Collaborative Research: Single-Input Control of Large Microrobot Swarms using Serial Addressing for Microassembly and Biomedical Applications

这个合作研究项目将为大群微型机器人创造一个实用的控制方案。这些机器人通常不超过几毫米长，并依赖于外部电源和控制信号。目前，有可能将群体作为一个整体引导到一个目的地（或者可能是一个期望的平均位置）。然而，要实现微型机器人群的全部潜在好处，需要能够同时向单个机器人或小型子组发送独立命令。 先前已经探索了响应于不同命令幅度的设备设计，但是随着可独立寻址的机器人数量的增长，这种方法很快变得不切实际。这个可扩展性问题可以通过串行寻址方案来克服。这里，对于控制信号只有几个不同的值。每个可独立寻址的机器人子集都与唯一的信号值序列相关联，并且只有当控制信号包含该特定序列时才会改变其行为。本项目考虑在实施这一计划时出现的两个基本问题。首先是需要板载计算和存储器，允许机器人识别独特的序列，并根据检测到的序列改变机器人的状态。第二是需要一个推进机制，耦合到机器人的状态，以允许差分引导朝向目标配置。该项目将推进两个创新的工程平台，以满足这两个需求。第一种是静电驱动的，在平面衬底上操作，适合于结构化任务，如微组装。第二种是磁致动的，在液体体积中操作，并且适用于生物医学应用，例如药物输送。该项目的技术方面得到了外联活动的补充，包括将在IEEE机器人和自动化国际会议上举行的年度微型机器人移动竞赛，该会议是学术界和工业界的首要机器人会议。该项目的成果将通过为微型机器人群提供新的诊断和治疗用途来增强国民健康。他们还将促进国家的繁荣，使新的类别的微装配机器人。本项目旨在开发一个实用的控制方案，同时控制大量的微型机器人。这将通过使用微机电系统（MEMS）来机电地和磁性地解码嵌入在单个全局控制信号中的序列，并且将这种序列的重新配置耦合到个体微型机器人轨迹的修改来实现。这种机载序列解码将通过机载基于物理的有限状态机（PFSM）来完成，PFSM可以接受嵌入在控制信号中的控制序列，并相应地改变微型机器人的行为。该项目将使用静电和磁的方法来实现PFSM，并将其“接受”状态耦合到推进机制，以调节单独的轨迹。应力工程静电开关，这将锁存响应于一个预编程的控制电压序列，将被用来实现PFSM上的静电平台。电永磁电路将用于在磁性平台上实现PFSM，该电路响应于一系列全局磁场而改变其磁矩。该项目将开发基于PFSM的多微型机器人控制理论，构建静电和磁微型PFSM平台，并通过对静电和磁动力微型机器人群实施基于PFSM的控制来验证概念。开发的理论和方法将铺平道路的控制大型微型机器人群众多的生物医学和微装配应用。这个奖项反映了NSF的法定使命，并已被认为是值得通过评估使用基金会的智力价值和更广泛的影响审查标准的支持。

This collaborative research project will create a practical control scheme for large swarms of microrobots. These robots are typically no more than a few millimeters in length, and rely on an external power source and control signal. Currently, it is possible to steer the swarm as a whole to a single destination (or perhaps, to a desired average location). However, realizing the full potential benefits of microrobot swarms will require the ability to simultaneously send independent commands, either to individual robots or to small subgroups. Device designs have previously been explored that respond to different command amplitudes, however this approach quickly becomes impractical as the number of independently addressable robots grows. This scalability problem can be overcome using serial addressing schemes. Here, there are only a few distinct values for the control signal. Each independently addressable subset of robots is associated with a unique sequence of signal values, and will change its behavior only if the control signal contains that specific sequence. This project considers two fundamental issues that arise in implementing such a scheme. First is the need for on-board computation and memory allowing the robots to recognize the unique sequence and to change the robots state based on the detection of such sequence. Second is the need for a propulsive mechanism that couples to the robot state to allow differential guidance towards a target configuration. This project will advance two innovative engineering platforms that meet both needs. The first is electrostatically actuated, operates on a planar substrate, and is suitable for structured tasks such as microassembly. The second is magnetically actuated, operates in a liquid volume, and is suitable for biomedical applications such as drug delivery. The technical aspects of the project are complimented by outreach activities, including an annual microrobotics mobility competition to be held at the IEEE International Conference on Robotics and Automation -- a premier robotics conference for academia and industry. The results from this project will enhance the national health, by enabling new diagnostic and therapeutic uses for microrobot swarms. They will also promote the national prosperity, by enabling new classes of microassembly robots.This project aims to develop a practical control scheme to simultaneously control large numbers of microrobots. This will be achieved by using microelectromechanical systems (MEMS) to electromechanically and magnetically decode a sequence embedded in the single global control signal, and couple the reconfiguration of such sequence to the modification of the individual microrobot trajectories. This on-board sequence decoding will be accomplished through sets of on-board physics-based finite state machines (PFSM) that can accept a control sequence embedded in the control signal and change the behavior of the microrobots accordingly. The project will use both electrostatic and magnetic approaches to implement PFSMs, and to couple their "accept" state to the propulsion mechanism to modulate individual trajectories. Sets of stress-engineered electrostatic switches, which will latch in response to a pre-programmed control voltage sequence, will be used to implement PFSM on the electrostatic platform. Electro-permanent magnetic circuits, which change their magnetic moment in response to a sequence of global magnetic field, will be used to implement PFSM on the magnetic platform. The project will develop the theory for PFSM-based multi-microrobot control, construct both electrostatic and magnetic microrobotic PFSM platforms, and validate the concept by implementing the PFSM-based control on swarms of electrostatically and magnetically powered microrobots. The developed theory and approach will pave way for control of large microrobot swarms for numerous biomedical and microassembly applications.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.