Swallowable and morphable robotic device

Abstract

A swallowable robotic device may include a first cap and a second cap. The robotic device may further include an origami shell having a Kresling tower structure configured to compress from an expanded state to a contracted state in response to a torsional force. A first end of the origami shell is attached to the first cap and a second end of the origami shell is attached to the second cap. An actuator disposed within the origami shell may twist the first cap and cause the first end of the origami shell to collapse toward the second end and thereby vary an effective volume in the origami shell.

Description

PATENT Atty. Dkt. No. 70663-02SWALLOWABLE AND MORPHABLE ROBOTIC DEVICEGOVERNMENT RIGHTS

[0001] This invention was made with government support under 2018-67007-28439 and 2019-67021-28990 awarded by the United States Department of Agriculture. The government has certain rights in the invention.CROSS REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U. S. Provisional Application No.63 / 725,362 filed November 26, 2024, the entirety of which is incorporated by reference herein.TECHNICAL FIELD

[0003] This disclosure relates to robotic devices and, in particular, to submersible robotic devices.BACKGROUND

[0004] The exploration of biomedical environments through robotic systems holds transformative potential for medical diagnosis, monitoring, and intervention. However, the complex nature of these environments — characterized by fluid-filled, unstructured internal organs with particle-rich media — presents unique challenges that push the boundaries of conventional robotics. While underwater robots offer promising capabilities for navigating such environments, their successful deployment in biomedical applications demands a fundamental rethinking of locomotion strategies to ensure both safety and sustained operation.

[0005] Traditional underwater robots, designed for large-scale aquatic environments such as oceans and lakes, rely on mechanical actuators such as propellers and jets to achieve precise six degrees of freedom (DOF) control. WhilePATENT Atty. Dkt. No. 70663-02 effective in open water, these systems are unsuitable for biomedical applications due to several critical limitations. First, their substantial power requirements necessitate large batteries or external power sources, which are impractical for robots operating at the sub-meter scale within the human and animal body. Second, the mechanical components are susceptible to clogging or malfunction in particle-rich biological fluids, and their exposed moving parts pose significant risks to delicate internal tissues.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The embodiments may be better understood with reference to the following drawings and description. The components in the figures are not necessarily to scale. Moreover, in the figures, like-referenced numerals designate corresponding parts throughout the different views.

[0007] FIG. 1 illustrates an example of a robotic device.

[0008] FIG. 2A-D illustrates a second example of the robotic device showing various dimensions and angles.

[0009] FIG. 3 illustrates an exploded view of a top cap.

[0010] FIG. 4 illustrates an exploded view of a top cap without an outer shell.

[0011] FIG. 5 illustrates an exploded view of a bottom cap.

[0012] FIG. 6 illustrates an exploded view of a bottom cap without an other shell.

[0013] FIG. 7 illustrates an example of an actuator for a robotic device.

[0014] FIG. 8 illustrates an example of a worm drive connecting a motor to a spinning wheel.

[0015] FIG. 9 illustrates steps for fabrication and assembly of the robotic device according to some embodiments.

[0016] FIG. 10A-B illustrates charts showing energy conception of a robotic device during various states of operation according to some embodiments.

[0017] FIG. 11 illustrates a demonstration of deploying a robotic device into a stratified rumen environment.PATENT Atty. Dkt. No. 70663-02 DETAILED DESCRIPTION

[0018] Particle-rich environments present significant challenges for traditional underwater actuation mechanisms such as propellers and jet orifices. These mechanisms expose their actuators to the environment and rely on relatively high speeds to generate propulsion, making them susceptible to clogging and failure due to particle interactions.

[0019] Previous attempts at harnessing buoyancy for robotics locomotion have relied on energy-intensive mechanisms such as compression tanks and pumps, or employ bulky hardware such as proton exchange membranes (PEMs)and electrolyzers, which are impractical for in-vivo applications due to their size, complexity, and safety concerns. The displacement of fluid and reactants in both of these approaches wastes energy while also limiting the lifespan of the robot.

