Tagged a github release of Bullet Physics and PyBullet, both version 3.05. The release was used for our motion imitation research, and also includes various improvements for the finite-element-method (FEM) deformable simulation, by Xuchen Han and Chuyuan Fu .
Assistive Gym leverages PyBullet for physical human-robot interaction and assistive robotics. Assistive Gym currently supports four collaborative robots and six physically assistive tasks. It also supports learning-based control algorithms, and includes models of human motion, human preferences, robot base pose optimization, and realistic pose-dependent human joint limits
Paper Link: https://ras.papercept.net/proceedings/ICRA20/1572.pdf Zackory Erickson, Vamsee Gangaram, Ariel Kapusta, C. Karen Liu, and Charles C. Kemp, “Assistive Gym: A Physics Simulation Framework for Assistive Robotics”, IEEE International Conference on Robotics and Automation (ICRA), 2020.
In “Learning Agile Robotic Locomotion Skills by Imitating Animals”, we present a framework that takes a reference motion clip recorded from an animal (a dog, in this case) and uses RL to train a control policy that enables a robot to imitate the motion in the real world. By providing the system with different reference motions, we are able to train a quadruped robot to perform a diverse set of agile behaviors, ranging from fast walking gaits to dynamic hops and turns. The policies are trained primarily in simulation, and then transferred to the real world using a latent space adaptation technique that can efficiently adapt a policy using only a few minutes of data from the real robot. All simulations are performed using PyBullet.
Facebook AI Habitat is a new open source simulation platform created by Facebook AI that’s designed to train embodied agents (such as virtual robots) in photo-realistic 3D environments. The latest version adds Bullet Physics.
TossingBot, a new paper by Google Robotics (Andy Zeng, Shuran Song, Johnny Lee, Alberto Rodriguez, Thomas Funkhouser
) using PyBullet.
We investigate whether a robot arm can learn to pick and throw arbitrary objects into selected boxes quickly and accurately. Throwing has the potential to increase the physical reachability and picking speed of a robot arm. However, precisely throwing arbitrary objects in unstructured settings presents many challenges: from acquiring reliable pre-throw conditions (e.g. initial pose of object in manipulator) to handling varying object-centric properties (e.g. mass distribution, friction, shape) and dynamics (e.g. aerodynamics). In this work, we propose an end-to-end formulation that jointly learns to infer control parameters for grasping and throwing motion primitives from visual observations (images of arbitrary objects in a bin) through trial and error. Within this formulation, we investigate the synergies between grasping and throwing (i.e., learning grasps that enable more accurate throws) and between simulation and deep learning (i.e., using deep networks to predict residuals on top of control parameters predicted by a physics simulator). The resulting system, TossingBot, is able to grasp and throw arbitrary objects into boxes located outside its maximum reach range at 500+ mean picks per hour (600+ grasps per hour with 85% throwing accuracy); and generalizes to new objects and target locations.