Biological Robots: Perspectives on an Emerging Interdisciplinary Field

Abstract

Advances in science and engineering often reveal the limitations of classical approaches initially used to understand, predict, and control phenomena. With progress, conceptual categories must often be re-evaluated to better track recently discovered invariants across disciplines. It is essential to refine frameworks and resolve conflicting boundaries between disciplines such that they better facilitate, not restrict, experimental approaches and capabilities. In this essay, we address specific questions and critiques which have arisen in response to our research program, which lies at the intersection of developmental biology, computer science, and robotics. In the context of biological machines and robots, we explore changes across concepts and previously distinct fields that are driven by recent advances in materials, information, and life sciences. Herein, each author provides their own perspective on the subject, framed by their own disciplinary training. We argue that as with computation, certain aspects of developmental biology and robotics are not tied to specific materials; rather, the consilience of these fields can help to shed light on issues of multiscale control, self-assembly, and relationships between form and function. We hope new fields can emerge as boundaries arising from technological limitations are overcome, furthering practical applications from regenerative medicine to useful synthetic living machines.

Keywords: animal cap; biorobot; computer science; developmental biology; embryo; robot; synthetic bioengineering; xenobot.

FIG. 1.

Sample of designs produced in the research program. Actuation can be achieved through motile cilia generated flow, contractile muscle tissue, or a combination of the two. Morphology can be generated via compression, sculpting, or by layering specific tissue types during the construction process (far right panel, red indicates muscle, green indicates epidermis). 5th panel modified and reprinted with permission from ref. [8], National Academy of Sciences of the United States of America.

FIG. 2.

Sample xenobot designs optimized by machine learning methods. Left: xenobots designed for locomotion using myocardial tissue (red). With permission from ref. [8]. Right: xenobots designed for kinematic self-replication. With permission from ref. [9].

FIG. 3.

Engineering with agential materials. Engineering has traditionally been carried out with passive materials (A), which can only be dependent on to keep their shape and other physical properties. These must be carefully managed for each desired functionality, giving rise to a perception of robotics as the manual arrangement of parts toward each goal. However, increasingly, engineering has moved toward active matter (B) and computational media (C) as well recognized in soft robotics; now, biorobotics enters a new phase transition where the material is “agential”—it is composed of subunits (D, living cells) which themselves were whole organisms once and thus have many built-in competencies and agendas, including problem-solving in physiological, metabolic, and morphological problem spaces.^,,, This means that robots are now not only constructed by physical (or even genetic) rewiring (E) but more akin to behavior-shaping (F), using signals and environments to achieve desired system-level behavior. In contrast and complement to 3D-printing and similar approaches designed for building with passive matter (G), which also works with cells (H), collective intelligence of living systems at all scales (such as that of an ant swarm, I) can be used to manipulate the collective behavior of cells (J) in anatomical morphospace, by signaling that alters the collective's navigation policy of that space: just as instructive signals from other cells cause frog ectodermal cells to be a two-dimensional barrier in standard embryos (K), techniques such as subtraction (of other cells and their signals) and stimuli can achieve guided self-assembly toward novel form and function (L). 3D, three-dimensional. All images by Jeremy Guay of Peregrine Creative, used with permission.