[0020] The robotic deice herein has a morphable robot structure (origami-inspired) that can be implanted in the rumen of livestock, such as cows, to monitor their health and methane output in real time. The robotic device employs an encapsulated design that adjusts buoyancy through internal shape transformation, without expelling or ingesting fluids. This self-contained mechanism minimizes the risk of clogging in particle-rich environments and reduces the potential for harm to surrounding tissues. Moreover, the bi-stable nature of the Kresling pattern enables the robot to maintain a stable configuration with zero energy consumption, which is critical for long-term deployment in biomedical settings where power resources are limited. Its particleresistant, soft encapsulated structure allows it to navigate the dense, heterogeneous layers of the rumen without causing harm or disruption. The energy-efficient buoyancybased locomotion enables sustained operation and data collection across different strata, providing unprecedented insights into the rumen’s internal processes. By facilitating precise vertical movement and stable positioning, the robotic device can perform detailed spatial and temporal sampling, advancing our understanding of this complex biological system.

[0021] Alternatively or in addition, a technical advantage is that the robotic device has a swallowable and morphing structure. The structure uses origami folds to morph its shape so that it can be swallowed by the animals. This is a major leap for technology as it eliminates the need for cannulation of animals to study rumen dynamics. ThePATENT Atty. Dkt. No. 70663-02 morphing of its shape also allows the robot to change its density so that it can traverse the depth of the rumen, being the first such machine to be able to do so.

[0022] Another technical advantage is that the robotic device has buoyancy-based locomotion. The morphing shape of the robot allows it to change its density and the distribution of buoyant forces. This enables the robot to move in the rumen, capturing the nuances of the various layers for the very first time. Moreover, with this approach, we ensure that the robot also utilizes the contracting motions of the rumen to achieve full 3D locomotion capabilities.

[0023] FIG. 1 illustrates an example of a robotic device 100. The robotic device 100 may include a first cap 102 and a second cap 104. The robotic device 100 may further include an origami shell 106 having a first end and a second end. A first end of the origami shell is attached to the first cap 102 and a second end attached to the second cap 104.

[0024] The origami shell 106 may have a Kresling tower structure configured morph between a fully expanded state, a fully contracted state, or a partially expanded state in response to torsional forces applied the ends of the origami shell. Thus, the origami shell 106 may expand and contract along a longitudinal direction L oriented with respect to the caps. Since the ends of the origami shell 106 are attached to the caps of the robotic device, one or both of the caps may be twisted to cause expansion / contraction of the robotic device 100. When the robotic device 100 is placed in fluid, expansion of the shell 106 increases the volume of surrounding fluid that is displaced, thereby increasing buoyancy. Along the same lines, contraction of the shell 106 decreases buoyancy of the robotic device 100.

[0025] The robotic device 100 may include an actuator 108. The actuator 108 may be engaged to twist the first cap 102 and / or origami shell and cause the ends of the origami shell 106 to collapse toward or expand away each other. For example, the actuator 108 may be partially or completely disposed within the origami shell 106. The actuator 108 may be connected to the first cap 102 and twist the cap, which causes the origami shell 106 to collapse toward the second cap 104, thereby varying an effective volume in the robotic device and causing the robotic device to become less buoyant.

[0026] FIG. 2A-D illustrates a second example of the robotic device showing various dimensions and angles. FIG. 2A illustrates a sideview of the robotic device,PATENT Atty. Dkt. No. 70663-02 showing the top cap 102, the bottom cap 104, the origami shell 106 and the actuator 108. FIG. 2B shows a sideview of the robotic device 100, showing cables 202 connecting the actuator 108 to the top cap 102. FIG. 2C illustrates a sideview of the robotic device 100, showing the Kresling tower structure. FIG. 2D illustrates a top view of the robotic device 100 in a compressed configuration.

[0027] The origami shell 106 is a Kresling tower structure. Constructed from triangular units linked in repeating arrays, the Kresling tower can rotate at its joints, as depicted in FIG. 2A-D. This design forms a cylindrical structure with bellows-like capacity to contract and expand along the longitudinal reference line L. Each triangle in the array includes one valley fold and one mountain fold, contributing to the robustness and flexibility of the structure. This geometric configuration enhances the structural integrity and adaptability of the robot, enabling it to operate effectively in challenging environments.

