Scribd
Upload
95 views

ACSC 368 - Artificial Intelligence: Coursework 1

This document discusses evaluation and cost functions for solving a blocks world problem using different search algorithms. The evaluation function counts the number of blocks not on position p3. The cost function assigns a cost of 2 minutes to move a block to an adjacent position or 3 minutes to move further. Depth-first, breadth-first, hill-climbing, and best-path search algorithms are to be applied to develop the search trees in order of state expansion.

Uploaded by

Zerihun Bekele
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
95 views

ACSC 368 - Artificial Intelligence: Coursework 1

This document discusses evaluation and cost functions for solving a blocks world problem using different search algorithms. The evaluation function counts the number of blocks not on position p3. The cost function assigns a cost of 2 minutes to move a block to an adjacent position or 3 minutes to move further. Depth-first, breadth-first, hill-climbing, and best-path search algorithms are to be applied to develop the search trees in order of state expansion.

Uploaded by

Zerihun Bekele
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

ACSC 368 – Artificial Intelligence: Coursework 1

Recall the ‘Blocks World’ problem from Homework 1. We can use the following
evaluation and cost functions in solving it:

Evaluation function

Since in our goal state all blocks are on p3, our evaluation function can be defined as:

The number of blocks not on p3.

So for example the evaluation value of the state [on(a,p1), on(b,p2)] will be 2 and that
of the state [on(a,p3), on(b,p2)] will be 1. (The evaluation value of the goal state will
of course be 0.)

Cost Function

We calculate the cost of transitioning from one state to another in terms of the time it
takes to move a block. It is assumed that it takes 2 minutes to move a block to an
adjacent position and 3 minutes to move a block from either p1 or p3 to the position at
the other end (i.e. p3 or p1 respectively). So our cost function can be defined as:

2, if a block is being moved to an adjacent position,


cost =
3, otherwise.

Using this information (wherever needed) develop the search tree created by each of
the following search algorithms for this problem, making clear the order in which
states are expanded:

(a) Depth-first

(b) Breadth-first

(c) Hill-climbing

(d) Best-path. In this case you are also required to detail the path-lists considered by
the algorithm, as well as the ones resulting from the expansion of each path.

You might also like