Myriad: a real-world testbed to bridge trajectory optimization and deep learning

02/22/2022
by   Nikolaus H. R. Howe, et al.
0

We present Myriad, a testbed written in JAX for learning and planning in real-world continuous environments. The primary contributions of Myriad are threefold. First, Myriad provides machine learning practitioners access to trajectory optimization techniques for application within a typical automatic differentiation workflow. Second, Myriad presents many real-world optimal control problems, ranging from biology to medicine to engineering, for use by the machine learning community. Formulated in continuous space and time, these environments retain some of the complexity of real-world systems often abstracted away by standard benchmarks. As such, Myriad strives to serve as a stepping stone towards application of modern machine learning techniques for impactful real-world tasks. Finally, we use the Myriad repository to showcase a novel approach for learning and control tasks. Trained in a fully end-to-end fashion, our model leverages an implicit planning module over neural ordinary differential equations, enabling simultaneous learning and planning with complex environment dynamics.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/17/2018

Optimization Strategies for Real-Time Control of an Autonomous Melting Probe

We present an optimization-based approach for trajectory planning and co...
research
05/04/2023

Real-Time Spatial Trajectory Planning for Urban Environments Using Dynamic Optimization

Planning trajectories for automated vehicles in urban environments requi...
research
07/10/2023

Generalizing Graph ODE for Learning Complex System Dynamics across Environments

Learning multi-agent system dynamics has been extensively studied for va...
research
03/17/2023

Recent Developments in Machine Learning Methods for Stochastic Control and Games

Stochastic optimal control and games have found a wide range of applicat...
research
03/02/2023

Continuous Implicit SDF Based Any-shape Robot Trajectory Optimization

Optimization-based trajectory generation methods are widely used in whol...
research
02/15/2020

Jelly Bean World: A Testbed for Never-Ending Learning

Machine learning has shown growing success in recent years. However, cur...
research
06/08/2023

AI Enhanced Control Engineering Methods

AI and machine learning based approaches are becoming ubiquitous in almo...

Please sign up or login with your details

Forgot password? Click here to reset