-
Notifications
You must be signed in to change notification settings - Fork 1
Home
This benchmark is composed of two major parts: the physics engine and the algorithm library. And they are connected by the agent module.
We use a physics engine to simulate the movement and interaction of robots and return measurements of sensors attached to robots. The physics engine simulates in discrete time.
Relating modules: world
, environment
.
The algorithm library contains all components a robot needs to give a control input. The current implementation includes model, estimator, planner, and controller. The algorithm library can be enriched easily by following the interface rules.
Relating modules: model
, estimator
, planner
, controller
.
An agent receives measurements of sensors from the physics engine and gives a control input by a combination of various algorithms. A user can build an agent by specifying which algorithms to use and hyperparameters of algorithms. The algorithms may have dependencies on other algorithms, all the required algorithms should also be used.
Relating modules: agent
.