Robust Electromagnetic Control of Microrobots Under Force and Localization Uncertainties

Abstract

Abstract-Microrobots are promising tools for micromanipulation and minimally invasive interventions. Robust electromagnetic control of microrobots can be achieved through precisely modeled magnetic steering systems and accurate localization. Error-free modeling and position information, however, are not realistic assumptions, and microrobots need to be controlled under force and localization uncertainties. In this paper, methods to account for these types of uncertainties are presented. Initially, the uncertainties in electromagnetic force generation of a new class of manipulation systems are quantified. Subsequently, a drag-force uncertainty model for linear dynamics is proposed. This model can be employed for microrobots whose fluid dynamics are not well understood. A set of performance measures is introduced in the design of controllers, and a PID and a robust H∞ controller are synthesized and evaluated through simulations. To demonstrate the capabilities of the synthesized controllers under localization and force uncertainties, low update rates are considered. The H∞ controller can provably respect the performance measures under higher uncertainties than the PID controller, and its performance is further quantified through experiments in a prototype electromagnetic control system. Note to practitioners: This paper is motivated by the problem of robust magnetic microrobot control for in vivo medical applications and micromanipulation. Existing approaches do not account for navigation system uncertainty, and rely on high imaging rates that are unrealistic for some applications (e.g., involving x-rays). We investigate the uncertainties in magnetic force generation and fluidic forces for microdevices that move in viscous fluid, considering state-of-the-art electromagnetic control systems capable of 5 degree-of-freedom control. These systems have not been examined before from this perspective. We demonstrate how to estimate the uncertainty in electromagnetic force generation due to localization and calibration inaccuracies, as well as the uncertainty in drag force due to microrobot shape complexity. Finally, we show that we can robustly control permanent magnetic microrobots in low viscosity oils, thus enabling more accurate and less restricted manipulation.

Navigation of a rolling microrobot in cluttered environments for automated crystal harvesting

“…In [19], the authors used H ∞ control to achieve 3D motion control for a magnetic body actuated with magnetic field gradients. This approach assumes linear drag forces in low Reynolds number fluid regimes and a fully-actuated system and is therefore not applicable to nonholonomic microrobots.…”

Uncalibrated Visual Servo Control of Magnetically Actuated Microrobots in a Fluid Environment

“…In [14] Marino et al compare an H∞ controller with a PID controller for a linear uncertain dynamical model for electromagnetic steering control of a 1000 μm microrobot in low viscosity oil using the OctoMag system [15]. Tracking errors on the order of 270-490 μm are reported using the H∞ controller.…”

A robust controller for micro-sized agents: The prescribed performance approach

“…On the contrary, closed-loop control can deal with these perturbations effectively. For microrobotic systems, different closed-loop techniques such as linear control [5], adaptive control [9], H ∞ control [10] and model predictive control [11] have been investigated. For (standard) linear controllers, problems such as instabilities and undesired oscillations have been reported [12], [13].…”