Morphological computation and decentralized learning in a swarm of sterically interacting robots

Abstract

Whereas naturally occurring swarms thrive when crowded, physical interactions in robotic swarms are either avoided or carefully controlled, thus limiting their operational density. Here, we present a mechanical design rule that allows robots to act in a collision-dominated environment. We introduce Morphobots, a robotic swarm platform developed to implement embodied computation through a morpho-functional design. By engineering a three-dimensional printed exoskeleton, we encode a reorientation response to an external body force (such as gravity) or a surface force (such as a collision). We show that the force orientation response is generic and can augment existing swarm robotic platforms (e.g., Kilobots) as well as custom robots even 10 times larger. At the individual level, the exoskeleton improves motility and stability and also allows encoding of two contrasting dynamical behaviors in response to an external force or a collision (including collision with a wall or a movable obstacle and on a dynamically tilting plane). This force orientation response adds a mechanical layer to the robot’s sense-act cycle at the swarm level, leveraging steric interactions for collective phototaxis when crowded. Enabling collisions also promotes information flow, facilitating online distributed learning. Each robot runs an embedded algorithm that ultimately optimizes collective performance. We identify an effective parameter that controls the force orientation response and explore its implications in swarms that transition from dilute to crowded. Experimenting with physical swarms (of up to 64 robots) and simulated swarms (of up to 8192 agents) shows that the effect of morphological computation increases with growing swarm size.

A Mechanical Origin of Cooperative Transport

“…Here, with each collision, robots tend to turn into the payload and progressively push it until reaching the perimeter of the arena (150 cm diameter). We further find the effect to increase with payload diameter (2a = 7 − 32 cm), swarm size (N = 1 − 53 robots) in both a custom-made and a modified commercial multi-robot platform 41,47 (see Fig. 4 and Supporting Information Section I B 4).…”

“…An inspection of the contact dynamics of an active particle with the ground (Supporting Videos 3 and 4) allows us to derive from first principles the pivotal contribution of mechanics to force-alignment (Fig. 2), thereby extending, and generalizing previous phenomenological descriptions for the equations of motion of selfpropelled particles 22,28,[35][36][37][43][44][45][46][47] . We find that the force-alignment is intrinsic to active particles, and can be described by a signed, charge-like parameter with units of curvature, which we term "curvity".…”

A self-organizing robotic aggregate using solid and liquid-like collective states

“…Still, although robotic systems with a single soft body are able to achieve highly variable and adaptive behaviors (16), the inherent need to model materials in the large deformation limit makes the design difficult. Furthermore, inspired by flocking observed in nature (17,18), multi-agent swarm approaches have been successful in generating autonomous, as well as resilient, robotic systems by using a large number of individual robotic units where large-scale collective behaviors emerge through local rules between neighboring units (19)(20)(21)(22)(23). Such noncentralized control is able to absorb local perturbations in the inherent noise of the collective organization, and the failure of a few in a swarm of similar units does not necessarily affect the capabilities at the collective scale (23,24).…”

Micro/Nanorobotic Swarms: From Fundamentals to Functionalities

“…10,11 In the early 2000s, various types of macroscale (>10 −1 m) modular robotic systems and robotic swarms consisting of a few robots (e.g., Kheperas, 12 sbots, 13−15 and Jasmines 16 ) began to appear, and the macroscale robotic swarm was systematically reviewed. 17 Since then, macroscale and mesoscale (10 −3 m to 10 −1 m) robotic swarms with homogeneous agents, such as Slimebots, 18 Alices, 19 epucks, 20 Kilobots, 21 Bubblebots, 22 Bristle-bots, 23,24 Particlebots, 25 mindless rodlike robots, 26 Rainbow-bots, 27 and Morphobots, 28 and heterogeneous robotic swarms, such as Swarmanoid, 29 have been developed, inspired by natural swarm behaviors like insect pattern formations and phototaxis. To date, for large-scale robots, swarm coordination has been designed for cooperative tasks, such as directional motion, 21,25,30 obstacle traversal, 25,31,32 cargo allocation, 15,25,31,33 environment exploration, 34−36 and viral testing.…”

“…Since the 1990s, researchers have been leveraging natural swarm behaviors and intelligence for developing swarms in robotics, where groups of robots coordinate and cooperatively perform tasks. , In the 1990s, the development of algorithms had already begun for robotic swarms, including pattern generation, navigation, , and materials design. , In the early 2000s, various types of macroscale (>10 –1 m) modular robotic systems and robotic swarms consisting of a few robots (e.g., Kheperas, s-bots, − and Jasmines) began to appear, and the macroscale robotic swarm was systematically reviewed . Since then, macroscale and mesoscale (10 –3 m to 10 –1 m) robotic swarms with homogeneous agents, such as Slimebots, Alices, e-pucks, Kilobots, Bubblebots, Bristle-bots, , Particle-bots, mindless rodlike robots, Rainbow-bots, and Morphobots, and heterogeneous robotic swarms, such as Swarmanoid, have been developed, inspired by natural swarm behaviors like insect pattern formations and phototaxis. To date, for large-scale robots, swarm coordination has been designed for cooperative tasks, such as directional motion, ,, obstacle traversal, ,, cargo allocation, ,,, environment exploration, − and viral testing …”