GridEnvironment6.py

def get_agent_states(self):
            
            start_state  = None
            target_state = None
            start_state = list(map(lambda x:x[0] if len(x) > 0 else None, np.where(insert_grid_map==4)))
            
            target_state = list(map(lambda x:x[0] if len(x) > 0 else None, np.where(insert_grid_map==3)))
            
            if (start_state == [None, None] or target_state == [None, None]):
                sys.exit('Start or Target state not specified')
            
            if (start_state == self.obstacle_mask):
                sys.exit('Obstacle encountered GAME OVER !!!')
                
                return start_state, target_state
            
 def _render_(self, mode='human', close=False):
            
            if self.verbose == False:
                
                return
            img = self.observation
            fig = plt.figure(self.fig_num)
            plt.clf()
            plt.imshow(img)
            fig.canvas.draw()
            plt.pause(0.0002)
            return
        
        #configuring agent's states
        
        def _get_agent_state(self):

            return self.agent_state
        
        def get_agent__start_target_state(self):
            
            return self.start_state, self.target_state
        
        
        def jump_to_a_state(self, to_state):
            
            if self.current_map[to_state[0], to_state[1]] == self.current_map[self.obstacle_mask[0], self.obstacle_mask[1]]:
                    sys.exit("GAME OVER")
                    
            elif self.current_map[to_state[0], to_state[1]] == 0:
                
                if self.current_map[agent_state[0], agent_state[1]] == 4:
                    
                    self.current_map[agent_state[0], agent_state[1]] += self.actions_pos_dict[actions][2]
                    self.observation = self.grid_map_to_observation(self.current_map)
                    self.agent_state = [to_state[0], to_state[1]]
                    self._render_()
                    return (self.observation, agent_state)
                    
                if self.current_map[agent_state[0], agent_state[1]]== 6:
                    
                    self.current_map[agent_state[0], agent_state[1]] += self.actions_pos_dict[actions][1]
                    self.observation = self.grid_map_to_observation(self.current_map)
                    self.agent_state = [to_state[0], to_state[1]]
                    self._render_()
                    return (self.observation, agent_state)
                
                if self.current_map[agent_state[0], agent_state[1]]== 7:
                    
                    self.current_map[agent_state[0], agent_state[1]] += self.actions_pos_dict[actions][3]
                    self.observation = self.grid_map_to_observation(self.current_map)
                    self.agent_state = [to_state[0], to_state[1]]
                    self._render_()
                    return (self.observation, agent_state)
            
            elif self.current_map[to_state[0], to_state[1]] is None:
                
                    return ("Invalid state")
                
            """
            please add the to_state if condition for the remaining states i.e 4, 1, 3, and 7.
            
            """
            
            def close(self):
                plt.close(1)
                return