Possible Wumpus Worlds

HilaryHolz

1. walk to the point shown in 7.4(a)
   1. percept at [1,1] is [ None, None, None, None, None ]
   2. move([1,1],[2,1])
   3. percept at [2,1] is [ None, Breeze, None, None, None ]
   4. l90
   5. l90
   6. move([2,1],[1,1])
   7. r90
   8. move([1,1],[1,2])
   9. percept at [1,2] is [ Stench, None, None, None, None ]

-- HilaryHolz - 20 Oct 2008

MichaelKhuong

- Michael's Solutions after finding out what and how to do it after Hilary's example in class.

N = nothing.

W = Wumpus.

P = Pit.

KavithaAshwin

Section 7.2:-

* Kavitha's Report on Wumpus World @ "Mutualdiscovery.WumpusWorld.ppt"

Where: "[S=Stench,B=Breeze,G=Gold,P=Pit,W=Wumpus]"

-- KavithaAshwin - 21 Oct 2008

FaridNabavi

• Possible Wumpus Worlds

Possible Wumpus Worlds (SUSHANT KADADI)

• Possible_Wumpus_Worlds.ppt: Possible Wumpus World Models

People.Jyothirmai Vissapragada=

Introduction to Wumpus World A variety of "worlds" are being used as examples for Knowledge Representation, Reasoning, and Planning. Among them the Vacuum World, the Block World, and the Wumpus World. We will examine the Wumpus World and in this context introduce the Situation Calculus, the Frame Problem, and a variety of axioms. The Wumpus World was introduced by Genesereth, and is discussed in Russell-Norvig. The Wumpus World is a simple world (as is the Block World) for which to represent knowledge and to reason. It is a cave with a number of rooms, represented as a 4x4 square. The Wumpus world is a grid of squares surrounded by walls, where each square can contain agents and objects. The agent (you) always starts in the lower left corner, a square that will be labeled [1,1]. The agent's task is to find the gold, return to [1,1] and climb out of the cave. Report on Agent Syntax Lab - PART -A About Row-major order and Column-major Order Row- major order and Column-major order and two types techniques in arranging and storing arrays.Rows are identified by the first index of a two-dimensional array and columns by the second index.Column-major order is a similar method of flattening arrays onto linear memory, but the columns are listed in sequence.

When using row-major order, the difference between addresses of array cells in increasing rows is larger than addresses of cells in increasing columns. For example, consider this 2×3 array: 1 2 3 4 5 6

Column-major order is a similar method of flattening arrays onto linear memory, but the columns are listed in sequence. The programming languages Fortran and MATLAB use column-major ordering. The array

1 2 3 4 5 6 Treating a row-major array as a column-major array is the same as transposing it. Because performing a transpose requires data movement, and is quite difficult to do in-place for non-square matrices, such transpositions are rarely performed explicitly. For example, software libraries for linear algebra, such as the BLAS, typically provide options to specify that certain matrices are to be interpreted in transposed order to avoid the necessity of data movement. Column-major oder Column-major order is similar method of flattening arrays into linear memory, but the columns are listed in sequence. Let us consider the above example again 1 2 3 4 5 6 In a column-major order the tuples would be represented as : (1,1)(2,1)(1,2)(2,2)(1,3)(2,3) When stored in the memory, it would look like : 1 4 2 5 2 6 Order used by Python Python use row-major order. Knowing this would help a programmer to undestand the manner in which arrays are processed by the Python interpreter. Partially observable The wumpus is partially observable because it does not have the information of the full environment. The vacuum cannot tell if they are a wumpus near by because of it lack of information about the environment. Deterministic The agent is deterministic because it has the ability to call back on it pervious action and determine what the best logical steps to move on to not get eaten by the wumpus. Sequential The agent is sequential because all of it past and future decision could be effected if it does something wrong. So in essential, the agent has to think ahead of what are some of the possible move(s) that it can perform to accomplish its task and not get eaten by the wumpus. Static The environment that the agent has to travel to is static. Everything in the environment is already predefined when the program starts. Discrete The state of the environment is discrete because it has a finite number of moves before there no more moves left to be done. Single agent The agent is a single agent. The agent is the intelligent object that moves around the environment while the other stuffs in the environment is set to do a static set of tasks. Agent Syntax,Propositions,percepts and actions Agent in a particular location or square A[i,j] Wumpus in a particular location or square W[i,j] Pit in a particular location or square P[i,j] Gold particular location or square G[i,j] If a wumpus is killed in a particular location or square W[i,j] Percepts at a particular square [i,j] [Stench, Breeze, Glitter, Bump, Scream][i,j] If there is either a breeze or a stench from a particular square [i,j] then the agent perceives this from the adjacent square. That is [i+1,j],[i,j+1] or [i-1,j],[i,j-1] This is the case for a wumpus and a pit too. 7.1 Describe the wumpus world according to the properties of task environments listed in Chapter 2 From closer observation of the environment and the agent, we described the wumpus world as : • Partially observable - The agent does not have access to the complete state of the environment at each point in time. The agent knows the state of the current location. From the perceptions received at the current point, the agent can come to know the state of the neighboring locations only • We could not conclude whether the wumpus world is deterministic or stochastic. Further analysis showed that the environment being strategic might be a better option. But again, it depends upon the type of the agent and it's capabilities. • The wumpus world environment is sequential. It cannot be episodic because the actions at each state is not atomic. In the environment of our consideration, the perceptions or the experience of the agent at one particular state has it's impact on the subsequent states and actions.

