An Introduction to Hybrid Control Architecture based on Multi-agent System for a Hexacopter Autonomy Optimization

摘要

An unmanned aerial vehicle UAV (also known as a drone) refers to a pilotless aircraft. Next generation of this kind of flying robots will be designed to fly autonomously at long range and high endurance for multi-complex purposes as surveillance, reconnaissance, electronic warfare and target acquisition. For operating in uncontrolled and dynamic environments, the robot must constantly reconfigure itself to adapt to the external conditions and its own goals. The need of control architecture to manage these reconfigurations becomes the first order of business to improve the autonomy of such robots. Such control is achieved through the use of computer systems. But "truly" autonomous UAV is done by placing a computer onboard the UAV with efficient control architecture, allowing it to process all commands itself. The most popular system meeting these criteria is the Raspberry Pi and Arduino families of microcontrollers. In the other hand an array of sensors including a GPS, an IMU and a camera are used. Through using of software state of the art and modern telecommunication networks, this aerial robots will be capable of intelligent navigation (multi-agent control architecture) combining GPS (Global Positioning System) in outdoor environment through the use of a machine interface employing mapping software maintaining an internet connection through 4G LTE and image processing in indoor space utilizing various image processing algorithms. An illustration of this study will be given in an application of control of an autonomous hexacopter developed by the team architecture of systems, in the national engineering school of electricity and mechanic in Casablanca in Morocco.