Configuration Optimization and Advanced Docking Design for Two Classes of Reconfigurable Robotic Systems

两类可重构机器人系统的配置优化和先进对接设计

• 批准号: RGPIN-2014-04596
• 负责人: Melek, William
• 金额: $ 1.75万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=659513
• 关键词: Configuration; Optimization; Advanced; Docking; Design

Modular and Reconfigurable Robots (MRR) and Modular and Self-Reconfigurable Robots (MSRR) offer a great economic advantage stemming from the potential of lowering the overall cost by building complex robotic structures from few rudimentary mass-produced modules. MRR technology can provide the manufacturing and automotive industries with flexible solutions to lower cost of the capital equipment and provide the much needed reusability feature. Furthermore, the ability of MSRR systems to efficiently reconfigure its morphology by rearranging the connectivity of its modules enables robots in this class to adapt to changes in the environment which makes such systems very attractive in applications related to surveillance, autonomous exploration, and search and rescue. The goal of this research program is to advance the state-of-the-art in design and utilization of MRR and MSRR, enabling wider scale integration of such systems in applications areas such as manufacturing, exploration, and military. To do so, the applicant has identified the following research objectives for the NSERC DG program being applied for that need to be tackled in order to realize this goal: (i) Development of task-based configuration optimization (TBCO) algorithm for serial-type MRR: this algorithm will utilize a multi-solution IK solver developed by the applicant to identify the most suitable MRR configuration for a given task. In order to boost the computational performance and speed of the TBCO solvers for MRR proposed by the applicant, a novel Memetic Algorithm (MA) will be developed as the core optimization algorithm, (ii) Development of advanced autonomous docking methods and pose optimization algorithms for MSRR in formation: When several mobile MSRRs are linked via dexterous interfaces, they can perform complex tasks through decentralized collaboration. However, in order to enhance the utility of MSRR, and in particular its subcategory referred to Mobile Configuration Change (MCC) class in applications related to exploration and navigation in unstructured environments, maneuverability, dexterity, and ability to navigate on uneven surfaces has to be significantly enhanced. Hence, another key objective of the proposed research is make significant advances towards this goal through: (1) development and experimental validation of a task-based genetic algorithm-based pose optimization algorithm for MSRRs linked in chained formation by dexterous serial arms. To validate our proposed task-based pose optimization algorithm, a formation of three mobile manipulators (existing at the applicant's research lab) serially connected through their on-board mechanical manipulators will be utilized, and (2) development of a novel mechanism for autonomous docking of MSRR modules based on electromagnetic/permanent magnet male-female interfaces with passive spherical joint and an Extended Kalman Filter (EKF) estimation algorithm. The EKF is proposed for estimation of the relative position and orientation of two docking mobile modules which utilize IR sensors and wheel encoders in indoor applications. Mobile manipulators in the applicant's research lab will be retrofitted with the proposed docking mechanism and the process will be validated experimentally. For outdoor applications, the methodology will be extended by using a camera and LED markers instead of IR sensing. The applicant has a proven record, capable team, and state-of-the-art laboratory to succeed in accomplishing the above-mentioned research objectives in order to enable stakeholder researchers and industries to reap the benefits of such modular engineering systems.

模块化和可重构机器人(MRR)和模块化和自重构机器人(MSRR)提供了巨大的经济优势，这源于通过利用少量基本的大规模生产模块来构建复杂的机器人结构来降低总体成本的潜力。MRR技术可以为制造业和汽车行业提供灵活的解决方案，以降低资本设备的成本，并提供急需的可重用性特征。此外，MSRR系统通过重新安排其模块的连接性来有效地重新配置其形态的能力，使这类机器人能够适应环境的变化，这使得此类系统在与监视、自主探索和搜救相关的应用中非常有吸引力。这项研究计划的目标是推进MRR和MSRR设计和使用的最先进水平，使此类系统能够在制造、勘探和军事等应用领域进行更广泛的集成。为此，申请人为正在申请的NSERC DG计划确定了以下研究目标，为了实现这一目标需要解决这些目标：(I)为序列型MRR开发基于任务的配置优化(TBCO)算法：该算法将利用申请人开发的多解IK求解器来确定给定任务的最合适的MRR配置。为了提高本申请者提出的用于MRR的TBCO解算器的计算性能和速度，将开发一种新的Memtic算法(MA)作为核心优化算法，(Ii)开发先进的MSRR编队自主对接方法和姿态优化算法：当多个移动MSRR通过灵活的接口连接时，它们可以通过分散协作来执行复杂的任务。然而，为了增强MSRR的实用性，特别是其子类别指的是移动配置变化(MCC)类别，在与非结构化环境中的探索和导航相关的应用中，机动性、灵活性和在不平坦表面上导航的能力必须显著增强。因此，本研究的另一个关键目标是通过以下方式取得重大进展：(1)开发和实验验证基于任务的基于遗传算法的MSRR姿态优化算法。为了验证我们提出的基于任务的位姿优化算法，将使用三个移动机械手(存在于申请人的研究实验室)通过其船上的机械臂串联的编队，以及(2)开发一种基于被动球面关节的电磁/永磁式凸凹式接口和扩展卡尔曼滤波(EKF)估计算法的MSRR模块自主对接的新机制。EKF用于估计室内应用中使用红外传感器和轮盘编码器的两个对接移动模块的相对位置和方向。申请者研究实验室中的移动机械手将改装拟议的对接机构，该过程将进行实验验证。对于户外应用，该方法将扩展为使用相机和LED标记，而不是红外传感。申请人拥有经过验证的记录、有能力的团队和最先进的实验室，能够成功实现上述研究目标，使利益相关者研究人员和行业能够从这种模块化工程系统中获益。