Untethered Resonant Magnetic Microrobots

One of the principle challenges of building mobile microrobots is the design of the propulsion system. Due to difficulties in storing energy at these scales, mechanisms are needed for harvesting energy from the environment for wireless applications. This project entails using the interactive forces between small magnetic bodies in a uniform magnetic field to drive the mechanism to resonance. This energy is then used to move the robot through its environment. These small microrobots are capable of being driven forward, backward, and turning in place.

Methods for transmitting power wirelessly to devices at this scale are being actively researched around the world. Current wireless communication or power transmission methods typically involve inductive coupling between antennas. The efficiency of this method decreases as the area of the coils involved shrinks, making inductive coupling a less attractive method for transmitting power to sub-mm microrobots. In addition to this power loss during transmission, such systems require circuitry and electronics to drive the actuators. This leads to further inefficiencies. Our method of bypassing these problems is to transmit the energy directly to a mechanical structure and use the structure to power the actuator. By designing actuators that rely on frequency components of the controlling field, several actuators can be operated independently in the same field or on the same robot.

The actuator relies on the interactions between soft magnetic materials in a uniform field. The magnetic microrobot consists to two rectangular nickel bodies. One is connected to a gold frame that rests on the substrate below, suspended on small stiction reducing dimples. Attached to the frame, but elevated 5μm above the level of the substrate is a meander spring that supports the second nickel body which is also elevated above the substrate. When placed in a uniform magnetic field, the robot aligns the combined long axis of the nickel bodies with the field due to shape anisotropies.

The magnetic field also creates an attractive force between the two bodies, causing the spring to deflect and the gap between them to narrow. When the field is turned off, the attractive forces decrease to the level of those generated by the magnetic remanence of the nickel which is significantly lower. When excited with a low amplitude oscillating magnetic field the system can be driven to resonate at its resonant frequency which depends on the spring constant and weight of the oscillating mass. When the amplitude of the driving field is increased the amplitude of the oscillations increase until impact occurs between the two bodies. During this impact, under the correct operating conditions, the second body transfers its momentum to the first body and reverses its motion to begin the oscillatory cycle again.

If the device was actuated on a frictionless surface it would simply vibrate in place without any exhibiting any controllable motion. To generate motion in a particular direction, a method for creating a friction difference between the two directions of the stroke of the resonant body is required. This is currently achieved by driving the robots on an engineered substrate, where the unidirectional force is caused by a phase-locked electrostatic clamping signal. For prototyping purposes, we use electrostatic clamping to the substrate through the use of electrodes in the substrate and charge separation. By applying a magnetic field with a square wave signal at a desired frequency and phase offset, the frictional forces are increased when the oscillatory body begins to travel forward and released before impact. This allows the device to slide forward before clamping again when the next cycle starts. The ability to control the clamping waveform also allows for a convenient velocity control mechanism. The effective velocity of the microrobot can be controlled by releasing the clamping for differing duty cycles similar to a pulse-width modulation controller.

Additional experiments have also demonstrated the devices ability to move on a non-engineered substrate, due to inhomogeneities in the part of the device that contacts the substrate. Current research is investigating these differences so that new devices can be engineered to drive on a variety of surfaces.

Images