The current investigation has been carried out on a unique approach i.e. supervised machine learning algorithm. The proposed scaled conjugate gradient (SCG) training method is used for trained the flight simulated data and to evaluate all the stability and control derivatives parameter estimation has been used for longitudinal short period mode. For the parameters like angle of attack, Pitch rate, elevator deflection as neural network inputs, lift and moment coefficients as output respectively considered in investigation. The aim is to train the delta method using scaled conjugate gradient to learn and predict aerodynamic derivative coefficients. This approach is applied on simulated flight data with various neural network model (SCG) thoroughly evaluated and validate with true flight data. It has been observed that the mean square error (MSE) value is very minimal for scaled conjugate gradient (SCG) as compared to Levenberg-Marquardt Optimization (LMO) method. This method’s accuracy is benchmarked against standard backpropagation algorithms to analyze the problem and predict aerodynamic parameters using neural network-based methods. In this study, all the models are compared based on their training performance and standard deviation of their estimates and actual values. The suggested SCG model will reduce the effort cost as well as experimental time of researchers to find accurate parameter estimated derivatives.

Although numerous models of aerial‐robots/drones and planet landers have been demonstrated to fly autonomously in natural and manmade environments for far‐reaching scientific and technological change to help humankind, no drone flowed yet with a push‐pull dual‐head electromagnetic propulsion (EMP) system ensuring unrestricted flight endurance. The precept of EMP is eminent as it hastens an object using a flowing electrical current, either to charge a field or opposing a magnetic field for locomotive use. Here we present a proof of concept to demonstrate the capability of a newly invented variable speed propulsion and energy conversion system for uninterruptedly generating the reciprocating motion of a magnetic piston by a solar battery‐driven polarity changer timing circuit together with a laser‐based timing circuit for redundancy. The reciprocating crusade of the magnetic piston could be successfully converted into rotary motion to generate continuous power for various multidisciplinary applications on Earth and the other planets effectively in lieu of long‐lasting rechargeable batteries. The proposed concept can also be used for devising various systems and subsystems from underwater to outer space by using light weight supermalloy, which ensures high magnetic fluxes with relatively low current. Lucrative applications include nanomaterial‐enabled locomotive applications. Aerial robots regulated with the EMP system could be used for fostering pharmaceutical florae in the international space station for drug design to negate the recently discovered phenomenon of Sanal flow choking in cardiovascular system at gravity and/or microgravity conditions. This concept is originated from mind and aimed for lucrative product developments of yocto to yotta scale energy systems and beyond by discovering new magnetic materials through the fourth industrial revolution. Herein, a proof of the concept of an unrestricted solar‐battery driven push‐pull electromagnetic propulsion and energy conversion device by use of light weight high power magnetic materials is presented.

Double Delta method along with Bayesian Regularization is proposed for the estimation of aircraft’s parameters. This method draws its inspiration from Taylor series expansion and existing Delta method for extracting aircraft’s parameters. Two models namely Feedforward Neural Network-Levenberg Marquardt and Feedforward Neural Network-Bayesian Regularization are separately trained by presenting flight data of aircraft’s longitudinal motion and control variables at input nodes and corresponding coefficients at output node of the models. For parameter estimation, the separate input files of longitudinal variables by adding small perturbation are prepared and given to trained neural models to predict the aerodynamic coefficients. Thereafter, the Double Delta method uses perturbed input files and corresponding predicted outputs for extracting the parameters. For the validation, the parameters are also extracted by implementing Delta method for both models and compared with the Double Delta method results. Superior performance in terms of mappings, MSE, and standard deviation is observed for the case of Double Delta method- Feedforward Neural Network-Bayesian regularization model compared to Delta Method- Feedforward Neural Network-Levenberg Marquardt, Delta Method-Feedforward Neural Network- Bayesian Regularization and Double Delta method- Feedforward Neural Network- Levenberg Marquardt models.