Genetic Scheduling and Reinforcement Learning in Multirobot Systems for Intelligent Warehouses

Algorithm 2

-learning for path planning.

(1) Initialize the total episodes of learning Maxepi, maximum step number Maxstep, discount rate ;

(2) Initialize the coordinate of the starting and ending point;

(3) Initialize based on the actual grid world;

(4)  for to Maxepi do

(5)  Initialize the learning rate ;

(6)  for to Maxstep do

(7)    Decrease the learning rate gradually;

(8)    Choose from by greedy action selection;

(9)    Take action , observe , ;

(10)  

(11)     

(12)     if is a goal state then

(13)     Break;

(14)     end if

(15)   end for

(16) end for