Sys Agent: move_right_up(0) move_right_down(1) move_left_up(2) move_left_down(3) nop(4) 
Env Agent: move_right(0) move_left(1) move_up(2) move_down(3) move_right_up(4) move_right_down(5) move_left_up(6) move_left_down(7) nop(8) 

EnvAct           SysAct           hx      hy      px      py      
