Parareal with a learned coarse model for robotic manipulation
- PDF / 1,519,708 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 46 Downloads / 226 Views
PINT 2019
Parareal with a learned coarse model for robotic manipulation Wisdom Agboh1
· Oliver Grainger2 · Daniel Ruprecht3
· Mehmet Dogar1
Received: 6 December 2019 / Accepted: 27 August 2020 © The Author(s) 2020
Abstract A key component of many robotics model-based planning and control algorithms is physics predictions, that is, forecasting a sequence of states given an initial state and a sequence of controls. This process is slow and a major computational bottleneck for robotics planning algorithms. Parallel-in-time integration methods can help to leverage parallel computing to accelerate physics predictions and thus planning. The Parareal algorithm iterates between a coarse serial integrator and a fine parallel integrator. A key challenge is to devise a coarse model that is computationally cheap but accurate enough for Parareal to converge quickly. Here, we investigate the use of a deep neural network physics model as a coarse model for Parareal in the context of robotic manipulation. In simulated experiments using the physics engine Mujoco as fine propagator we show that the learned coarse model leads to faster Parareal convergence than a coarse physics-based model. We further show that the learned coarse model allows to apply Parareal to scenarios with multiple objects, where the physics-based coarse model is not applicable. Finally, we conduct experiments on a real robot and show that Parareal predictions are close to real-world physics predictions for robotic pushing of multiple objects. Code (https://doi.org/10.5281/zenodo.3779085) and videos (https://youtu. be/wCh2o1rf-gA) are publicly available. Keywords Parallel-in-time · Parareal · Manipulation · Robotics · Planning · Neural network · Model-predictive control · Learning
1 Introduction We present a method for fast and accurate physics predictions during non-prehensile manipulation planning and control. Communicated by Robert Speck. This project was supported by an EPSRC studentship (1879668) and EPSRC grants EP/R031193/1 and EP/P019560/1.
B
Wisdom Agboh [email protected] Oliver Grainger [email protected] Daniel Ruprecht [email protected] Mehmet Dogar [email protected]
1
School of Computing, University of Leeds, Leeds, UK
2
School of Mechanical Engineering, University of Leeds, Leeds, UK
3
Lehrstuhl Computational Mathematics, Institut für Mathematik, Technische Universität Hamburg, Hamburg, Germany
An example scenario is shown in Fig. 1, where a robot arm pushes the marked cylindrical object into a target zone without pushing the other three objects off the table. We are interested in predicting the motion of the objects in a fast and accurate way. Physics engines like Mujoco [37] and Drake [36] solve Newton’s equation to predict motion. They are accurate but slow. Coarse models can be built by introducing simplifying assumptions, trading accuracy for solution speed but their lack of precision will eventually compromise the robot’s chance of completing a given task successfully. Given an initial state and a sequence of controls,
Data Loading...
