Bulletin of the American Physical Society
APS March Meeting 2023
Volume 68, Number 3
Las Vegas, Nevada (March 5-10)
Virtual (March 20-22); Time Zone: Pacific Time
Session Y10: Robophysics IVFocus
|
Hide Abstracts |
Sponsoring Units: DBIO Chair: Chen Li, Johns Hopkins University Room: Room 202 |
Friday, March 10, 2023 8:00AM - 8:36AM |
Y10.00001: Limit cycles turn odd active matter into robots Invited Speaker: Corentin Coulais Controlling how waves propagate, attenuate and amplify in simple, scalable geometric structures is a daunting challenge for science and technology. In this talk, I will discuss how odd active matter---media in which energy conservation and chiral symmetries are simultaneously broken---can be used to steer mechanical waves in unprecedented ways. Combining experiments on mechanical lattices of distributed robots with wave physics and continuum mechanics, I will discuss the emergence of unidirectionally amplified waves, of topological waves and of one-way solitons in odd active matter. I will further show how these odd waves can be used to induce locomotion and unusual responses to impacts and hence pave the way towards a novel generation of distributed robots with autonomous and adaptable locomotion. |
Friday, March 10, 2023 8:36AM - 8:48AM |
Y10.00002: Coordinated collective behavior through contact interactions and adaptive oscillators Nick G Gravish The collective behavior of animals and robots, such as flocking and schooling, often involves long-range sensing within low-density groups. However, when collectives are in high density configurations their vision and communication capabilities may be compromised. Instead of vision and communication, individuals in high density collectives may leverage physical interactions to coordinate their movements. In this talk we present design principles for high density collectives to achieve phase synchronized motion through contact interactions. This approach relies on appropriately chosen adaptive oscillators that govern each individual's movement, and that can synchronize through contact interactions between robots. We demonstrate these capabilities through two sets of experiments: 1) worm-inspired collective undulatory movement in which undulatory gaits synchronize through contact, and 2) ant-inspired collective manipulation in which pushing robots synchronize their collective pushing phases. To study the stability of these collective behaviors we introduce the contact return-map, which examines the evolution of phases through contact to next contact. The contact return-map provides the ability to design oscillator feedback rules that enable optimal synchronization and collective movement. We lastly discuss how these principles may apply to biological collectives that operate in close-proximity. |
Friday, March 10, 2023 8:48AM - 9:00AM |
Y10.00003: Stigmergic formation of soft matter structures in a robophysical collective Daniel Daniel Soto, Joonha Hwang, Michael D Goodisman, Daniel I Goldman Biological collectives like ants and termites manipulate environmental soft materials (soil, twigs, leaf litter) to create 3D structures that can be redesigned to suit the swarm's needs, despite no individual having knowledge of the global design. To discover principles by which groups perform such tasks, we develop a robophysical system where robotic agents manipulate soft material to create simple structures in response to environmental cues. We use a geometrically cohesive material composed of U-shaped particles which can form 3D structures such as walls, slopes, and arches. Further, we develop a scalable mudskipper-inspired robophysical model (length = 24 cm, width and height = 16 cm) to traverse the complex terrain formed during material manipulation. The robots move at 12 cm/s and negotiate 5 cm tall obstacles via a combination of limb, wheel, and tail actuation. These agents operate within a dark arena (length = 1.8 m, width = 1.2 m) in which light cues dictate different behaviors based on color intensity. A single agent transports ~10% of the initial staple mound over 4 hours. We aim to assign these robots with low level tasks such as material transport coordinated by the environment (i.e., "stigmergy") and posit that the resulting structures will facilitate higher level tasks. Behaviors discovered during structure formations will provide insight on the methods of collective construction in soft matter environments and develop hypotheses for how complex habitats emerge in biological collectives. |
Friday, March 10, 2023 9:00AM - 9:12AM |
Y10.00004: Towards a Colloidal Crystal Robot Gripper Corwin B Kerr, Nipuli K Gunaratne, Philipp Schönhöfer, Sharon C Glotzer Colloidal particles can be manufactured with varying shape, activity, or interaction patchiness to assemble into a wide variety of crystals. Converting a passive colloidal crystal into a crystalline colloidal robot that changes morphology and performs robotic tasks requires a blueprint to coordinate particle motion within a crystal. Such a machine could unlock new possibilities for manipulating microscale objects but requires a way to move matter in dense systems. Building on the work of VanSaders et. al., we study a strategy to control the formation and migration of dislocations, which carry plastic deformation in crystals, by implanting a cluster of cyclically swelling and deswelling colloids into a finite 2D colloidal crystallite. Across dozens of swelling cycles, the sequential motion of dislocations gradually changes the crystallite’s shape. In this computational project, we decompose the geometry of the cluster of swelling particles into basic elements such as angle, edge length and cluster asymmetry. We show how these geometric elements control where dislocations form and how they move in the 2D crystallite. As one example, we show how these design rules can be combined to create a colloidal robotic gripper to trap an object. |