[0028] The Kresling tower is modeled as a left-handed helix, with structural parameters defined by the number of sides n, the lengths of the sides at the base and top polygons (denoted as a and b, respectively), and the angles between adjacent sides that describe the tower’s configuration. Due to the non-rigidly foldable nature of the Kresling tower, we make assumptions regarding its mechanical behavior during deployment. Specifically, assume that the mountain folds (solid lines in FIG. 2C (e.g., crease AA') remain rigid, while deformation occurs along the valley folds (dashed lines, e.g., crease A'B). This selective flexibility allows the Kresling tower to achieve controlled compression through torsional forces.

[0029] The shape morphing is achieved by torsional forces acting on the Kresling tower’s base end (or top end, as the case may be) to compress its height. The kinematic relationship between the torsion angle ( ) and the corresponding height (h) during compression is mathematically represented as:> / 7>2 _ ij2 <7(7?.) = 2 sin1-: - - Eq. 12r

[0030] where r is the constant radius of the tower. The volume of Kresling tower, Vk, can be simplified as the volume of an oblique polygonal prism subtracting the volume of n pyramids:PATENT Atty. Dkt. No. 70663-0214 = An - ri X VtEq. 2Vrn = / Sdh = nr2sin( — )co$i — )h Eq. 3Jo ''n;;

[0031] Here, n is the number of sides (we use n = 6 in our design). Vmrepresents the volume of the oblique polygonal prism, and S is the area of the radial cross-section of the polygonal prism. Each triangular pyramid, contributing to Vt, is formed by four adjacent vertices, such as A, B, A', and B' (see FIG. 2C). The volume of each triangular pyramid is calculated as:v; = X AB) ■ Eq. 4

[0032] Combining these expressions, the volume of the Kresling tower as a function of height h and external pressure pout is:Vk(h.Pmt). — n n.. Eq. 5 sin - ' - sin —;■n 2 n

[0033] The total volume of gas inside the robot, Vg(h), is:Vfl t — Tfci / l- l Vj "t" -r i p VQ Eq. 6

[0034] where Vb denotes the volumes of the top and bottom cylindrical sections, and Vo is the volume occupied by the actuation system. The volumetric change VPis induced by the differential pressure between the internal and external environments (Ap = pin - pout), causing bulging deformations on each triangular unit. We model this as VP= 2nCfStAP, where Ct is a constant coefficient and St represents the area of each triangular unit (FIG. 2C).PATENT Atty. Dkt. No. 70663-02

[0035] To achieve precise and controlled torsional compression of the Kresling tower, we utilize the principles of tensegrity using a cable-based actuation system (FIG. 2B). The actuation system comprises a geared DC motor with a spur gearbox and a worm gearbox, three pulling cables, and a spinning wheel mounted on the motor’s output shaft, which is centrally aligned with the robot’s central axis. The cables are evenly attached to the vertices of the top cylindrical section and wound around the spinning wheel. As the motor rotates, it winds the cables around the spinning wheel, inducing torsional forces that twist and compress the Kresling tower, thereby altering its effective volume.

[0036] By maintaining an equilibrium between tensile elements (the cables) and compressive components (the Kresling tower, acting like the struts in a tensegrity structure), the robot retains its structural integrity while achieving uniform contraction along the valley creases. The uniform distribution of the cables ensures even load distribution, minimizing the risk of overstressing any single cable.

[0037] The relationship between the actuator angle 0aand the tower height h is:- [h - hs)2- y r'l - rD / r* - 0WQ + (<p - do) Eq. 7

[0038] where LCD is the cable length, rais the radius of the spinning wheel, and rt represents the distance between the cable fixed points in the top section and the central axis. During operation, as the spinning wheel rotates, the cable winds around it with the winding angle 0w = 0wo + 0a - (4> - 4>o), where 0wo and o are the initial values of 0Wand, respectively. And, hsis the distance between the cable fixing points on the spinning wheel and the bottom contours of the cylindrical section.

[0039] FIG. 3 illustrates an exploded view of the top cap 102. The top cap 102 may include a seal board 302, a top board 304, and a top cover 306, and an outer shell 308. FIG. 4 illustrates an exploded view of the top cap 102 without the outer shell 308 for clarity. Reference is made to FIG. 3 and FIG. 4 in the following description the example of the top cap 102.

