Chapter: 7 Planning:

Q) Discuss Goal Stack planning – S21, S24, S25 (4M)

Goal Stack Planning
 Goal Stack Planning is one of the simplest planning algorithms that is
designed to handle problems which include compound goals.
 It utilizes STRIPS as a formal language for specifying and manipulating
the world with which it is working.
 This approach uses a Stack for plan generation. The stack can contain
Sub-goal and actions described using predicates. The Sub-goals can be
solved one by one in any order.
 It starts by pushing the unsatisfied goals into the stack.
 Then it pushes the individual sub-goals into the stack and its pops an
element out of the stack.
 When popping an element out of the stack the element could be,
o either a predicate describing a situation about our world or
o it could be an action that can be applied to our world under
consideration.
 So, a decision has to be made based on the kind of element we are
popping out from the stack.
 If it is a Predicate, then compare it with the description of the current
world, if it is satisfied or is already present in our current situation then
there is nothing to do because it’s already true.
 On the contrary if the Predicate is not true then we have to select and
push relevant action satisfying the predicate to the Stack.
 So after pushing the relevant action into the stack its precondition should
also have to be pushed into the stack.
 In order to apply an operation its precondition has to be satisfied, i.e.,
the present situation of the world should be suitable enough to apply an
operation.
 For that, the preconditions are pushed into the stack once after an action
is pushed.
Q) Explain planning problem. – S22 (3M)
A Planning Problem in Artificial Intelligence refers to the process of finding a sequence of
actions that leads from an initial state to a goal state.
Explanation:
 Planning is an important part of AI that deals with deciding a series of actions to
achieve a specific goal.
 The planner uses a model of the world, usually described in terms of states, actions,
and goals.
 Each action has:
o Preconditions – conditions that must be true before the action is performed.
o Effects – conditions that become true after the action is performed.
Components of a Planning Problem:
1. Initial State:
The starting situation of the agent or world.
2. Goal State:
The desired situation or objective that the agent wants to achieve.
3. Actions/Operators:
The set of actions available, each with preconditions and effects.
4. Plan:
The sequence of actions that transforms the initial state into the goal state.
Example:
If a robot wants to move from room A to room C through room B:
 Initial state: Robot in room A
 Goal state: Robot in room C
 Actions: Move(A→B), Move(B→C)
 Plan: [Move(A→B), Move(B→C)]

Q) Explain how planning is different from search procedure? – S22 (3M)

1. Planning:
 Planning explicitly represents the states, goals, and actions using
predicates and logic-based descriptions.
 It focuses on what actions should be performed and in what order to
achieve the goal.
  Uses knowledge representation (like STRIPS) to describe the world and
action effects.
 Handles complex and compound goals efficiently.
 Example: Goal Stack Planning, Partial Order Planning.
2. Search Procedure:
 Search represents the problem as a graph or tree of states connected by
actions.
 It focuses on how to traverse the state space to reach the goal.
 Operates at a lower level — often just considers states and transitions, not
logical predicates.
 Common examples: Breadth-First Search, Depth-First Search, A*
algorithm.
Q) Explain the components of planning system. –W22, W24 (3/4M)
A Planning System in Artificial Intelligence is designed to generate a sequence of actions that
transforms the initial state of the world into the desired goal state.
It consists of several key components that work together to achieve this.
1. Initial State
 It describes the starting situation of the agent or the environment.
 Defines all facts that are true at the beginning of the problem.
 Example: Robot is in Room A, Door is closed.
2. Goal State
 Specifies the desired situation that the planner must achieve.
 Defines conditions that must be true when the plan is complete.
 Example: Robot should be in Room B, Door should be open.
3. Actions / Operators
 Represent the possible actions the agent can perform.
 Each action has:
o Preconditions – Conditions that must be true before the action can be applied.
o Effects – The result of performing the action (changes in the world state).
 Example:
o Action: Open Door
 o Preconditions: Door is closed
o Effects: Door becomes open
4. Plan
 A sequence of actions generated by the planner that leads from the initial state to the
goal state.
 The plan must ensure all preconditions are satisfied before each action is executed.
5. World Model / Knowledge Base
 Describes the representation of the environment in which planning occurs.
 Includes facts, relationships, and constraints among different objects and actions.
6. Planning Algorithm (Planner)
 The mechanism or procedure used to search for or construct a valid plan.
 Examples: Goal Stack Planning, STRIPS Planner, Partial Order Planning, etc.

Q) Explain Hierarchical planning in brief. – S25 (3 M)

Hierarchical Planning
 In order to solve hard problems, a problem solver may have to generate long plans.
 It is important to be able to eliminate some of the details of the problem until a
solution that addresses the main issues is found.
 Then an attempt can be made to fill in the appropriate details.
 Early attempts to do this involved the use of macro operators, in which larger
operators were built from smaller ones.
 In this approach, no details were eliminated from actual descriptions of the operators.
Definition:
Hierarchical Planning is a planning approach in which a complex problem is solved by
breaking it down into smaller sub-problems or by planning at different levels of abstraction.
Explanation:
 In difficult problems, generating a full detailed plan at once can be very complex.
 Hierarchical planning simplifies this by solving the problem at a higher (abstract)
level first, ignoring unnecessary details.
 Once a general (high-level) plan is created, the details are filled in at lower levels.
 This reduces complexity and makes planning faster and more manageable.
Example:
If a robot needs to clean a house,
  High-level plan: Clean the house.
 Sub-goals: Clean kitchen, clean bedroom, clean hall.
 Each sub-goal can further be divided into smaller actions like “pick broom,” “sweep
floor,” etc.

Q) Explain Reactive System.

Reactive Systems
 The idea of reactive systems is to avoid planning altogether, and instead, use the
observable situation as a clue to which one can simply react.
 A reactive system must have an access to a knowledge base of some sort that
describes what actions should be taken under what circumstances.
 A reactive system is very different from the other kinds of planning systems we have
discussed. Because it chooses actions one at a time.
 It does not anticipate and select an entire action sequence before it does the first thing.
 The example is a Thermostat. The job of the thermostat is to keep the temperature
constant inside a room.
 Reactive systems are capable of complex behaviors.
 The main advantage reactive systems have over traditional planners is that they
operate robustly in domains that are difficult to model completely and accurately.
 Reactive systems dispense with modeling altogether and base their actions directly on
their perception of the world.
 Another advantage of reactive systems is that they are extremely responsive since
they avoid the combinatorial explosion involved in deliberative planning.
 This makes them attractive for real-time tasks such as driving and walking.