An In-depth Look at STRIPS in AI

An In-depth Look at STRIPS in AI

Artificial Intelligence (AI) is a broad field with many sub-disciplines and methodologies. One such methodology is STRIPS (Stanford Research Institute Problem Solver), a system for automated planning developed in the late 1960s at the Stanford Research Institute. STRIPS has played a significant role in the history of AI and continues to influence contemporary AI planning systems. This article explores what STRIPS is, how it works, and its applications and implications in the AI field.

What is STRIPS?

STRIPS, an acronym for Stanford Research Institute Problem Solver, is a critical component in the realm of Artificial Intelligence (AI). Essentially, it is both a language and a mechanism developed with the objective of outlining actions within AI systems. STRIPS is predominantly used in planning systems, where it equips an AI with the capacity to outline a series of actions necessary to achieve a predefined goal, given an initial state.

The elegance of STRIPS lies in its simplicity. It employs logical representations to depict the world, facilitating an understanding of the actions that can be performed and the desired outcomes of these actions. The world, in the context of STRIPS, is modeled as a collection of logical propositions. A logical proposition refers to a statement about the world which can either be true or false. For example, the statement “the sky is blue” can be a proposition as it can be true under clear weather conditions and false when it’s cloudy or dark.

In the world defined by STRIPS, actions serve as transformations on these propositions. Each action within the system has a specific set of preconditions that must be met for the action to take place. Moreover, every action will result in certain propositions being added to the current state (making them true), and certain others being removed (making them false).

For instance, consider an AI system that is navigating a virtual world with rooms and doors. A proposition could be “AI is in Room A”. An action could be “AI moves to Room B”, which has the precondition “A door connects Room A and Room B”. Upon performing this action, the proposition “AI is in Room A” would be removed and the proposition “AI is in Room B” would be added.

Thus, through its unique language and mechanism, STRIPS provides a powerful framework for automated planning in AI. It empowers an AI system to predict the outcomes of its actions and select the best actions to accomplish its goals, given its understanding of the current state of the world.

How Does STRIPS Work?

In order to understand the functioning of STRIPS (Stanford Research Institute Problem Solver) within the realm of Artificial Intelligence, it is important to first grasp the concept of ‘states and propositions’, ‘actions’, and the ‘planning’ process.

States and Propositions

At its very core, STRIPS begins by defining the world as a set of propositions. A proposition, in the context of STRIPS, is a statement about the world that can be evaluated as true or false. These propositions together, at a particular instance, define the ‘state’ of the world. The state, therefore, is a snapshot of the world at a given time, represented by the collective truth values of the propositions.

As an example, consider a simplified scenario where a robotic arm is tasked to manipulate blocks. A proposition in this context could be “Block1 is on the table”. In this case, the state of the world includes the position of Block1 and the status of the robot’s interaction with it.


Once the state of the world is defined, STRIPS then moves towards actions. In the STRIPS framework, actions are the primary agents of change. An action, when executed, transforms the world from one state to another. Each action within STRIPS is defined with three key components:

  1. Preconditions: Preconditions are the set of propositions that need to be true in the current state of the world for the action to be executable. These set the stage for the action to occur.
  2. Add list: The add list consists of propositions that would be true in the state of the world after the action has been performed. These propositions are the result of the changes brought about by the action.
  3. Delete list: The delete list includes the propositions that would be rendered false after the action.

In the case of the action “Pickup(Block1)”, for example, the preconditions could be “Block1 is on the table” and “Robot hand is empty”. The add list would contain the proposition “Robot hand holds Block1”, whereas the delete list would include “Block1 is on the table” and “Robot hand is empty”.


The last pillar of the STRIPS mechanism is the planning process. Given an initial state, a goal state, and a set of available actions, the STRIPS planning system attempts to devise a sequence of actions that can successfully transition the world from the initial to the goal state.

The goal state, similar to the initial state, is a set of propositions that we wish to be true. The planning process, therefore, involves navigating the space of all possible action sequences to identify a path that leads from the initial to the goal state.

In sum, the STRIPS framework offers a robust model for defining the world’s state, laying out the possible actions, and planning a path from a given initial state to a desired goal state, making it an effective tool for problem-solving in AI.

Applications of STRIPS

STRIPS (Stanford Research Institute Problem Solver) has found a significant place in the field of artificial intelligence due to its powerful problem-solving capabilities. Its influence spans a broad spectrum of AI applications, especially in planning and problem-solving tasks. Here are some of the domains where STRIPS has been extensively applied:


STRIPS is also extensively used in the logistics domain. Here, it helps in planning and routing, which are essential for supply chain and transportation management. For example, in a warehouse, a STRIPS-based system can help plan the optimal sequence of actions to pick up items from different locations and deliver them to the packing area. Similarly, in transportation networks, STRIPS can be used to plan the best routes for delivering packages, considering various factors such as traffic, distance, and delivery times.

Game AI

In the gaming industry, STRIPS has been utilized in creating AI opponents or non-player characters (NPCs). Here, it helps in making the NPCs more intelligent by enabling them to plan and execute actions based on the current state of the game world. This enhances the realism and the challenge of the game, leading to a more engaging experience for the players.

Despite its inception dating back to the 1960s, the principles of STRIPS continue to play a substantial role in contemporary AI systems. Many modern planning languages and systems have their roots in STRIPS and have expanded upon its basic framework to handle more complex and dynamic environments. They incorporate additional features such as conditional effects, probabilities, or time constraints, which are necessary to deal with the increasing complexity of today’s AI applications. Thus, STRIPS continues to serve as a fundamental building block in the ever-evolving landscape of artificial intelligence.


In conclusion, STRIPS (Stanford Research Institute Problem Solver) has made a significant impact on the field of artificial intelligence since its inception. Its ability to represent the state of the world as a set of propositions, describe actions as transformations of these propositions, and plan sequences of actions to reach a goal state has revolutionized the way AI systems solve problems.

While STRIPS itself is a relatively simple and elegant framework, it has served as the foundation for many of the more sophisticated and powerful AI systems we see today. From robotics to logistics to game AI, the principles of STRIPS continue to be employed and expanded upon, demonstrating their enduring value.

As AI continues to advance, it is likely that the influence of STRIPS will persist, guiding the development of new techniques and approaches for planning and problem-solving in complex, dynamic environments. The enduring relevance of STRIPS in the fast-evolving AI landscape is a testament to its robustness and versatility, making it a key component in the toolkit of any AI practitioner.

Further Reading

  1. STRIPS: A New Approach to the Application of Theorem Proving to Problem Solving: This is the original paper where STRIPS was first introduced. It’s a valuable resource for anyone wanting to delve into the historical and theoretical foundations of STRIPS.
  2. Artificial Intelligence: A Modern Approach: This is a comprehensive AI textbook by Stuart Russell and Peter Norvig. It covers a broad range of AI topics, including planning and STRIPS.
  3. STRIPS Wikipedia Page: The Wikipedia page on STRIPS provides a good overview of the subject and includes numerous links to related topics.
  4. Artificial Intelligence: Structures and Strategies for Complex Problem Solving: This book by George F. Luger has a detailed section on STRIPS and other AI planning methods. It presents a broad introduction to AI with a focus on problem-solving strategies.
  5. Introduction to Artificial Intelligence Planning: This thesis by Drew McDermott provides a comprehensive introduction to AI planning, including the role of systems like STRIPS.