[0040] The seal board 302 may receive an end of the origami shell 106. By way of example, an end of the origami shell 106 may include a lip. The lip may have a greaterPATENT Atty. Dkt. No. 70663-02 diameter than the Kresling tower structure. The end of the origami shell may pass through the opening of the seal board 302.

[0041] The seal board 302 may have a seat 310 which receives the end of the origami shell 106. For example, the seal board 302 may have a ring shaped outer diameter. An opening 312 centrally defined by seal board 302 may have a hex shape (or other n-gon which matches the structure of the origami shell 106). At least a portion of a span between an edge opening and the outer circumference of the seal board may form the seat. The seal board 302 may secure to the top board 304. The end of the origami shell 106 may be retained between the seal board 302 and the top board 304. A seal 314 (i.e. a ring seal) may be placed between the end of the origami shell 106 and the top plate 304. The seal 314 may help with pressure regulation within the origami shell 106.

[0042] The top plate 304 may include a plurality of standoffs 316 which receive the top cover 306. The top cover 306 may fasten to the standoffs 316. In some examples, the top cap 102 may include an outer shell 308. The outer shell 308 may circumferentially wrap around the seal board 302, top board 304, and any fasteners used to attach the seal board 302 to the top board 304. The outer shell 308 may include a material, such as silicone, or the like. The outer shell 308 may include holes through which the standoffs 316 of the top board 304 extend.

[0043] A compartment 318 may defined between the top cap 306 and the top board 304 or between the top cap 306 and the outer shell 308. The standoffs 316 of the top board 304 may provide the space for the compartment 318. Electronics may be placed within the compartment. For example, a controller may be placed within the compartment. The compartment may be completely sealed from the environment to keep liquid from the electronics. The electronics may include, controller(s), batteries, or any other electronics, sensors, or power sources. Wiring for between sensors, power sources, controllers, and other electronics may be wireless or fed through top cap, bottom cap, and / or origami shell.

[0044] FIG. 5 illustrates an exploded view of the bottom cap 104. The bottom cap 104 may include a seal board 502, a bottom board 504, and a bottom cover 506, and an outer shell 508. FIG. 6 illustrates an exploded view of the bottom cap 104 without the outer shell 508 for clarity. Reference is made to FIG. 5 and FIG. 6 in the following description the example of the bottom cap 104.PATENT Atty. Dkt. No. 70663-02

[0045] The seal board 502 may receive an end of the origami shell 106. (the oragami shell 106 is not shown in FIG. 5 and FIG. 6) By way of example, an end of the origami shell 106 may include a lip. The lip may have a greater diameter than the central origami structure (i.e. the Kresling tower structure). The end of the origami shell may pass through the opening of the seal board 502.

[0046] The seal board 502 may have a seat 510 which receives the end of the origami shell 106. For example, the seal board 502 may have a ring shaped outer diameter. An opening 512 centrally defined by seal board 502 may have a hex shape (or other n-gon which matches the structure of the origami shell 106). At least a portion of a span between an edge opening and the outer circumference of the seal board 502 may form the seat 510. The seal board 502 may secure to the bottom board 504. The end of the origami shell 106 may be retained between the seal board 502 and the bottom board 504. A seal 514 (i.e. a ring seal) may be placed between the end of the origami shell 106 and the bottom plate 504. The seal may be rubber or other elastic and flexible substance to provide a fluid-tight seal. The seal 514 may help with pressure regulation within the origami shell 106.

[0047] The bottom plate 504 may include a plurality of standoffs 516 which receive the bottom cover 506. The bottom cover 506 may fasten to the standoffs 516. In some examples, the bottom cap 102 may include an outer shell 508. The outer shell 508 may circumferentially wrap around the seal board 502, bottom board 504, and any fasteners used to attach the seal board 502 to the bottom board 504. The outer shell 508 may include a material, such as silicone, or the like. The outer shell 508 may include holes through which the standoffs 516 of the bottom board 504 extend.

