Quadrotor intercept trajectory planning and simulation

Abstract

Quadrotor drones pose a safety hazard when operated in or near controlled airspace. A hazardous quadrotor could be intercepted and removed by another quadrotor. In this thesis, we seek to determine if optimal control methods outperform missile control methods when applied to a quadrotor drone performing an intercept with a moving target. This is achieved by simulating the intercept of a target with a quadrotor and comparing the performance of several on-line trajectory planners. Two missile control-based trajectory planners, pursuit guidance and proportional navigation, are compared against an optimal control trajectory planner. The time and energy used by a simulated quadrotor to intercept a target are the performance measures used for comparison. The trajectory planners use a three-degree of freedom model, and the simulated quadrotor uses a six-degree of freedom model. Each trajectory planner is compared in a crossing, head-on, and tail-chase geometry. All of the on-line results are compared to an off-line optimal solution. The results show that the off-line optimal control method performs better than the on-line trajectory planners, regardless of intercept geometry type. The proportional navigation planner has the best performance of the on-line trajectory planners.