Research On Object Motion Based On Spatial Coding And Gradient Manifold

This thesis is based on the self-evolving intelligent system project of unmanned aerial vehi-cle(UAV)basketball competition.The main difficulties of the project are: fast and accurate po-sitioning of large indoor scenes,prediction of moving objects' trajectory,and hover positioning of aircraft attitude changes.This thesis has conducted in-depth research on the above-mentioned problems,and achieved the following results.(1)Fast and accurate positioning of large indoor scenes.This thesis designs and builds a UAV positioning system based on 3D motion capture.Considering the problem of construc-tion cost,this thesis proposes an indoor rapid positioning technology based on spatial coding.This technology uses the permutation and combination of different light colors in the encoding position as the characteristic information of the characteristic light source,and uses the spatial encoding and decoding technology to quickly and accurately realize indoor positioning through the look-up table method.This thesis details the realization process of this method and gives the calculation method of related parameters.The feasibility of the technology is proved through simulation experiments and physical simulation experiments.In addition,the genetic algorithm is used to optimize the positioning technology,which takes into account the positioning time and accuracy when there are many combinations of characteristic light sources.(2)The trajectory prediction of moving objects.Based on the stability of the motion trajectory,this thesis classifies the moving objects according to the Lyapunov index.Aiming at the simple mechanical motion in the convergence system,such as the rigid body motion of basketball and drone involved in the project,the curve equation is approximated by Newton's law of motion and physical meaning modeling.For chaotic systems,such as the complex gas flow field generated by the rotation of UAV blades,simulation experiments show that although the trajectory is unstable,due to the determination of the attractor,the manifold trajectory can be predicted by the Lyapunov exponent to achieve short-term prediction.In addition,this thesis has obtained the conclusion that motion is the space-time deformation of matter driven by the energy field gradient.(3)Research on hovering positioning problem of aircraft attitude change.This thesis deeply researches the problem that the aircraft cannot hover and locates when the aircraft has an attitude angle,and proposes a method to separate the propeller attitude control from the aircraft fuselage attitude control,and realizes the separate control of the actual aircraft attitude through experiments.At the same time,in order to solve the problem that the racquet is located above the propeller,the change in the trajectory of the basketball movement may cause the failure of receiving the ball,and the effective receiving area of the racquet below the propeller is reduced.This thesis explores and simulates the stability and controllability of the rotating airflow of the aircraft.It proves that the airflow field is controllable and reproducible,and demonstrates that it is feasible to use an ion generator to provide power for a rotorless aircraft.In short,starting from the project,this thesis provides a new method for realizing rapid positioning of large indoor scenes and accurately predicting the trajectory of objects.It ex-plores the movement realization process in depth,and provides a new perspective for research in related fields.