Abstract
In the surveillance of interested regions by unmanned aerial vehicle (UAV), system performance relies greatly on the motion control strategy of the UAV and the operation characteristics of the onboard sensors. This paper investigates the 2D path planning problem for the lightweight UAV synthetic aperture radar (SAR) system in an environment of multiple regions of interest (ROIs), the sizes of which are comparable to the radar swath width. Taking into account the special requirements of the SAR system on the motion of the platform, we model path planning for UAV SAR as a constrained multiobjective optimization problem (MOP). Based on the fact that the UAV route can be designed in the map image, an image-based path planner is proposed in this paper. First, the neighboring ROIs are merged by the morphological operation. Then, the parts of routes for data collection of the ROIs can be located according to the geometric features of the ROIs and the observation geometry of UAV SAR. Lastly, the route segments for ROIs surveillance are connected by a path planning algorithm named the sampling-based sparse A* search (SSAS) algorithm. Simulation experiments in real scenarios demonstrate that the proposed sensor-oriented path planner can improve the reconnaissance performance of lightweight UAV SAR greatly compared with the conventional zigzag path planner.
Highlights
Day and night airborne observation is a mission of great significance in both military and civil applications
There are no widely accepted benchmark problems in the field of unmanned aerial vehicle (UAV) path planning, so the simulated environment is generally used in the existing literature [39,40,41,42,43,44]
Considering the regions of interest (ROIs) can be located with the aid of the classification technique, we utilized a classification map as the input of the path planner in the simulation experiments
Summary
Day and night airborne observation is a mission of great significance in both military and civil applications. The motivating problem of this paper is to monitor multiple dispersedly distributed regions of interest (ROIs) using a single lightweight UAV synthetic aperture radar (SAR). The areas illuminated by the radar beam are mainly determined by the motion strategy of the radar platform This is because the track of radar footprint is highly coupled with the route of the airplane, especially for the lightweight UAV SAR system without beam steering capability. With a variety of factors to be accounted for, the sensor-oriented path planning for UAV SAR under study is a constrained multiobjective optimization problem (MOP) in essence [21]. The path planning problem can be decoupled into two subproblems, i.e., locating the SAR-oriented path segments and designing the UAV-oriented sub-routes.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