[0048] A compartment 518 may defined between the bottom cap 506 and the bottom board 504 or between the bottom cap 506 and the outer shell 508. The standoffs 516 of the bottom board 504 may provide the space for the compartment 518. Electronics may be placed within the compartment. For example, a pressure sensor may be placed within the compartment to measure water sensor. Thus, the bottom cap may include a hole which allows fluid into the bottom compartment.

[0049] The top plate may extend around the actuator 108. The actuator may attach to the bottom plate 504.PATENT Atty. Dkt. No. 70663-02

[0050] FIG. 7 illustrates an example of an actuator 108. The actuator 108 may connect to the cap 102 via cables 702. In the example illustrates in FIG 3, the actuator connects to the top plate of the top cap 102.

[0051] The actuator may include a motor 704 and a spinning wheel 706 driven by the motor 704. The cables 702 may circumferential connect to the spinning wheel 706 and extend to the top cap 102. Rotation of the spinning in a first direction may apply torsion to the top cap and cause contraction of the oragami structure (not shown in FIG. 7), thereby causing the top cap and bottom cap of the robotic device to move closer. Rotation of the spinning wheel in a second direction may cause expansion (where the top cap and bottom cap of the robotic device move further apart).

[0052] The actuator 108 may include a chassis 708 that attaches to the bottom cap (bottom cap not shown in FIG 7). The motor 704 and any intervening gears may be affixed to the chassis. It should be appreciated that the bottom cap may include the chassis, or the chassis may affix to the bottom cap.

[0053] FIG. 8 illustrates an example of a worm drive 802 connecting the motor 704 to the spinning wheel 706. The motor 704 may spin a first gear paired with a second gear attached to the worm drive 802. The worm drive 802 may be paired with a toothed portion of the wheel 706. Accordingly, the motor 704 may be oriented along a different axis (i.e. perpendicular) to the spinning wheel 706.

[0054] FIG. 9 illustrates steps for fabrication and assembly according to some embodiments. A technical challenge in the fabrication is to ensure a robust and chemically neutral product that can withstand high-pressure operation. To this end, for structural integrity and to prevent air or water leakage, a customized mold specifically designed for the Kresling tower was developed. In various experimentation, a Kresling tower was created characterized by a radius r of 3.7cm, a short edge a of 4 cm, and a long edge b of 6.3cm, with each dimension maintaining a uniform thickness of 3 mm. The mold comprising an inside mold 902 and an outside mold 904 may be printed using Polylactic acid material (Bambu Lab PLA-CF). A silicone rubber compound (Dragon Skin 30) was selected for the Kresling tower due to its robustness and chemical neutrality. An injection syringe 906 loaded with the compound 908 may be injected in the space 910 between the inner mold and outer mold. After drying, the and mold removal, a Kresling tower shell 912 is formed.PATENT Atty. Dkt. No. 70663-02

[0055] The Kresling tower shell was mounted onto a 3D-printed bottom board 914. A seal board was used in conjunction with the bottom board to securely clamp the lower edge of the Kresling tower, preventing slippage. To prevent air leakage through the printed parts and their connections, we performed a secondary molding process, wrapping the bottom board with a 2-mm-thick outer shell 916 that adhered to the previously molded Kresling tower shell. The cylindrical bottom board was designed with specific mounts for the actuator 918, which includes a 12V DC motor (Pololu 1000:1 Micro Metal Gearmotor), a spur gearbox (1.6:1) for transmission adjustments in tight spaces, and a worm gearbox with a 40:1 reduction ratio to increase output torque. Several electronic components were also mounted on the bottom board, including a motor driver (Adafruit TB6612 1.2A DC / Stepper Motor Driver) and a pressure sensor (SparkFun MS5803-14BA) used to measure the internal air pressure.

[0056] The output shaft of the actuator was connected to a spinning wheel, to which one end of each steel cable was attached. The spinning wheel was designed to allow the cables to wind around it without slipping during operation. The other ends of the cables were secured to the top board. A seal board was used to clamp the upper edge of the Kresling tower to the top board, similar to the bottom assembly. A third molding process was then performed, wrapping the top board with an outer shell 920 to ensure that the internal chamber of the Kresling tower was completely sealed.

