Ans(a)- If the environment is fully observable and episodic, then the simple reflex agent can make the right decision to maximise its profit.
	
Ans(b)- No, its not. The robot when surrounded by more than one squares with maximum dirt, then it makes a random choice and this randomness 
can result in total different moves of the robot.

Ans(c)- No, such an environment cannot be designed as given any environment, there has to be atleast a single optimal path that maximises the 
result, and that path can be retraced by a randomized agent even though the probability of that can be very less.
	Hence, there cannot be an environment for which the randomized agent will always perform poorly.

Ans(d)- The robot can calculate how much was the difference in the sensed and the actual dirt (assuming it can weigh the total dirt inside it, 
then difference of value after and before sucking will give the weight of dirt present on the cell). Initially the agent will traverse for a 
short period of time assuming that the input is correct and then updating in the memory the actual value of dirt in the cell and the error in 
the measurement in that cell. So, after some time when it will read the new cell, it will try to see the difference in the observed value with 
the averaged value of the neighbouring cells (visited and observed both) taking into account the error calculated at the visited ones. This 
will be done for all the cells to which it can move. Then, assuming the continuity in the distribution of dirt (i.e. the differnce in the 
dirt value of a cell and those of its neighbouring cells can't be very high), it will calculate a possible value of dirt on the cell by 
assigning some probability of error in the sensing using the thinking as stated above. Then, out of those possible values, the next move 
will be decided. The calculation of possible dirt values can be approximated by running some test cases initially by the experimenter and 
then coming up with a function or a distribution that applies in most of the test case environments.

Ans(e)- In this case, the rational agent should calculate the probability of reapperance of dirt on the cleaned cells and when 
calculating the moves, it should not ignore the cells but should consider the high probability cells to be uncleaned (or unseen). i.e. in 
this case, agent should not refrain to make a move to a cell surrounded by clean cells if the probability of reappearance of dirt in the 
surrounding cells is not very low.