USA National Research Council points out that among all priority technology areas in manufacturing "reconfigurable and adaptable systems" are most important. There exist several manufacturing approaches aimed at increasing production flexibility, such as lean manufacturing, and just-in-time production. In the field of robotics, recent efforts exploit modular and reconfigurable robots (MRRs) for the purpose of enabling manufacturing flexibility. Moreover, in recent years, several manufacturing tools suppliers began to offer MRR-based solutions for design of versatile facilities that offer improved dexterity over task-specific hardware for automation. This need for MRRs will continue to grow in order to meet the projected demand for adaptable manufacturing systems in upcoming years. Therefore, the applicant's research program will focus on advancing motion planning and control methodologies of MRR to facilitate wider scale use of those systems in advanced manufacturing applications. First, the research program will focus on developing intelligent path planning methods to solve the Inverse Kinematics (IK) problem for MRRs. By incorporating adaptive niching methods, the proposed approach will be able to find multiple solutions of the IK problem for configurations that have no closed form analytical solution. Secondly, the program will target the development of configuration-independent robust control methods for MRRs. Finally, the research program will focus on development of configuration optimization strategies for MRRs that utilize novel GA-based task-based algorithms. The proposed algorithms will utilize the multi-solution IK solver developed by the applicant during the research program to identify the most suitable MRR configuration for a given task. Through the successful execution of the aforementioned milestones, this research program will enable the utilization of MRRs in a wide variety of manufacturing applications.