MICROROBOTS. Richard Yeh, Ezekiel J. J. Kruglick, Kristofer S. J. Pister. University of California at Los Angeles, Dept. of Electrical Engineering, Engineering IV, ...
MICROELECTROMECHANICAL COMPONENTS FOR ARTICULATED MICROROBOTS Richard Yeh, Ezekiel J. J. Kruglick, Kristofer S. J. Pister University of California at Los Angeles, Dept. of Electrical Engineering, Engineering IV, 54-148 Los Angeles, CA 90024, USA
SUMMARY We propose to create a class of articulated micromanipulator robots with multiple degrees of freedom, workspaces on the order of a cubic millimeter, and payloads on the order of a milligram. We have created rigid links, mechanical couplings, and large-force, large-displacement micromotors. Hollow triangular beams made from rotated microhinged polysilicon plates can withstand axial loads of up to 2.3 gm. Mechanical couplings are used to rotate folded structures off the substrate with less than 2µN of force. Linear electrostatic stepper motors with an estimated force of 6.5µN at 35V and a travel of 40µm have also been demonstrated.
φ3
joint
φ2
link
φ1
end-effector
base
(a) Fig. 1.
(b)
Robot manipulator. (a) Structure of an articulated manipulator with revolute joints and mechanical links. (b) Photograph of an operational macroscopic articulated manipulator.
INTRODUCTION There are many approaches to developing microrobots [1-3]. One of the more successful examples of “microrobots” is insects, which have articulated limbs. In 1992, Suzuki, Shimoyama et. al., introduced the concept of creating insect-like microrobots with exoskeletons made from polysilicon plates, and flexible polyimide joints [2]. In 1994, we continued in that direction by proposing a class of articulated microrobots using hollow triangular beams (HTB’s) as links, microhinges [4] as revolute joints, and linear electrostatic stepper motors for actuation [5]. Figure 1a shows a diagram of our proposed articulated manipulator, which consists of robot links connected by revolute joints. Figure 1b shows a picture of a macroscopic remotecontrolled model of our manipulator with three HTB’s. Figure 2 shows a SEM picture of an HTB and a lever arm, each connected to a stepper motor.
RIGID MICROROBOT LINKS A surface micromachined link, made from thin film polysilicon, must be rigid enough to support the weight of its payload and also have the requisite shape to allow joints to have independent axes of rotation. HTB’s have been designed and tested to satisfy these conditions. Figure 3a shows a version of an HTB in CAD layout. The HTB is created with three plates connected with scissor hinges, which allow two plates to rotate with respect to one another. This allows us to use probe tip
stepper motor
HTB shuttle
lever arm
push-rod
microhinges un-assembled HTB Fig. 2.
SEM picture of 1-DOF robotic test structures.
manipulators to fold the two side plates up and over the center plate, creating a hollow triangular beam. The two side plates have snaplocks (Fig. 3b) along the top edges to fasten the plates in position. Figure 3c shows a SEM picture of HTB’s that are assembled and rotated out of the substrate.
Strength & Stiffness To find the strength of HTB’s in the axial direction, a glass box was placed on top of four 432µm tall HTB’s (Fig. 4). Water was injected slowly from a syringe into
snaplocks
side plate
center plate
side plate
scissor hinges
HTB rotated by 90o
glass box
test chip
(a)
substrate hinges
Fig. 4.
Loading test on four corner HTB’s.
center plate
substrate hinges
upper left upper right lower left lower right Fig. 5. Pictures of the remains of four corner HTB’s after collapse. (b) Fig. 3.
(c)
HTB. (a) IC CAD layout of HTB with scissor hinges, substrate hinges, snaplocks, center plate, and side plates. This version has a right triangular cross-section. (b) SEM picture of a snaplock. (c) HTB’s that are rotated from the substrate.
the glass box until the HTB’s collapsed. Videotape of the collapse indicated that when the critical load was reached, the snaplocks broke first. Next, the scissor hinges broke, freeing the side plates from the center plate. Figure 5 shows the remains of the four center plates after a loading test. All plates broke in approximately the same place, while all the substrate hinges remained intact. A load of 8.7-9.3gm was required to break the four HTB’s. If all twelve plates that make up these four HTB’s were independent, the plates would fail by buckling at a calculated total critical load of only 0.4gm. Similarly, we expect the HTB to have a higher spring constant compared to flat beams. A fixed-end flat polysilicon beam 1mm long, 100µm wide, and 2µm thick has a spring constant of approximately 0.04 N/m under small deflections. However, if we could create an HTB with rigid corners, the spring constant would be increased to 1,400 N/m. Since in reality, the HTB does not have rigid corners due to snaplocks and scissor hinges, we do not expect the overall stiffness of the HTB to be as high as an HTB with rigid corners.
