Artificial Intelligence for Robotics – Reading Response

In the 1970s, the U.S. Army proposed the “OODA” decision-making model: the Observation, Judgment, Decision, and Action loop. With breakthroughs in artificial intelligence in the world and the accelerated transfer to the military, the “+intelligence” model will enable the “OODA” to unleash impressive capabilities on the battlefield by embedding intelligent technologies into the “OODA. “Govers explains how the OODA cycle can be used to connect artificial intelligence to robotics. He claims that robots have two “OODA” loops: “an inner loop or introspective loop, and an outward-looking environment sensor loop”. Govers explains the relationship between these two loops metaphorically: “The lower, inner loop takes priority over the slower, outer loop, just as the autonomic parts of your body (heartbeat, breathing, eating) take precedence over your task functions (going to work, paying bills, mowing the lawn).”

Reactive robotic architecture known as subsumption is closely related to behavior-based robotics. Subsumption architecture relates sensory data to action choices in an intimate and bottom-up manner as opposed to using symbolic mental representations of the environment to direct behavior. It accomplishes this by breaking down the main activity into its component behaviors. A hierarchy of layers has been created to arrange these sub-behaviors. Each layer implements a certain degree of behavioral competency, and higher levels have the capacity to subsume lower levels (= integrate/combine lower levels into a more complete whole) in order to produce functional behavior. The layers all receive sensor data in concurrently, operate together, and provide results. These outputs may consist of instructions for actuators or signals that block or inhibit other layers. Govers utilizes his version of the subsumption architecture to enable robots to have several layers of goals at once. According to his description, “this type of approach has become a standard across unmanned vehicles and robotics.”