[0057] A spherical cap-shaped bottom cover was placed beneath the bottom board, creating a space that allows liquid to enter. For example, a gap may be formed between the bottom cover and the bottom board. This space houses another pressure sensor (SparkFun MS5803-14BA) used to measure the external water pressure and determine the robot’s depth. Similarly, a spherical cap shaped cover was placed above the top board, housing the microcontroller (Arduino Nano ESP32) and a current sensor (Adafruit INA260) used for measuring energy consumption. These were sealed for a leakproof design. The assembled robot has a diameter of 9.6cm, with a height of 12.7cm when fully compressed and 15.5cm when fully extended. In other embodiments the sensor may be placed in the top cap and the microcontroller may be placed in the bottom cap.

[0058] Buoyancy-based Locomotion Through Shape Morphing - The design of robotic device enables precise control over its internal volume through shape morphing, which we harness to implement a buoyancy-based locomotion strategy. ByPATENT Atty. Dkt. No. 70663-02 varying its volume, the robotic device alters the amount of fluid it displaces, thereby changing its net buoyancy force. This allows the robot to float, sink, or achieve neutral buoyancy at specific depths without relying on traditional propulsion mechanisms.

[0059] To analyze the dynamics of the robotic device’s vertical motion, we model the internal gas as an ideal gas undergoing isothermal processes. Applying Boyle’s Law, we relate the internal gas pressure pin and volume Vgto the initial conditions by PinVg= poVg.o, where po is the atmospheric pressure at the surface and Vg,o is the initial internal gas volume.

[0060] The net vertical force acting on the robot is given byTTet = Ft, - Fg- = m,.arEq. 8

[0061] where Fb is the buoyancy force, Fgis the gravitational force, Fd is the drag force, mr is the robot’s mass, and aris its vertical acceleration.

[0062] The buoyancy force is determined by Archimedes’ principle: Fb = pt gVd, where pt is the fluid density, g is the gravitational acceleration, and Vd is the volume of fluid displaced by the robot. The displaced volume Vd depends on whether the robot is fully submerged or partially submerged and is given byVd = + V, - Vf Eq. 9

[0063] where Vs is the volume of the robot’s structure excluding the internal gas, and Vt is the volume above the fluid surface when the robot is partially submerged (with Vt = 0 when fully submerged).

[0064] The gravitational force is simply Fg= mrg. The drag force opposing the robot’s motion is modeled as Fd = (1 / 2)CdPfAvr2, where Cd is the drag coefficient, A is the cross-sectional area perpendicular to the motion, and vris the robot’s vertical velocity. By adjusting its internal volume Vgthrough shape morphing, NAUTILUS can control the net buoyancy force, allowing for precise vertical movement within the fluid.

[0065] Energy Consumption Analysis - The energy consumption during the robot’s sinking, stationary, and ascending phases was analyzed in some embodiments to assess its efficiency. Initially, at the surface, the robot has a total volume Vr = Vgj + Vs. To initiate sinking, the robot reduces its volume to Vr,s such that Vr,s< mr / pf, ensuring that the buoyancy force becomes less than the gravitational force (Fb < Fg). The energy consumed during this volume reduction isPATENT Atty. Dkt. No. 70663-02H'-.i = J pin. Vr);dVr4- TfWa(Vr,s) - 6»0(E,;)] Eq. 10

[0066] where pin(Vr) is the internal pressure as a function of the robot’s volume Vr, 0a(Vr) is the actuation angle corresponding to Vr, and Tt is the friction torque of the actuator.

[0067] As the robot descends to a target depth dt.t, the increased external pressure compresses the robot further, reducing its volume to Vr,s,t= Vr,s- AVP, where AVP= VP(pOut,i) - VP(pOut,t) represents the volumetric change due to pressureinduced deformation, and pout,i and pOut,t are the external pressures at the initial and target depths, respectively.

[0068] To maintain neutral buoyancy at depth, the robot adjusts its volume back to Vr,t= mr / pf by reversing the actuation, consuming additional energy:M’s, 2 = I Pin{^d).d\'r+ Tf[da[\'r,t) ~sEq. 11Jvr s.t

[0069] The total energy consumed during sinking is Ws= Ws,i + Ws, 2.