Articulated Joints and Multiple DOF The triangular cross-section of an HTB simplifies the design of multiple degree of freedom (DOF) structures. Each end of the HTB has three edges available for the placement of microhinge joints. By placing a joint at the correct edges in a series of HTB’s, we can create a manipulator with multiple DOF.
Since scissor hinges are used as revolute joints, their maximum angles of rotation affect the workspace of our manipulator. Various versions of microhinge designs gave measured maximum angles of rotation ranging from 111o to 146o. Deflection past the maximum angles of rotation is still possible as the two polysilicon plates connected by the scissor hinges bend elastically. Preliminary tests indicate that applying a force of approximately 1,000µN causes the hinges to fracture.
FRICTION Mechanical couplings transmit the actuator force to the HTB’s. By using a sliding shuttle and push-rod, we have manually rotated structures out of the plane of the substrate. Figure 6 shows push-rods and shuttles coupled to 2 DOF lever arms. Each lever arm has its own pushrod and shuttle and can be manually rotated by pushing or pulling a ring at the end of the shuttle. Figure 7 shows an HTB manually rotated out of the substrate by a push-rod and shuttle. To use these push-rods, shuttles, and lever arms as mechanical couplings, an actuator must be able to overcome the friction caused by the moving parts. The mass of our mechanical coupling test structures are estimated to be on the order of a few µgm (a few nN). For such light masses, the friction is no longer dependent on the mass of the structures but on the adhesive forces between the surfaces [6]. Preliminary tests on friction requirements were conducted on the following test structures: • shuttle • shuttle + push-rod • shuttle + push-rod + lever arm • shuttle + push-rod + HTB (Fig. 7) • shuttle + push-rod + lever arm + HTB (Fig. 8). Applied forces were visually measured by the deflection of a serpentine spring attached to the shuttles. To prevent
Fig. 6.
SEM picture of 2 DOF lever arms with push-rod and shuttle.
surface liquid adsorption, all samples were dehydrated in an oven at 120oC for 45 minutes. The maximum static friction for all the different test structures was less than 2µN (up to 22µN observed for unbaked samples) despite the differences in mass and the angles of rotation of the coupled structure.
Fig. 7.
SEM picture of an HTB rotated with a push-rod and shuttle.
Fig. 8.
SEM picture of a lever arm coupled to an HTB for mechanical coupling.
ACTUATION To meet the requirements of large motion, large force, and low power consumption, we have developed gap-closing electrostatic actuators. Electrostatic actuators consume less power compared to thermal actuators, and can be fabricated from CMOS materials. However, the typical electrostatic forces and displacements available with CMOS compatible voltages are small. To boost the force generated from electrostatic actuators and at the same time be able to produce large displacements, we have designed a stepper motor that uses gap-closing actuators in an attachment/detachment stepping scheme.
Gap-Closing Electrostatic Actuators The electrostatic force between two gap-closing beams is: 1 2 lt F e = --- εV ( α ) ----2- d 2
where α is a fringe field factor, roughly 2 for this geometry [7], ε is the dielectric permittivity, V is the applied voltage, l is the overlap length of the beam, t is the beam thickness, and d is the gap between the two parallel capacitive plates (Fig. 9). Due to Paschen’s Law, we can generate large electric fields without breakdown [8] by operating with small gaps. A parallel array of 10 actuators with beams that are 100µm long, 2µm thick, and separated by an initial gap of 2µm can generate a force of 5.4 µN at 35 V. If 10% of the force is used to overcome the spring attached to the rotor beam, then the actuator array will produce almost 5µN of force. The force-to-area ratio of the actuator array is 0.2mN/mm2 with 35 V.
