For RL task you often need to access the state of the robots multiple time each time step(e.g observation and rewards). Should you read from the ArticulationView multiple times or read once and store in an internal buffer? What is more efficient/best practice.
Hi,
For RL task you often need to access the state of the robots multiple time each time step(e.g observation and rewards). Should you read from the ArticulationView multiple times or read once and store in an internal buffer? What is more efficient/best practice.