Friday, March 10, 2023 9:12AM - 9:24AM |
Y10.00005: Swarms of Optically Controllable Electrokinetic Microrobots William H Reinhardt, Lucas C Hanson, Scott Shrager, Tarunyaa Sivakumar, Maya Lassiter, Marc Z Miskin Electrochemical propulsion is a high speed, low energy method of locomotion for microscopic robots. While current devices that operate using this mechanism boast high speed and chemical control, they lack compatibility with standard silicon microelectronics. This limitation makes it difficult to perform tasks like sensing and swarming as devices can’t easily collect and store information or be reprogrammed to change behaviors. Here we show a new class of photovoltaic-powered, electrochemically propelled microrobots that are compatible with CMOS electronics and are fabricated in parallel using standard lithography techniques. The microrobots receive and process basic commands, can be selectively controlled, and locomote at speeds over 300 microns/s when illuminated by light. These results represent an alternative, low power locomotion mechanism for microscopic robots and help pave the way for future studies towards emergent behaviors in swarms of microrobots. |
Friday, March 10, 2023 9:24AM - 9:36AM |
Y10.00006: Granulobots: Leveraging mechanical properties of a decentralized, multi-unit, dense robotic aggregate for sensorless tasks. Baudouin Saintyves, Matthew Spenko, Heinrich M Jaeger Designing robotic systems that can autonomously interact with their environment to solve tasks remains a major challenge. Conventional approaches use multi-sensor feedback loops to control displacements. This is often coupled with algorithmic and hardware complexity, which tend to be detrimental to energy efficiency, reliability, and form factor , all key aspects in autonomous systems. Here we use a new decentralized and modular robotic platform that we recently developed, Granulobot, to demonstrate tasks based on aggregate mechanical properties. Granulobots consist of active, gear-like particles that magnetically interact with each other and produce torques. They can self-assemble into aggregates that can reconfigure in real-time. The apparent complexity of a system with many degrees of freedom is made an advantage by leveraging the material-like properties of aggregates. In particular, aggregates can transition between rigid and liquid-like states with a wide range of effective viscosities. This enables the robot to grab objects as well as move in complex environments through holes and over obstacles by setting a single ”material” parameter, and without centralized control or real-time sensor-based feedback. Such minimal control, enabled by the modular design of Granulobots, advances robotic autonomy by exploring a morphological form of computation and feedback that greatly reduces the number of control parameters that must be taken care of by the embedded systems or operators. |
Friday, March 10, 2023 9:36AM - 9:48AM |
Y10.00007: Fabricating Reprogrammable Solar Powered Microscopic Robots Maya Lassiter We show progress towards programmable, autonomous microscopic robots. Each robot has sensors, memory, solar cells as power supplies and a microcontroller integrated together, all fabricated in low-power silicon electronics. Electrochemical actuators are fabricated on the electrical systems enabling locomotion and sensor readout. The custom microcontroller on each robot can be programmed by sending 11-bit instructions via optical signals to a receiver, allowing a user to define the robot's behavior. All the steps used to build these machines are carried out massively in parallel, allowing for swarms of tiny, programmable robots. Long term, we hope that programmability will enable a wide range of complex, user-customizable behaviors and new studies of locomotion or swarming at the microscale. |
Friday, March 10, 2023 9:48AM - 10:00AM |
Y10.00008: Collective chase: emergent dynamics of Braitenberg's vehicle 2b (aggression) Rundong Yang, Wei Zhou, Nick G Gravish The swarming, flocking, and schooling behaviors of animals, in which agents tend to align orientations, have provided myriad insights into the physics of collective systems. In this talk, we study the dynamics of a different collective system, inspired by Braitenberg's "aggression" vehicles. Individuals move with constant velocity and align their orientation towards the position of their neighbors. This simple rule of motion leads to collective chase dynamics that are markedly different from flocking. We first analyze chaser pairs and demonstrate several emergent states including straight-line and circular chasing. Using two mobile robots with light-dependent feedback control we validated these observations in experiment. In simulations with larger groups, we study the role of sensory input by restricting the visual range of the individuals (vision-cone angle) and observe vision-dependent emergent behaviors such as the number of chasing clusters and the average agent-agent distance in each cluster. Lastly, we demonstrate how chasers can converge to close proximity clusters through imposed oscillatory motion. This method shows a concise way for compact swarm aggregation without complex communication and precise distance measurement requirements which are commonly used in previous research. |
Friday, March 10, 2023 10:00AM - 10:12AM |