Linear Electrostatic Stepper Motors Although the gap-closing actuator can generate high forces with small gaps, its range of motion is limited to that gap. To increase the range of motion, we have employed the actuators in the stepper motor with an attachment/detachment stepping cycle [5]. One version of a stepper motor employs two actuator arrays and a shoe on each side of a shuttle. The actuator arrays provide the shoes with bi-directional motion with a travel equal to the actuator gap in each direction. The shoes, located on each side of the shuttle, attach and detach from the shuttle electrostatically. The stepper motor cycle begins with the left shoe attached to the shuttle. An actuator array connected to the left shoe closes its gap and pulls the shoe and the attached shuttle by 2µm. Next, the right shoe attaches to the shuttle and its actuator array holds the attached shuttle at its position. Then the left shoe detaches from the shuttle and is returned to its initial position by attached
springs. The actuator of the right shoe then closes its gap to advance the shuttle by another 2µm. Finally, the left shoe re-attaches itself to the shuttle and the cycle repeats. In a stepping cycle, the shuttle is advanced (or reversed) by two times the actuator gap. Figure 10 shows a SEM picture of a stepper motor that displaced the shuttle by 40µm in 10 stepper cycles before breaking one of its anchoring springs. Based on actuator geometry, the force at 35V is calculated to be 6.8µN. By measuring the deflection of the beams supporting the shuttle, the force generated is estimated to be at least 6.5µN.
stator beam rotor beam overlap length
spring
gap Fig. 9.
Diagram of a gap-closing electrostatic actuator. When sufficient voltage is applied, the electrostatic force will close the gap.
CONCLUSION Working components for microrobots have been demonstrated. We have created rigid modular links that can withstand up to 2.3 gm of weight before buckling. These links can be connected with microhinges as revolute joints to achieve multiple DOF. 1-DOF mechanical couplings for the links have been demonstrated and frictional forces were measured. A linear electrostatic stepper motor with gap-closing actuators could be used to actuate our microrobots. We have demonstrated a motor that can generate 6.5µN of force with 35V and displace a shuttle by 40µm. The friction force was found to be less than the stepper motor force. Stepper motors connected to rotating structures are shown in figure 2. These motors should be able to actuate the articulated manipulator microrobot.
ACKNOWLEDGEMENT The authors would like to thank Kevin Dang and Richard Liu for building 3-D robot models. Also thanks to Floy Chang, Henry Wong, and Gisela Lin for their valuable assistance. All structures shown were fabricated by ARPA-supported MUMPs process at MCNC. This work was supported in part by NSF and ARPA under award IRI-9321718. Layout and SEM pictures are available through ftp and http from synergy.icsl.ucla.edu.
REFERENCES [1] [2]
[3]
[4]
Smits, J.G., Sensors and Actuators A, Vol. 35, (1992), pp. 129-135. Suzuki, K.; Shimoyama, I.; Miura, H.; and Ezura, Y., IEEE Workshop on Micro Electro Mechanical Systems, MEMS ’92 (Travemunde, Germany, Feb. 4-7, 1992), pp. 190-195. Takeshima, N.; and Fujita, H., ASME Winter Annual Meeting, Micromechanical Sensors, Actuators, and Systems (Atlanta, Georgia, Dec. 16, 1991), Vol. 32, pp. 203-209. Burgett, S.R.; Pister, K.S.J.; and Fearing, R.S., ASME Winter Annual Meeting, Dynamic Systems and Control, Micromechanical Systems, (Anaheim,
actuator array
shuttle
shoe
Fig. 10. SEM picture of a stepper motor with 4 actuator arrays (only 2 bonded), two shoes, and one shuttle. [5]
[6]
[7]
[8]
California, Nov. 8-13, 1992), pp. 1-11. Yeh, R.; Kruglick, E.J.J.; and Pister, K.S.J., ASME Winter Annual Meeting, Micromechanical Systems (Chicago, Illinois, 1994), DSC-vol 55-2, pp. 747-754. Kaneko, R., Proceedings, IEEE Workshop on Micro Electro Mechanical Systems (Nara, Japan, Jan. 30-Feb. 2, 1991), pp. 1-14. Boyd., M.R., Crary, S.B., and Giles, M.D., SolidState Sensor and Actuator Workshop (Hilton Head, South Carolina, Jun. 13-16, 1994), pp. 123-126. Gray, T.S., Applied Electronics; a first course in electronics, electron tubes, and associated circuits, 2nd edition, The MIT Press, Cambridge, Massachusetts, 1995, pp. 153-155.
