A Contribution to the Development of Sense and Avoid Systems for Autonomus Flight Missions
Buch in deiner Nähe kaufen
Beschreibung
This work addresses the development of concepts for the integration of Sense and Avoid systems into Unmanned Aircraft Systems (UAS). The necessity of such systems is given due to the fact that a large number of unmanned aircraft requires the presence of a pilot which can intervene immediately in hazardous situations. The capability of deciding independently in such situations and avoiding possible collisions increases the level of autonomy and is an essential prerequisite for these aircraft to be admitted to the civil airspace.
In this context, distinct approaches are taken into account, which can contribute to the implementation of a Sense and Avoid system for unmanned aircraft. The first method illustrates a flight path planning which is based on the generation of B-Spline curves. The B-Splines are applied not only to the planning of a flight path at the beginning of a mission but also to the generation of diverse avoidance maneuvers. The selection of appropriate maneuvers follows the conventional Traffic Alert and Collision Avoidance Systems (TCAS). Depending on the altitude, speed and the distances to the intruder aircraft, this system issues different recommendations as Resolution Advisory (RA). Similarly, the B-Spline flight path planning issues avoidance maneuvers which are dependent on distances, curve parameters as well as the constellation of the waypoints. In addition to this, the consideration of flight mechanics is important. This investigation is essential so that only maneuvers which are manageable in flight mechanical context are generated and taken into account.
An alternative approach compared to the B-Spline based flight path planning is the combination of the Distance Transformation with the so-called A* search algorithm. This distance transformation is widely used in digital image processing. The search algorithm is a common method in robotics in order to determine an optimum path to the goal.
By means of the transformation, each node in a map is allocated with the distance to the nearest obstacle and a discrete grading of the contrast in the map occurs where the regions around the obstacles are represented by dark and the obstacle-free regions by bright areas, respectively. This method provides the basis for the application of the A* search algorithm. By selecting a safety distance, it is possible that the algorithm can determine an optimum path by taking this threshold into account. For this purpose, a cost function is calculated for each step of the search. This function consists of the cost for the movement from the starting point to the actual position and the cost for the movement from the actual position to the goal. An efficient and collision-free path can be determined on the basis of the stepwise calculation of the cost function and the following comparison of the cost of the neighboring nodes.
Within the scope of this thesis, the B-Spline based flight path planning is integrated into an existing MATLAB/SIMULINK model of the unmanned aircraft type T200 of the Institut für Luft- und Raumfahrtsysteme (ILR). The detection of obstacles by means of a laser range finder and the algorithms for the path planning are programmed as level- 2 m-file s-function and are embedded into the simulation model. The conceived path planning method is investigated by means of distinct missions. These scenarios comprise level flight without and with wind influences as well as a combined climb and descent flight. Thereby, the main focus is on the lateral avoidance. The results show that the introduced method is capable of generating adequate avoidance maneuvers complying with flight mechanical aspects.
Autor*in
Jang-Bum Park
Themen in »A Contribution to the Development of Sense and Avoid Systems for Autonomus Flight Missions«
Autonomus Flight Missions Avoidance maneuser Flight path planning Sense and Avoid