A DIFFERENTIAL GAME APPROACH TO ENERGY-EFFICIENT UAV PATH PLANNING IN CIVILIAN SENSOR ENVIRONMENTS

Abstract

This paper develops a trajectory optimization framework for unmanned aerial ve-hicles (UAVs) in sensor-rich or restricted environments. The objective is to reach designatedtargets while minimizing control effort and limiting exposure to ground-based sensors. TheUAV motion is modeled in continuous time using azimuth and elevation angles as controlinputs, and the problem is posed as a nonzero-sum differential game. A direct colloca-tion scheme with polynomial approximations and Gaussian quadrature is employed to solvethe resulting nonlinear program efficiently. Numerical simulations confirm that the methodgenerates smooth, dynamically feasible, and energy-efficient trajectories. The framework issuitable for real-time applications such as environmental monitoring, infrastructure inspec-tion, and search-and-rescue missions, and provides a foundation for future integration withadaptive learning and probabilistic sensor models.