Algorithms Storage System

This paper focuses on the development and application of a particular subsystem that complements the architecture designed for planning, supervising, navigating and controlling autonomous flying vehicles. The algorithm storage system (ASS) was design to facilitate switching efficiently between different ways of performing a task. It was developed for a 10m long blimp that had 4 operational stages with a different vehicle model for each one; hence the need of an architectural subsystem that could manage the different maneuvers that the blimp had to carry out. The system can be easily transferred to any automated vehicle and is an important component in the development of high level artificial intelligence. The paper discusses possible applications such as reinforcement learning, machine learning, multiagent learning and experience gathering.