|M.Sc Student||Tsukerman Alexander|
|Subject||Trajectory Planning and Guidance for Civil Autonomous Aerial|
|Department||Department of Autonomous Systems and Robotics||Supervisor||PROF. Tal Shima|
|Full Thesis text|
This work addresses the concept of civil aerial refueling which was suggested in the framework of a European project to reduce environmental impact of air transportation. According to previous studies, this futuristic concept, is supposed to reduce fuel consumption of passenger aircraft by utilizing aerial refueling operations. The basic idea was that splitting a flight mission into sections would allow a passenger aircraft (the Receiver) to take-off with much less fuel on board than required to fly the whole mission. Dedicated Tanker aircraft will intercept the passenger aircraft along its route and transfer the required amount of fuel to finish the mission. The aircraft approach to close proximity, prior to the deployment of the refueling system, has proven to be extremely difficult and demanding task for pilots. Therefore, to ensure smooth and safe approach maneuver, automation is required. The rendezvous condition, resulting from a successful approach maneuver, is defined as zero relative position, zero relative velocity, and zero relative acceleration at the end of the maneuver. In this work optimal guidance laws, that aim at enforcing the rendezvous condition, are developed
and applied in realistic simulation model of the civil aerial refueling scenario. Inspired by missile interception optimal guidance theory, two novel guidance laws are developed to impose the terminal rendezvous condition while minimizing the required control effort. The guidance law aggressiveness is determined by a specific selection of weight coefficients allowing a certain amount of flexibility for the designer. The first guidance law is designed to deliver acceleration commands as part of a dual loop architecture. The inner autopilot loop is designed to execute the acceleration commands of the outer guidance loop. The guidance law is developed based on simple kinematic model augmented with a model of the inner autopilot loop dynamic response to guidance acceleration commands. The obtained guidance law is written in closed-from and analytic solutions are given for the time varying guidance gains and compared to other well -known guidance laws. The second guidance law is developed as a single loop autopilot-guidance controller. This integrated guidance and control law is based on an open-loop aircraft dynamic model. This architecture issues combined guidance and autopilot commands using full-state feedback. The guidance performance is evaluated using a high-fidelity simulation model available from previous studies on aerial refueling. It consists of detailed nonlinear aircraft simulation models of four-engined civil transport aircraft, including engine and actuator dynamics and nonlinear aerodynamic models. The simulation
environment also includes a model for the wake interaction between the aircraft. Both guidance laws were examined under various atmospheric and flight conditions and demonstrated the ability to steer the Tanker towards rendezvous as required by the guidance objectives. Also, to compare the performance of both guidance laws a dedicated performance criterion is defined. Numerous simulations were carried out under various design parameters and flight conditions to compare the defined performance criteria. The study illustrated the benefits and drawbacks of specific selection design wake coefficients and the overall guidance performance. The benefit of applying both guidance laws in a combined manner is also discussed.