Motor System Improvements

Improvements we would like to add to the motor system.

These are the things we would like to implement:

• Use a different policy for learning than for inference #learning
• Bottom-Up exploration policy for surface agent #learning
• Top-down exploration policy #learning #numsteps
• Bottom up policies for distant agent #numsteps
• Calculate saliency map in view finder and use it to saccade to region of interest #numsteps #multiobj
• Learn policy using RL (using simplified action space) #numsteps
• Decompose goals into subgoals & communicate #goalpolicy
• Interprete goal states in motor system & switch policies #goalpolicy

Please see the instructions here if you would like to tackle one of these tasks.