Skip to content

Tool for managing clusters of receding horizon controllers on CPU. Aimed at facilitating the implementation of the training phase of Reinforcement Learning-based Model Predictive Control policies.

License

Notifications You must be signed in to change notification settings

AndrePatri/CoClusterBridge

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

License

icon.svg

If you have a beautiful CPU-based controller developed, for example, with CPU-based Trajectory Optimization tools and want to train and deploy a *Reinforcement Learning-based Model Predictive Control* policy without having to rewrite your whole controller on GPU, then this might be the right tool for you.

CoClusterBridge is a tool for bridging parallel simulations (typically GPU-based), with a cluster of CPU-based (receding-horizon) controllers while properly handling controllers synchronization and triggering. It also supports full CPU operation. For an example application, please have a look at IBRIDO.

About

Tool for managing clusters of receding horizon controllers on CPU. Aimed at facilitating the implementation of the training phase of Reinforcement Learning-based Model Predictive Control policies.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published