[0070] While stationary at depth, the robot consumes no energy (Wr= 0) due to the bi-stable nature of the Kresling mechanism, which maintains its configuration without continuous power input.

[0071] To ascend, the robot increases its volume to Vr,t> mr / pt, generating a net upward force. The energy consumed during this volume increase is / ■vt fWf = / Pfo(K), dVr + Tf[dAV.,f) - &a(Vr.t)] Eq. 12

[0072] Therefore, the consumed energy during sinking is:IT = w.! + w,2Eq. 13

[0073] FIG. 10A-B illustrates charts showing energy conception of the robotic device during various states of operation according to some embodiments. FIG. 10A illustrates a chart showing energy consumption during sinking, maintaining depth, and ascending back to the surface according to some embodiments. FIG. 10B illustrates a charge showing total energy consumption at a target depth of 0.40 m (including sinkingPATENT Atty. Dkt. No. 70663-02 and ascending) for different robot diameters according to some embodiments. The marker indicates the current robot diameter.

[0074] Analysis reveals that sinking consumes more energy than ascending due to the additional energy required to adjust the volume at depth (Ws,2), as illustrated in FIG. 10A. Energy consumption increases with target depth, as larger volume changes are needed to counteract the increased external pressure. NAUTILUS consumes approximately 30J to reach a target depth of 0.40m, including both sinking and ascending phases, as depicted in FIG. 10A. With a 12V, WOOmAh battery weighing 100 g, the robot can perform up to 1,440 operation cycles, demonstrating its low energy consumption. The absence of energy consumption while stationary enables long-term monitoring without depleting power reserves.

[0075] Moreover, simulations indicate that reducing the diameter of the robot decreases energy consumption, which is advantageous for practical applications. For example, decreasing the robot’s diameter to 2.0cm can reduce energy consumption to approximately 1% of the current level (FIG. 10B). A robot of this size could be directly ingested by a cow, facilitating easier deployment and access to the rumen for in-vivo applications.

[0076] Controller Design

[0077] To achieve precise depth control, we implement a PID controller that adjusts the actuator angle 0a, thereby varying the internal gas volume Vgand controlling the buoyancy. The control input is the actuator torque Ta, calculated based on the depth error between the desired depth dt.tand the current depth dr: / '* d TO= Kp(df.t-df)+-Ki / (dfyt~df},d:t,+Kd— (df.t-df) Eq. 14

[0078] where KP, Ki, and Kd are the proportional, integral, and derivative gains, respectively.

[0079] The PID controller continuously adjusts Ta to minimize depth error, ensuring that the robot reaches and maintains the target depth without excessive overshoot or oscillations. By experimentally tuning the gains, we achieved a balance between responsiveness and stability. Through this control strategy, the robot demonstrated precise depth regulation with minimal steady-state error, achieving a depth resolutionPATENT Atty. Dkt. No. 70663-02 of 0.5 cm. The system is robust to environmental variations, making it suitable for in-vivo applications where fluid densities and dynamic conditions may change.

[0080] FIG. 11 illustrates a demonstration of deploying the robotic device 100 into a stratified rumen environment 1102. The cow may swallow the robotic device. The stratified rumen environment 502 shown in FIG. 11 is for a cow, but other livestock are possible. The rumen is segmented into three distinct layers from top to bottom: gas, solid hay, and liquid. The overall volume of a cow’s rumen is approximately 184 gallons, with dimensions ranging from 100-125 cm in width and 75-100 cm in height. The robotic device may be deployed into the rumen through a cannula hole.

[0081] To validate the robotic device’s ability to explore the in-vivo rumen environment, outdoor farm experiments were performed on a two-year-old cannulated cow. Before feeding, the robotic device was deployed into the rumen liquid layer through the cannula hole. Utilizing buoyancy-based locomotion, the robotic device moved to various depths within the rumen. To monitor and track the robot’s movement, depth markers were added to the cable connected to the robot.

[0082] A second action may be said to be "in response to" a first action independent of whether the second action results directly or indirectly from the first action. The second action may occur at a substantially later time than the first action and still be in response to the first action. Similarly, the second action may be said to be in response to the first action even if intervening actions take place between the first action and the second action, and even if one or more of the intervening actions directly cause the second action to be performed. For example, a second action may be in response to a first action if the first action sets a flag and a third action later initiates the second action whenever the flag is set.

