Contour Based Path Planning with B-Spline Trajectory Generation for Unmanned Aerial Vehicles (UAVs) over Hostile Terrain

Abstract

This research focuses on trajectory generation algorithms that take into account the stealthiness of autonomous UAVs; generating stealthy paths through a region laden with enemy radars. The algorithm is employed to estimate the risk cost of the navigational space and generate an optimized path based on the user-specified threshold altitude value. Thus the generated path is represented with a set of low-radar risk waypoints being the coordinates of its control points. The radar-aware path planner is then approximated using cubic B-splines by considering the least radar risk to the destination. Simulated results are presented, illustrating the potential benefits of such algorithms

A 3D Laboratory Test-Platform for Overhead Power Line Inspection

“…In contrast, a number of studies have directly applied curves to plan flight path. Among these studies, Dubins, Clothoid, PH and B-splined curves, respectively, have been used to fit to flight paths [15][16][17][18][19] . However, the path planned by Dubins curves is not feasible for flight, due to its discontinuous curvature.…”

Systematic vs. Non-Systematic Search for 3D Aircraft Conflict Resolution

“…These methods produce a resolution maneuver which minimizes a given cost, a function of deviation from the original trajectory, flight time, fuel consumption, or energy. Contributions belonging to this category include, but are not limited to [10][11][12][13][14][15]. However, these methods usually need to make one or more unrealistic assumptions when extrapolating the future positions of aircraft (for example, no uncertainty about the aircraft speed).…”

Stealth path planning for unmanned aerial vehicle operations in radar zones

“…Another technique used to effectively represent the alternate paths generated by a path planner is to use B-splines. B-splines allow a parametric curved path to be described by only a few control points rather than the entire curve segmented which could be as many as thousands of sections, depending on the length and curvature of the path [69]. Recent methods include modelling the trajectory using splines based on elastic band theory [70] [71] and interactive manipulation of control points for spline trajectory generation [72] [ 73].…”

“…A value closer to 0 puts emphasis on shortest paths, with little regard for the exposure to radars, while a value closer to 1 put emphasis on threat exposure at the expense of longer path length. For this case, the cost weighting parameter was chosen to give a reasonable trade-off between the path length and proximity of the path to threats [106]. The experimentally defined value for has to be in line with our algorithm objective which can be computed in linear time.…”

“…Subsequently, we employ our radar-aware path-planning method to find an optimal path on the radar region. The optimal path should carry minimal radar risks, and adhere to physical and motion constraints [69]. We construct approximation of the radar sites at distinct risk levels ≤ ≤ ℎ ℎ at ranges from the position of the UAV, ( 0 , 0 ) in the sense that the risk is measured for all waypoints in the region,…”