Our Approach

Veloxiti's Approach to Intelligent Decision Making

Our approach to building intelligent systems draws on ideas from three sources: a fighter pilot, a philosopher, and a computer scientist.


U.S. Air Force plane

Colonel John Boyd (U.S. Air Force) made two significant contributions as a fighter pilot and military strategist. They were Energy-Maneuverability theory and the OODA Loop. Col Boyd, a skilled U.S. jet fighter pilot in the Korean War, began developing the Energy-maneuverability theory in the early 1960s. He teamed with mathematician Thomas Christie at Eglin Air Force Base to use the base's high-speed computer to compare the performance envelopes of U.S. and Soviet aircraft from the Korean and Vietnam Wars.

Energy–maneuverability theory is a model of aircraft performance. It is useful in describing an aircraft's performance as the total of kinetic and potential energies or aircraft specific energy. It relates the thrust, weight, drag, wing area, and other flight characteristics of an aircraft into a quantitative model. This allows combat capabilities of various aircraft or prospective design trade-offs to be predicted and compared. Based on Boyd’s previous success he was kept on as a key military air-to-air combat strategist.

Boyd and Christie completed a two-volume report on their studies in 1964. Energy Maneuverability came to be accepted within the U.S. Air Force and brought about improvements in the requirements for the F-15 Eagle and later the F-16 Fighting Falcon fighters.

OODA Loop graphic

Boyd's next key concept was that of the decision cycle or OODA Loop, the process by which an entity (either an individual or an organization) reacts to an event. According to this idea, the key to victory is to be able to create situations wherein one can make appropriate decisions more quickly than one's opponent.

The construct was originally a theory of achieving success in air-to-air combat, developed out of Boyd's earlier Energy-maneuverability theory and his observations on air combat between MiG-15 and North American F-86 Sabre aircraft in Korea. Harry Hillaker (chief designer of the F-16) said of the OODA theory, "Time is the dominant parameter. The pilot who goes through the OODA cycle in the shortest time prevails because his opponent is caught responding to situations that have already changed." Boyd hypothesized that all intelligent organisms and organizations undergo a continuous cycle of interaction with their environment.

Boyd breaks this OODA cycle (Figure 1) down to four interrelated and overlapping processes through which one cycles continuously:

OODA graphic

Figure 1: The OODA Loop Enables Continuous Adaptation to Continuously Changing Situations

  • Observation: the collection of data by means of the senses

  • Orientation: the analysis and synthesis of data to form one's current mental perspective

  • Decision: the determination of a course of action based on one's current mental perspective

  • Action: the physical playing-out of decisions

Of course, while this is taking place, the situation may be changing. It is sometimes necessary to cancel a planned action in order to meet the changes. This decision cycle is thus known as the OODA loop. Boyd emphasized that this decision cycle is the central mechanism enabling adaptation (apart from natural selection) and is therefore critical to survival.

Boyd theorized that large organizations such as corporations, governments, or militaries possessed a hierarchy of OODA loops at tactical, grand-tactical (operational art), and strategic levels. In addition, he stated that most effective organizations have a highly decentralized chain of command that utilizes objective-driven orders, or directive control, rather than method-driven orders in order to harness the mental capacity and creative abilities of individual commanders at each level.

In 2003, this power to the edge concept took the form of a DOD publication "Power to the Edge: Command ... Control ... in the Information Age" by Dr. David S. Alberts and Richard E. Hayes. Boyd argued that such a structure creates a flexible "organic whole" that is quicker to adapt to rapidly changing situations.

Veloxiti’s approach to artificial intelligence has been heavily influenced by Boyd’s problem solving approach, so much so that our development toolkit explicitly models the OODA loop.


Michael Bratman is a Professor of Philosophy at Stanford University who has developed a theory of human reasoning known as the Belief-Desire-Intention (BDI) model. Broadly speaking Bratman is concerned with human action: how is it that humans can act rationally? Bratman argues that reasoning is fundamentally concerned with beliefs, desires, and intentions.

Bratman’s ideas are surprisingly consistent with John Boyd’s view of the world. Consider the case of a fighter pilot flying a Combat Air Patrol (CAP). The CAP mission focuses on defeating or destroying all enemy aircraft within a combat area. This mission end-state is a high-level goal – i.e., a Desire using Bratman’s terminology.

Bratman’s Beliefs represent what we know about the situation and accordingly are aspects of Observing and Orienting. Bratman’s Intentions are plans and are components of Deciding and Acting.

Bratman’s philosophical theory of reasoning has had a significant impact in the AI community and has led to development of a broad field of study referred to as BDI software architectures.

Decorative image of a person thinking.

  • Beliefs: characterize what an agent (i.e., a human or computer system) considers to be true about the world

  • Desires: represent an agent’s goals or preferred end states and are a necessary element in planning and deciding. We need to know what we want to accomplish in order to develop a plan

  • Intentions: are expressed in the plans developed for achieving desires


Like Michael Bratman, John McCarthy taught at Stanford University. Not only did McCarthy invent the term “artificial intelligence”, he also created the LISP programming language, created one of the earliest chess playing programs, and influenced the development of artificial intelligence in innumerable ways. In his essay “What is Artificial Intelligence?” McCarthy provides the following definition of AI and intelligence:

“It [artificial intelligence] is the science and engineering of making intelligent machines, especially intelligent computer programs…Intelligence is the computational part of the ability to achieve goals in the world. Varying kinds and degrees of intelligence occur in people, many animals, and some machines.”

John Boyd’s OODA loop provides an approach for adapting continuously to continuously changing situations. Michael Bratman’s BDI model describes rational action in terms of beliefs, desires, and intentions. Taken together, Boyd and Bratman offer a practical model for making Intelligent Decisions. John McCarthy identifies the kinds of problems that must be addressed in order to build Thinking Software.

Artificial Intelligence graphic

Artificial Intelligence graphic

McCarthy divides the discipline of Artificial Intelligence into 12 branches, as shown in Table 1. Veloxiti’s knowledge-based systems cannot be mapped directly to any single branch of McCarthy’s AI tree. However, we can certainly describe Veloxiti’s approach in terms of knowledge Representation technique, its Inferencing methods, and its Planning methodology.

Table 1: McCarthy’s Branches of AI

AI Branch

Quick Definition

Logical AI

Uses mathematically-based logical languages to reason about the world


AI programs often examine large numbers of possible solutions, and search techniques enable the discovery of good solutions

Pattern Recognition

AI techniques which involve comparing currently available data of some sort with patterns.   Pattern recognition is commonly used not only in vision systems but in many other applications such as problem diagnosis systems.


Techniques for representing information about the world in software


Techniques for inferring new information from existing facts.  Rule-based systems use inference as do many other techniques.

Common sense reasoning

Techniques for capturing and reasoning about the common sense world

Learning from experience

Techniques for learning about the world based on observation


Techniques for determining how to reach a goal.


The study of the kinds of knowledge required for solving problems in the world.


The study of things that exist in the world and their relationships.


Rules of thumb for making decisions

Genetic programming

Techniques for problem solving that involve selecting the best of many – possibly millions -- of generated candidate solutions.