Ashby J et al. (2022), One Soft Step: Bio-Inspired Artificial Muscle M...

ECB-ART-50531

Front Robot AI 2022 Jan 06;8:792831. doi: 10.3389/frobt.2021.792831.

Show Gene links Show Anatomy links

One Soft Step: Bio-Inspired Artificial Muscle Mechanisms for Space Applications.

Ashby J , Rosset S , Henke EM , Anderson IA .

Abstract
Soft robots, devices with deformable bodies and powered by soft actuators, may fill a hitherto unexplored niche in outer space. All space-bound payloads are heavily limited in terms of mass and volume, due to the cost of launch and the size of spacecraft. Being constructed from stretchable materials allows many possibilities for compacting soft robots for launch and later deploying into a much larger volume, through folding, rolling, and inflation. This morphability can also be beneficial for adapting to operation in different environments, providing versatility, and robustness. To be truly soft, a robot must be powered by soft actuators. Dielectric elastomer transducers (DETs) offer many advantages as artificial muscles. They are lightweight, have a high work density, and are capable of artificial proprioception. Taking inspiration from nature, in particular the starfish podia, we present here bio-inspired inflatable DET actuators powering low-mass robots capable of performing complex motion that can be compacted to a fraction of their operating size.

PubMed ID: 35096985
PMC ID: PMC8793852
Article link: Front Robot AI

Article Images: [+] show captions

	FIGURE 1. Left: Basic structure of the podium (tube foot) of the common starfish, Asterias Rubens, showing the muscles used for contracting and deforming the ampulla and podium. Right: A starfish righting itself on the sea floor, with close-up of the leg showing many podia in actuation [photographs by Iain Anderson].
	FIGURE 2. Operating principle of a Dielectric Elastomer Actuator. A is the area of the electrodes and z is the thickness of the dielectric layer, prior to actuation. dA represents the increase in area and dz is the decrease in thickness, after actuating. Applied voltage V is typically on the order of 3–5 kV for a membrane thickness of 0.1 mm.
	FIGURE 3. Top: The top electrodes of a MIDA model, showing the 4 segmented actuators and an electro-adhesive end effector. Left: The structure of the MIDA artificial tube foot actuator, inflated and undergoing actuation. Right: The latest prototype MIDA, inverted. Black line is 1 cm in length. See Supplemental Materials for a video of the actuator in motion.
	FIGURE 4. Structure of the MIDA test setup in the deflated state, red lines indicate the position of the edges of acrylic support frames. Top Left: Top view, upper membrane. Top Right: Top view, lower membrane. Bottom Left: Bottom view, upper membrane. Bottom Right: Bottom view, lower membrane. Small holes show bolt placement, air channel visible on Top Right and Bottom Left.
	FIGURE 5. Comparative diagram. (Ai–iv) shows the basic structure and motions of the tube foot of Asterias Rubens the common starfish, whilst (Bi–iv) shows the equivalent in our inflated actuator. Sections in red/solid are active muscles and DEAs, blue/striped are inactive, you can see the operation is inverted between the contracting biological muscle and our expanding artificial muscles. The operations of the ampulla and artificial ampulla are shown in figures (Ai–ii,Bi–ii) respectively, the arrows indicate the water/air flow.
	FIGURE 6. PILA, the phased inflatable locomotory actuator. (A)—a series of MIDA actuators in a line, the structure PILA is intended to emulate. The crosses and diamonds on (A–C) show points that travel along equivalent paths (see Figure 14). (B)—PILA in a deflated state, showing the relative size and position of the electrodes and a representative cross section showing the regions of different thickness (not to scale). (C)—the basic layout of PILA when inflated showing a cross section through the central axis, DETs are highlighted in various colours to show level of actuation at this point in the cycle. (D)—Finite element model of PILA undergoing a standard walking cycle using 3 kV peak to peak square wave (with DC offset 3.5 kV) actuation for the “feet” electrodes and 5 kV peak to peak (with 2.5 kV DC offset) sine wave actuation for the displacement electrodes. (E)—Physical prototype PILA. Black line is 1 cm in length. See Supplemental Materials for a video of the actuator in motion.
	FIGURE 7. Structure of the PILA test setup, red lines indicate the position of the edges of acrylic support frames. Top: Top view. Bottom: Bottom view. Small holes show bolt placement, large central hole shows air tube placement.
	FIGURE 8. Production of the PILA actuator. Starting at stage I with the silicone base layer (Wacker Elastosil®2030), this is plasma treated (II) to activate the surface for bonding with a similarly tre’ated second layer to produce the different thickness sections (III). A PET mask is used (IV) to selectively spray-coat the conductive electrode mixture onto the active areas (V). This process is then repeated on the other side for the ground electrodes (VII). Finally a thin layer of SEBS-g-MA (maleic anhydride grafted styrene-ethylene-butylene-styrene) co-block polymer is applied to reduce gas permeability.
	FIGURE 9. Finite element modelling of the actuation amplitude of the MIDA ampulla at varying pressures and applied voltages.
	FIGURE 10. Left: FEM simulated dome height against internal pressure, linear fit. Right, experimental set-up showing the dome height measurement, tracking marker on the end effector, and the laser displacement sensor head in the rear.
	FIGURE 11. FEM Comparison. Left: MIDA FEM model showing the voltages applied to the segments during a ‘circular’ actuation, I - VII show the progression of the cycle. Right: A comparison of simulated (circular markers) versus experimental data (square markers) during actuation at 5 kV, with 90° phase staggering between adjacent electrodes. Data is shown for 3 cycles.
	FIGURE 12. FEM Comparison 2. Simulated horizontal actuation (simultaneous actuation of two adjacent MIDA segments) under a 3 kV peak to peak sinusoidal load with 1.5 kV DC offset at varying pressure (blue) vs experimental actuation under the same conditions (orange) measure using the laser displacement sensor.
	FIGURE 13. A: Force produced by MIDA ampulla actuated by a 3 kV square wave at 0.5 Hz. B: Force produced by MIDA ampulla actuated by a 5 kV square wave at 0.5 Hz.
	FIGURE 14. Path taken by the end point of each foot. (A I–VI) shows the actuation of the DEAs in a “foot” during the cycle. Red segments are actuated, blue segments are relaxed. (B)—FEM data of the expected tip displacement during four cycles using a 3 kV square wave with an offset of 2 kV applied to the “feet” and a 5 kV peak-to-peak sinusoidal wave with a 2.5 kV DC offset applied to the smaller electrodes. I-VI show the progression of the cycle on each diagram.

References [+] :

Brochu, Advances in dielectric elastomers for actuators and artificial muscles. 2010, Pubmed