Advanced sensing and control for coordinated manipulation

提出的研究旨在发展对工业伺服机器设计的机电方法的基本理解，特别强调传感器和传感。在机器人和装配机等工业伺服系统的设计中，为了降低成本和效率，最近的趋势是使它们更小、更轻，这导致结构灵活性和动态不确定性的增加。此外，以合作方式操作两台或多台机器作为以更低成本完成更复杂任务的解决方案正受到越来越多的关注。这些趋势给伺服控制带来了挑战，特别是在传统机器的有限传感下，通常只依赖于电机位置信号。提出的研究试图通过利用现代传感器技术以及MEMS（微机电系统），DSP（数字信号处理）和无线通信等使能技术来提高传统方法的性能。本研究将研究在传统实践中未使用的驱动系统和相关伺服机器中的各种信号的传感，包括加速度，关节力和末端执行器运动。本研究的技术目标是：(1)开发一种富传感传动系统的通用架构，作为工业伺服系统的新范例；(2)开发利用传感信号提高性能的信号处理和决策算法；(3)通过包含伺服机器之间的传感信号通信，使这些算法适应多伺服机器的协调控制。基本方法将是采用多个低成本传感器，并通过仔细分析每个传感器的优势和局限性来探索可以实现的协同作用。这项研究产生的技术价值将对从事柔性自动化、装配和物料搬运的行业产生广泛的影响。此外，来自本研究的理论收益可能会带来我们设计和控制伺服系统的方式的范式转变。

The proposed research is aimed at developing a fundamental understanding of mechatronic approaches to the design of industrial servo machines, with particular emphasis on sensors and sensing. In the design of industrial servo systems such as robots and assembly machines, a recent trend is to make them smaller and lighter for cost reduction and efficiency, which leads to an increase in structural flexibility and dynamic uncertainty. Also, operating two or more machines in a cooperative manner is receiving more attention as a solution to accomplish more complex tasks at reduced cost. These trends are imposing challenges on servo control, especially under the limited sensing of conventional machines that typically rely on the motor position signal only. The proposed research attempts to improve the performance of conventional approaches by exploiting modern sensor technologies along with enabling technologies such as MEMS (Micro Electro Mechanical Systems), DSP (Digital Signal Processing) and wireless communication. The research will study sensing of various signals in the drive train and associated servo machines that are not utilized in conventional practice, including acceleration, joint force and end-effector motion. The technical objectives of this research are 1) to develop a general architecture of a sensing rich drive train as a new paradigm for industrial servo systems, 2) to develop signal processing and decision making algorithms that take advantages of the sensed signals for performance improvement, and 3) to accommodate these algorithms to the coordinated control of multiple servo machines by encompassing the communication of sensed signals between servo machines. The basic approach will be to employ multiple low cost sensors and to explore the synergy that can be achieved through careful analysis of advantages and limitations of each sensor. The technical value resulting from this research will generate a broad impact on industries engaged in flexible automation, assembly and material handling. Also, the theoretical benefits coming from this research may bring a paradigm shift in the way we design and control servo systems.