Y10.00009: Adaptable force chains in granular assemblies using variable stiffness particles Sven Witthaus, Atoosa Parsa, Nidhi Pashine, Jerry Zhang, Corey S O'Hern, Rebecca Kramer-Bottiglio Under an externally applied load, force chain networks form in granular assemblies that depend on, among other things, the contact network and stiffness of the grains. In this work, we fabricate variable stiffness particles, whose stiffness can be reversibly changed on demand to tune the force network in a packing. Each variable stiffness particle is made of a silicone shell that encapsulates a Field’s metal core. This eutectic alloy of bismuth, indium, and tin with a low melting point, exhibits a large drop in its elastic moduli after changing from solid to liquid. By sending electric current through co-located copper heaters, the Field’s metal can melt via Joule heating, which softens the particle. In the cool state, the Field’s metal particle modulus is 4 MPa but reduces to 1 MPa in the soft, heated state. To optimize the mechanical response of granular packings containing mixtures of soft and stiff particles, we employ evolutionary algorithms coupled with discrete element method simulations to dictate the patterning of grains that will yield a particular force output on the assembly boundary. The predicted designs were replicated in experiments using variable stiffness particles in a photoelastic container to measure the output forces between the particles and the boundary, with good matching between simulation and reality. We view this result as a first step toward making granular metamaterials made of robotic grains that can dynamically adapt their force chains, bulk moduli, and frequency response on demand. |
Friday, March 10, 2023 10:12AM - 10:24AM |
Y10.00010: Guided motion of kinked vesicles by active particles Sophie Y Lee, Philipp Schönhöfer, Sharon C Glotzer Collective motion of active particles has been studied extensively, yet effective strategies to navigate active particle swarms are still rare and rely on external guidance. We introduce a method to control the trajectories of swarms of active rod-like particles by confining them within a rigid bounding membrane (vesicle) of non-uniform shape. We show that the propelling agents spontaneously form clusters at the membrane wall and collectively propel the vesicle, turning it into an active superstructure. To steer the motion of the superstructure, we add a kink to the bounding membrane that directs the motion of the vesicle. We show how the system's geometrical and material properties, such as the aspect ratio and Péclet number of the active rods, along with the kink angle and flexibility of the membrane, determine the stacking of active particles at the kink. The stacking induces different types of vesicle motion, including kink-forward linear motion, kink-backward linear motion, and circular motion. Based on our findings, we designed vesicles with switchable and reversible motions triggered by external forces that change specific properties of the system. The observed phenomena suggest a promising strategy for steering vesicles and other confining soft matter systems and, conversely, a promising approach to particle transport and swarm control. Finally, we show how kinked vesicles can also serve to navigate active matter through complex and tortuous environments. |
Friday, March 10, 2023 10:24AM - 10:36AM |
Y10.00011: Emergent swarm robot behaviors induced by confinement and contact interactions Zhexin Shen, Ram Avinery, Hosain Bagheri, Daniel Soto, Daniel I Goldman Building on our study of phase changes in contact-mediated robot collectives [Li et al, Science Advances 2021], we study the dynamics and task capabilities of a confined programmable swarm of twelve disk-shaped (8 cm diameter) wheel-driven robots that make decisions based on local collisions. Each robot differentiates contact between other agents and environmental features (e.g., rigid walls, obstacles) via a force-sensitive resistor pad on its surface. The robots’ light sensors enable individual phototaxis via run-and-tumble dynamics. To enhance robot interactions and generate swarms of different collective mechanical properties, we place the swarm within a movable, flexible membrane which can actively change perimeter length. When the perimeter is small, the robots cluster and form a solid-like state; when expanded, the swarm functions like an active gas. The robots can control the perimeter and shape of the membrane via collisions and contact; the membrane can actively respond to the collisions to dynamically induce phase changes in the swarm. We posit that we can utilize the multiphase swarm dynamics in conjunction with membrane induced rigidity to perform tasks like object engulfment and transport. |
Friday, March 10, 2023 10:36AM - 10:48AM |
Y10.00012: Recipes for Emulating Nonlinear Active Materials with Robotics Kai Qian, Phoenix Stout, Gustav Blankenberg, Wei Zhou, Jihad E Alqasimi, Nicholas Gravish, Nicholas Boechler Within the area of wave propagation through nonlinear phononic media, there are still many practical challenges that remained to be solved, such as the physical realization of specific nonlinear and non-conservative multibody interactions. In this work, we demonstrate that programmable interactions between undulatory robots can be a versatile tool for studying nonlinear wave propagation that relaxes the limits necessitated by constituent materials and geometry. We perform simulations and experiments of simple virtually-coupled undulatory robots with one rotational degree of freedom that actively interact via torque control based on instantaneous position measurement. There is no need to integrate feedback signals and no central control of the system, except for local interactions. The theory and experiments in this study show how programmed robotic interactions can not only emulate the behaviors of nonlinear active materials with potential applications across multiple disciplines of physics and engineering, but also serve as a foundamental step in design and development of new robotic and mechanical metamaterials. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700