[0083] To clarify the use of and to hereby provide notice to the public, the phrases "at least one of , ,... and < N>" or "at least one of , ,... < N>, or combinations thereof" or ", ,... and / or < N>" are defined by the Applicant in the broadest sense, superseding any other implied definitions hereinbefore or hereinafter unless expressly asserted by the Applicant to the contrary, to mean one or more elements selected from the group comprising A, B,... and N. In other words, the phrases mean any combination of one or more of the elements A, B,... or N including any one element alone or the one element in combination with one or morePATENT Atty. Dkt. No. 70663-02 of the other elements which may also include, in combination, additional elements not listed.

[0084] While various embodiments have been described, it will be apparent to those of ordinary skill in the art that many more embodiments and implementations are possible. Accordingly, the embodiments described herein are examples, not the only possible embodiments and implementations.

Claims

PATENT Atty. Dkt. No. 70663-02 CLAIMSWhat is claimed is:

1. A robotic device, comprising:A first cap and a second capa origami shell having a Kresling tower structure configured to compress from an expanded state to a contracted state in response to a torsional force, wherein a first end of the origami shell is attached to the first cap and a second end of the origami shell is attached to the second cap; andan actuator disposed within the origami shell, the actuator configured to twist the first cap and cause the first end of the origami shell to collapse toward the second end and thereby vary an effective volume in the origami shell.

2. The robotic device of claim 1, wherein the actuator comprises a plurality of cables and a wheel, the cables attached to the wheel and the first cap, wherein rotating the wheel causes the first cap to rotate.

3. The robotic device of claim 2, wherein the actuator further comprises a worm gearbox and a motor, wherein the motor rotates the worm gearbox, which causes rotation of the wheel along an axis perpendicular to a rotational axis of the motor, the motor.

4. The robotic device of claim 1, wherein the origami shell comprises a plurality of valley folds and mountain folds that extend between the first end and the second end, wherein the valley folds are deformable and the mountain folds are rigid.

5. The robotic device of claim 1, further comprising a pressure sensor.

6. The robotic device of claim 5, wherein the pressure sensor is disposed in the second cap.PATENT Atty. Dkt. No. 70663-02 7. The robotic device of claim 5, further comprising a controller, wherein the controller is configured to adjust the actuator, based on input from the pressure sensor, to vary internal gas volume of the origami shell and thereby control buoyancy of the robotic device.

8. The robotic device of claim 7, wherein the controller is disposed in the first cap and the pressure sensor is disposed in the second cap.

9. The robotic device of claim 1, wherein the second cap has an inlet which allows fluid to enter the origami shell.

10. The robotic device of claim 1, wherein the second cap comprises a bottom board and a cover, wherein an inlet is defined between the bottom board and cover.

11. The robotic device of claim 10, wherein a pressure sensor is disposed in a compartment defined between the bottom board and the cover.

12. The robotic device of claim 1, wherein the first cap comprises a seal board and a top board, wherein the first end of the origami shell has a lip,wherein the origami shell extends through a hole centrally defined by the seal board,wherein the lip is retained between the seal board and top board, wherein cables from the actuator pass through the hole and attach to the top board.

13. The robotic device of claim 12, wherein the first cap further comprises a top cover, and the top board comprises a plurality of standoffs which receive the top cover, wherein the standoffs create a span between the top board and top cover which defines a compartment for housing electronics.

14. The robotic device of claim 1, wherein the second cap comprises a seal board and a bottom board, wherein the second end of the origami shell has a lip,PATENT Atty. Dkt. No. 70663-02 wherein the origami shell extends through a hole centrally defined by the seal board,wherein the lip is retained between the seal board and bottom board, wherein the actuator attaches to the bottom board.

15. The robotic device of claim 14, wherein the second cap further comprises a bottom cover and the bottom board comprises a plurality of standoffs which receive the bottom cover, wherein the standoffs create a span between the bottom board and bottom cover which defines a compartment for housing electronics.

16. the robotic device of claim 1, wherein the dimensions of the robotic device are sized sufficiently small to be swallowable by a cow.

Images