1 a Daptive Fuzzy Control of C O - Operating a Utonomous Robots

Balancing the conflicting demands imposed by a dynamic world on an autonomous robot requires a significant degree of adaptability. This paper describes a multi-layer control system for two co-operating mobile robots, which uses Fuzzy logic to adapt the relative importance of a set of reactive behaviours. The Fuzzy system therefore acts as an arbiter which smoothly interpolates control between conflicting behaviours. This allows the robots to successfully navigate out of potential minima, which are often associated with behaviour based control [12]. The adaptive mechanism itself is also modified by an array of vectors generated from an on-line analysis of the activity of each fuzzy rule. This considers the dynamic flow of the system through its matrix of fuzzy rules, and triggers a suppression of specific rules if the system is tending towards a stable state or point attractor. We propose that for a goal seeking agent in an unstructured environment maintaining a state of dynamic instability within the control system increases the probability of the agent reaching its goal. This work presents a new method for adapting a fuzzy logic controller within an autonomous agent. The total result is a significant increase in the degree of adaptation the robots can achieve based on their finite set of primary behaviours.