7.2 Construct a set of possible wumpus worlds (32 of them possible) In the classs, we came up with 32 possible wumpus worlds. We used the following notation for each state • V for Visited • OK for no Wumpus or Pit • B for Breeze • S for Stench • P for Pit • W for Wumpus Available Knowledge Base = ((1,1) = V.OK) ∧ ((2,1) = B.V.OK) ∧ ((1,2) = S,V,OK) Given statements : • α = "There is no pit in [2,2]" • β = "There is a wumpus in [1,3]"

Model (1,1) (2,1) (3,1) (1,2) (2,2) (1,3) KB? α β

1 V,OK B,V,OK OK S,V,OK OK OK no T F

2 V,OK B,V,OK OK S,V,OK OK P no T F 3 V,OK B,V,OK OK S,V,OK OK W no T T 4 V,OK B,V,OK OK S,V,OK OK P,W no T T 5 V,OK B,V,OK OK S,V,OK P OK no F F 6 V,OK B,V,OK OK S,V,OK P P no F F 7 V,OK B,V,OK OK S,V,OK P W no F T 8 V,OK B,V,OK OK S,V,OK P P,W no F T 9 V,OK B,V,OK OK S,V,OK W OK no T F 10 V,OK B,V,OK OK S,V,OK W P no T F 11 V,OK B,V,OK OK S,V,OK P,W OK no F F 12 V,OK B,V,OK OK S,V,OK P,W P no F F 13 V,OK B,V,OK P S,V,OK OK OK no T F 14 V,OK B,V,OK P S,V,OK OK P no T F 15 V,OK B,V,OK P S,V,OK OK W yes T T 16 V,OK B,V,OK P S,V,OK OK P,W no T T 17 V,OK B,V,OK P S,V,OK P OK no F F 18 V,OK B,V,OK P S,V,OK P P no F F 19 V,OK B,V,OK P S,V,OK P W no F T 20 V,OK B,V,OK P S,V,OK P P,W no F T 21 V,OK B,V,OK P S,V,OK W OK no T F 22 V,OK B,V,OK P S,V,OK W P no T F 23 V,OK B,V,OK P S,V,OK P,W OK no F T 24 V,OK B,V,OK P S,V,OK P,W P no F T 25 V,OK B,V,OK W S,V,OK OK OK no T F 26 V,OK B,V,OK W S,V,OK OK P no T F 27 V,OK B,V,OK W S,V,OK P OK no F F 28 V,OK B,V,OK W S,V,OK P P no F F 29 V,OK B,V,OK P,W S,V,OK OK OK no T F 30 V,OK B,V,OK P,W S,V,OK OK P no T F 31 V,OK B,V,OK P,W S,V,OK P OK no F F 32 V,OK B,V,OK P,W S,V,OK P P no F F

DevasenaKopparty

Here are 32 possible wumpus worlds.

For each Wumpus Environment, we use the following check words representing each grid"

• V for Visited
• Safe for no Wumpus or Pit
• B for Breeze
• S for Stench
• P for Pit
• W for Wumpus

Available Knowledge Base = ((1,1) = V.OK) ∧ ((2,1) = B.V.OK) ∧ ((1,2) = S,V,OK)

Given statements :

No_pit = There is no pit in the grid [2,2]

Is_Wumpus= wumpus is in the grid [1,3]

People.Shravani Nimma

* possiblewumpusworld.doc: Shravani's Possible Wumpus World Models

People.Kiran Nanjundan

* PossibleWumpusWorld1.ppt: Possible Wumpus world models

People.Niraj Soni

* Wumpus_World.doc: Niraj's Possible Wumpus World

People.Indraneel Yadav

* Posiible_Wumpus_World_by_indraneel.doc: Possible Wumpus Worlds by Indraneel Yadav

ShikhaSharma

In the wumpus world agents

• should collect gold from a rectangular world.
• should avoid the pits, which are detectable by its cool breeze.
• should kill the wumpus, which are detectable by its stench, by firing only 1 available arrow.

