← ArchivePaper2019

SoftCon: Simulation and Control of Soft-Bodied Animals with Biomimetic Actuators

Sehee Min, Jungdam Won, Seunghwan Lee, Jungnam Park, Jehee Lee

SIGGRAPH AsiaAcademic61 citesMuscles Motion Synthesis

Simulates and controls deformable invertebrate characters using muscle-like actuators with contact-rich locomotion.

Open the paper ↗Trace in the lineage graph

Abstract

We present a novel and general framework for the design and control of underwater soft-bodied animals. The whole body of an animal consisting of soft tissues is modeled by tetrahedral and triangular FEM meshes. The contraction of muscles embedded in the soft tissues actuates the body and limbs to move. We present a novel muscle excitation model that mimics the anatomy of muscular hydrostats and their muscle excitation patterns. Our deep reinforcement learning algorithm equipped with the muscle excitation model successfully learned the control policy of soft-bodied animals, which can be physically simulated in real-time, controlled interactively, and resilient to external perturbations. We demonstrate the effectiveness of our approach with various simulated animals including octopuses, lampreys, starfishes, stingrays and cuttlefishes. They learn diverse behaviors such as swimming, grasping, and escaping from a bottle. We also implemented a simple user interface system that allows the user to easily create their creatures.

How to read this

Category: Method: simulation and learned control of soft-bodied characters
Contributions: A general framework modeling whole soft-bodied animals with tetrahedral and triangular FEM meshes actuated by embedded muscles
A muscle excitation model mimicking muscular-hydrostat anatomy and excitation patterns
A deep reinforcement learning controller yielding real-time, interactive, perturbation-resilient behaviors across several invertebrates
Context: Relates to muscle-actuated physical character control (referenced Scalable Muscle-Actuated Human Simulation and Control), extending it from articulated humans to soft underwater invertebrates.Builds on: Scalable Muscle-Actuated Human Simulation and Control
Correctness: Demonstrated on simulated octopuses, lampreys, starfish, stingrays, and cuttlefish learning swimming, grasping, and escape; results are in simulation with a biomimetic actuator abstraction, so fidelity to real animal biomechanics and transfer beyond the shown creatures are not claimed.
Clarity: Readable with strong visual results; a first pass conveys the actuator and RL idea, a second pass is needed for the FEM and excitation-model details.
How to read it: On the first pass focus on the muscle excitation model and how the RL policy is set up; second pass for the FEM actuation coupling and training specifics if reproducing the control results.

Builds on

Scalable Muscle-Actuated Human Simulation and Control 2019

Built upon by

Nothing yet.

Related work

Generative GaitNet 2022 / SIGGRAPH
Computational Bodybuilding: Anatomically-Based Modeling of Human Bodies 2015 / SIGGRAPH
Character Controllers Using Motion VAEs 2020 / SIGGRAPH
DReCon: Data-Driven Responsive Control of Physics-Based Characters 2019 / TOG

Keywords

soft-bodied animals softcon finite element method projective dynamics muscle excitation model ap-mem muscular hydrostats biomimetic actuators deep reinforcement learning proximal policy optimization physics-based control hydrodynamics central pattern generator octopus corotational elasticity volume preservation nerve cord locomotion

This page summarises the entry and links to its original source. The archive never hosts or redistributes the publication itself.Show it in the full archive list →