Abstract
In this review we argue for the creation of a physics of moving systems -- a locomotion "robophysics" -- which we define as the pursuit of the discovery of principles of self generated motion. Robophysics can provide an important intellectual complement to the discipline of robotics, largely the domain of researchers from engineering and computer science. The essential idea is that we must complement study of complex robots in complex situations with systematic study of simplified robophysical devices in controlled laboratory settings and simplified theoretical models. We must thus use the methods of physics to examine successful and failed locomotion in simplified (abstracted) devices using parameter space exploration, systematic control, and techniques from dynamical systems. Using examples from our and other's research, we will discuss how such robophysical studies have begun to aid engineers in the creation of devices that begin to achieve life-like locomotor abilities on and within complex environments, have inspired interesting physics questions in low dimensional dynamical systems, geometric mechanics and soft matter physics, and have been useful to develop models for biological locomotion in complex terrain. The rapidly decreasing cost of constructing sophisticated robot models with easy access to significant computational power bodes well for scientists and engineers to engage in a discipline which can readily integrate experiment, theory and computation.
Introduction: The Manifesto for Robophysics
This review paper serves as a manifesto for Robophysics—the systematic, physics-oriented study of how self-propelled systems move through complex environments. By treating robots as "physical models" rather than just engineering tools, the authors argue we can discover the fundamental laws of locomotion that have remained elusive in traditional robotics and biology. Robophysics is an emerging discipline that uses simplified robotic models and the tools of soft matter physics to discover the fundamental principles of movement. By bridging the gap between engineering success and physical understanding, it provides a framework for building robots that can navigate "flowable" terrains like sand, mud, and rubble as effectively as animals.
The Challenge of Unstructured Environments
While we are excellent at building robots for paved roads or open water, our machines struggle immensely in natural, "unstructured" environments. Animals navigate trees, desert dunes, and mudflats with ease, but robots often get stuck or fail unpredictably in these settings. The robotics community has historically focused on optimization (getting a specific task done) rather than principles (understanding why a movement works). We lack a "minimal feature set" for robust locomotion—the equivalent of what thermodynamics provided for the development of the steam engine. Robophysics seeks to fill this gap by studying how simplified devices interact with complex, deformable substrates.
Methodology: Simplification and Abstraction
The paper outlines a multi-pronged approach to studying movement:
- Simplification & Abstraction: Instead of modeling every joint of a complex animal, researchers use "robophysical models"—simplified robots (like a 1-DOF jumper or a 7-link snake) to isolate specific physical variables.
- Templates and Anchors: They use "templates" (reduced-order models like the Spring Loaded Inverted Pendulum) to describe the target behavior and "anchors" (the actual hardware) to implement it.
- Terradynamics (RFT): The authors developed Resistive Force Theory (RFT) for granular media. Similar to how we use fluid dynamics for airplanes, RFT allows researchers to calculate the forces on a robot's limb as it moves through sand without needing to simulate every individual grain.
- Geometric Mechanics: They apply "gauge theory" frameworks to locomotion. By using Constraint Curvature Functions (CCFs), they can map a robot's internal shape changes to its net displacement, treating movement as a geometric problem.
- Automated Experimentation: Using systems like SCATTER, researchers can run thousands of trials autonomously, systematically varying parameters like ground compaction to find "bifurcations" where locomotion suddenly fails.
Key Insights and Results
- Failure as a Feature: In robophysics, studying why a robot fails (e.g., when sand "yields") is the key to discovering the physical limits of a gait.
- The "Stutter Jump": Automation revealed a "stutter jump" strategy—a small preliminary hop followed by a big push—that exploits a substrate's timing-dependent response better than a single big jump.
- Granular Added Mass: Jumping on sand creates a "cone" of jammed grains that moves with the foot, effectively increasing the robot's mass.
- SandBot and Sidewinding: By understanding "rotary walking" and contact length modulation, researchers enabled robots to move on loose sand at unprecedented speeds and navigate steep dunes without slipping.
Limitations and Future Directions
Current robophysical models focus primarily on steady-state, rhythmic gaits. Future work must address transient, impulsive movements and more complex environments like wet mud or snow. Additionally, the authors suggest "collective robophysics"—applying these principles to swarms of robots—as the next major frontier.
Reference: Aguilar, J., et al. (2016). A review on locomotion robophysics: the study of movement at the intersection of robotics, soft matter and dynamical systems. arXiv:1602.04712v1 [cs.RO].
Summary updated