Evolving soft robotic locomotion in PhysX

Abstract

Given the complexity of the problem, genetic algorithms are one of the more promising methods of discovering control schemes for soft robotics. Since physically embodied evolution is time consuming and expensive, an outstanding challenge lies in developing fast and suitably realistic simulations in which to evolve soft robot gaits. We describe two parallel methods of using NVidia's PhysX, a hardware-accelerated (GPGPU) physics engine, in order to evolve and optimize soft bodied gaits. The first method involves the evolution of open-loop gaits using a reduced-order lumped parameter model. The second method involves harnessing PhysX's soft-bodied material simulation capabilites. In each case we discuss the the challenges and possibilities involved in using the PhysX for evolutionary soft robotics.

Software toolkit for modeling, simulation, and control of soft robots

“…These simulation kernels come from the video-game industry and are often focused on articulated-rigid bodies. They have been successfully used to model and simulate soft robots as in the NASA Tensegrity Robotics Toolkit [19] or in [49] with the use of PhysX to evaluate the candidate solutions of genetic algorithms. The video-game based simulation frameworks are fast and efficient to compute rigidbody simulations as well as some kind of soft bodies.…”

Creating and Debugging Performance CUDA C

“…Rieffel et al [31] showed an nVidia 9800GT could be used to evolve movement in a soft robot. The target pneumatic robot was simulated using PhysX.…”

Unshackling evolution

“…However, few attempts have been made to evolve robots composed of soft materials [27], and most of those attempts are limited to only a few components. This paucity is due largely to the computational costs of simulating flexible materials and because many genetic encodings do not scale to large parameter spaces [5,18].…”