Planning Methods for Aerial Exploration and Ground Target Tracking

This thesis considers unmanned airborne surveillance systems equipped with electroopticalvision sensors. The aim is to increase the level of autonomy and improve thesystem performance by the use of planning methods for aerial exploration and targettracking. The general problem is very complex due to the “curse-of-dimensionality” andsuboptimal approaches are necessary in order to handle advanced surveillance missions.A general planning framework is proposed and the planner contains a high-level schedulerand a number of planning modes. Each mode consists of planning modules thatsolve smaller sub-tasks and in this thesis a number of these modules are developed. Inparticular, two major approaches are treated; information based planning, and Bayesiantarget search. In addition, the on-road target tracking problem is treated in detail and analgorithm based on the Particle filter is presented. In information based planning, different information measures are used to solve theoptimal trajectory planning problem for bearings-only estimation. Thus, the problem ishow to maneuver an unmanned aerial vehicle (UAV) to achieve the best possible estimateof a target location while observing it with a vision sensor. Approaches based on the Informationfilter and the differential entropy are proposed. The Information filter approachis also used to develop an exploration framework where the UAV flight trajectory and thesensor pointing direction are considered concurrently. In Bayesian target search, the aim is to find a target as quickly as possible givensome prior knowledge of where it might be. Methods based on both gradient search andcombinatorial optimization routines are proposed for the search problem where a UAV isequipped with a controllable vision sensor with limited field-of-view