Prediction of the parachute deploy for landing at the desired point
暂无分享,去创建一个
In this paper, we introduce how to predict the parachute deploy for landing at the desired point. The UAV-parachute system is required 9-DOF dynamic modeling, so we build up the equations of motion for this system. And then the input and the output data sets are trained to compose the neural network. The input data sets are the flight conditions such as the deploy position, UAV's velocity, and wind velocity and the output data sets are the landing points such as the cross range and the down range position that simulated by the 9-DOF dynamic modeling. Using the training input and output data sets we can build up the nonlinear function approximator for the neural network. So we can predict the deploy timing and conditions such as the deploy position, UAV's velocity for landing at the desired point.
[1] Gabriele Kotsis,et al. Using formal methods to verify safe deep stall landing of a MAV , 2011, 2011 IEEE/AIAA 30th Digital Avionics Systems Conference.
[2] Martin T. Hagan,et al. Neural network design , 1995 .
[3] Paul Zarchan,et al. Tactical and strategic missile guidance , 1990 .
[4] Om Prakash,et al. Modeling and Simulation of 9-DOF Parafoil-Payload System Flight Dynamics , 2006 .
[5] Youdan Kim,et al. Design of generalized conceptual guidance law using aim angle , 2004 .