Abstract
Coherent phantom track generation through controlling a group of electronic combat air vehicles is currently an area of great interest to the defense agency for the purpose of deceiving a radar network. However, generating an optimal or even feasible coherent phantom trajectory in real-time is challenging due to the high dimensionality of the problem and severe geometric constraints. This problem becomes even more difficult to solve when realistic kinematic and 6DOF dynamic constraints are considered. In this paper, the bio-inspired virtual motion camouflage based trajectory planning methodology, augmented with the derived early termination condition, is investigated to solve this constrained collaborative trajectory planning problem in two approaches: centralized and decentralized. In the centralized approach, one optimization loop is used to solve for the coherent trajectories for both phantom and electronic combat air vehicles. The virtual motion camouflage based formulation can help to dramatically reduce the problem dimension. In the decentralized approach, two optimization loops are designed. The first loop finds feasible phantom tracks based on the early termination condition and the equality and inequality constraints of the phantom track. Then the second loop uses the virtual motion camouflage based method to solve for the optimal electronic combat air vehicle trajectories based on the feasible phantom tracks obtained in the first loop. For both approaches, necessary conditions have been applied so that the initial and final velocities of the phantom and electronic combat air vehicles are guaranteed to be coherent. Through the proposed bio-inspired method, the dynamics models of the coherent pair of the phantom and actual vehicles can be represented by a single-degree-of-freedom vector, called the path control parameter vector. Thus a fast optimal phantom track design can be achieved. The fact that fewer equality constraints are involved in solving the formulated nonlinear programming problem will further make the convergence easier. Optimal solutions have been found in both centralized and decentralized simulations for several cases with different numbers of electronic combat air vehicles and varying numbers of discretization nodes. It is demonstrated that the decentralized approach can solve the problem much faster than the centralized one. Furthermore, the computational cost in the decentralized approach remains roughly the same for the cases when different numbers of nodes and different numbers of electronic combat air vehicles are involved. In comparison, the computational cost in the centralized approach increases dramatically when the number of nodes and/or the number of electronic combat air vehicles increases. It is concluded that the virtual motion camouflage based decentralized approach is appropriate for real-time implementation.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have