A simple knowledge based agent, accepts a precept and returns an action. Following possible Percepts are considered for

We start this problem after the agents third move with percept[Stench, None, None, None, None]. Till this stage agents has perceived nothing in [1,1], a breeze in [2,1] and a stench in [1,2].

Problem asks us to mark the worlds in which the KB is true and those in which following is true,

• α2 = "There is no pit in [2,2]"
• α3 = "There is a wumpus in [1,3]"

Since agent is now concerned with the contents of [3,1], [2,2] and [1,3]and we know that each of these can contain a pit and atmost one can contain a wumpus, we end up with 32 possible worlds.

The only world in which the KB criterion is satisfied is world no. 15. In this world both α2 and α3 are true, also there is no pit in [2,2] and also there is a wumpus in [1,3].

HsienErnLan

* Mutualdiscovery.WumpusWorld.doc: Please see the attachment.

Paul Trinh

• Possible_wumpus_worlds.doc: Please see the attached file below.

Karthik Sanyasi

• PossibleWumpusWorld1.doc: Possible Wumpus World

TapasyaAgarwal

A simple knowledge based agent, accepts a precept and returns an action. Following possible Percepts are considered for

“V” for agent has visited the square

“OK" for safe square, i.e. no Wumpus or Pit in the square

“B” for Breeze i.e. There is a pit adjacent to the current square

"S" for Stench i.e. There is a wumpus adjacent to the current square

"P" for Pit in the square

"W" for Wumpus in the square

We start this problem after the agents third move with percept[Stench, None, None, None, None]. Till this stage agents has perceived nothing in [1,1], a breeze in [2,1] and a stench in [1,2].

Problem asks us to mark the worlds in which the KB is true and those in which following is true,

• α2 = "There is no pit in [2,2]"
• α3 = "There is a wumpus in [1,3]"

Since agent is now concerned with the contents of [3,1], [2,2] and [1,3]and we know that each of these can contain a pit and atmost one can contain a wumpus, we end up with 32 possible worlds.

The only world in which the KB criterion is satisfied is world no. 15. In this world both α2 and α3 are true, also there is no pit in [2,2] and also there is a wumpus in [1,3].

SoyJoseph

• Wumpus_World.doc: Wumpus_World.doc

AnithaPadigapati

The wumpus world agents:

Should collect gold from a rectangular world. Avoid pits, which are detectable by its cool breeze. and should kill the wumpus.

A simple knowledge based agent, accepts a precept and returns an action. Following possible Percepts are considered for

“V” for agent has visited the square. “OK" for safe square, i.e. no Wumpus or Pit in the square. “B” for Breeze i.e. There is a pit adjacent to the current square. "S" for Stench i.e. There is a wumpus adjacent to the current square. "P" for Pit in the square. "W" for Wumpus in the square.

We start this problem after the agents third move with percept[Stench, None, None, None, None]. Till this stage agents has perceived nothing in [1,1], a breeze in [2,1] and a stench in [1,2].

AS agent is now concerned with the contents of [3,1], [2,2] and [1,3]and we know that each of these can contain a pit and atmost one can contain a wumpus, we end up with 32 possible worlds.

Only world in which the KB criterion is satisfied is world no. 15. In this world both α2 and α3 are true, also there is no pit in [2,2] and also there is a wumpus in [1,3].

GopikrishnaNagaraja

PROPERTIES OF TASK ENVIRONMENTS

Agent Syntax,Propositions,percepts and actions

• Agent in a particular location or square A[i,j]
• Wumpus in a particular location or square W[i,j]
• Pit in a particular location or square P[i,j]
• Gold particular location or square G[i,j]
• If a wumpus is killed in a particular location or square W[i,j]

I choose to explain in the way that concentrates on the actual agent movement from the start at square[1,1] to the square[1,3] where he kills the wumpus. The 32 possibility that are represented in a table in redundant information. what I mean to say is that focusing the path that the Agent takes will yield a more comprehensible explanation. the table format consumes time and slightly confuses the reader. One more point is that everyone ends up with the same table which will hinder the individuality of the concept behind it.

JasonSauer

The agent has perceived nothing in [1,1], a breeze in [2,1], and a stench in [1,2]. The following are the 32 possible worlds:

Anup Mohan

Since agent is now concerned with the contents of [3,1], [2,2] and [1,3]and we know that each of these can contain a pit and atmost one can contain a wumpus, we end up with 32 possible worlds.

The only world in which the KB criterion is satisfied is world no. 15. In this world both α2 and α3 are true, also there is no pit in [2,2] and also there is a wumpus in [1,3].