In recent times, Unmanned Aerial Systems (UAS) have been employed in an increasingly diverse range of applications. Numerous UAS market forecasts portray a burgeoning future, with many applications in both the military and civilian domains. Within the civilian realm, UAS are expected to be useful in performing a wide range of missions such as disaster monitoring (e.g. wildfires, earth-quakes, tsunamis and cyclones), search and support, and atmospheric observation. However, to realise these civilian applications, seamless operation of UAS within the National Air Space (NAS) will be required. Increasing the levels of onboard autonomy will help to address this requirement. Additionally, increased autonomy also reduces the impact of onboard failures, potentially lower operational costs, and decrease operator workload. Numerous intelligent control architectures do exist in the literature for mobile robots, space based robots and for UAS. These include: the WITAS project, Open Control Platform, Remote Agent and TRAC/ReACT. However, none of these are specifically targeted at providing the required support for a wide range of civilian UAS missions. Operation of UAS in the NAS for civil applications require robust methods for dealing with emergency scenarios such as performing forced landings and collision avoidance to preserve the safety of people and property. This paper presents a new multi layered intelligent control architecture. The highest layer provides deliberative reasoning and includes situational awareness and mission planning subsystems. The middle layers deals with navigational aspects (such as path planning and manoeuvre generation). Finally, there is a functional control layer which comprises sensor and actuator subsystems and provides reactive functionality to enable forced landings and collision avoidance. Collision avoidance and forced landing technologies are currently under development at the Australian Research Centre for Aerospace Automation (ARCAA).
展开▼
